Project Mask.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fca7aa89",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "import numpy as np\n",
    "import json\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "from mrcnn.config import Config\n",
    "import mrcnn.utils as utils\n",
    "import mrcnn.model as modellib\n",
    "import mrcnn.visualize as visualize\n",
    "\n",
    "import os\n",
    "import numpy as np\n",
    "import cv2\n",
    "import json\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "61d8f8cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: scikit-image in /home/michael/anaconda3/lib/python3.9/site-packages (0.19.2)\n",
      "Collecting scikit-image\n",
      "  Using cached scikit_image-0.19.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.0 MB)\n",
      "Requirement already satisfied: networkx>=2.2 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (2.8.4)\n",
      "Requirement already satisfied: pillow!=7.1.0,!=7.1.1,!=8.3.0,>=6.1.0 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (9.2.0)\n",
      "Requirement already satisfied: packaging>=20.0 in /home/michael/.local/lib/python3.9/site-packages (from scikit-image) (22.0)\n",
      "Requirement already satisfied: scipy>=1.4.1 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (1.9.1)\n",
      "Requirement already satisfied: PyWavelets>=1.1.1 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (1.3.0)\n",
      "Requirement already satisfied: tifffile>=2019.7.26 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (2021.7.2)\n",
      "Requirement already satisfied: numpy>=1.17.0 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (1.21.5)\n",
      "Requirement already satisfied: imageio>=2.4.1 in /home/michael/anaconda3/lib/python3.9/site-packages (from scikit-image) (2.19.3)\n",
      "Installing collected packages: scikit-image\n",
      "  Attempting uninstall: scikit-image\n",
      "    Found existing installation: scikit-image 0.19.2\n",
      "    Uninstalling scikit-image-0.19.2:\n",
      "      Successfully uninstalled scikit-image-0.19.2\n",
      "Successfully installed scikit-image-0.19.3\n"
     ]
    }
   ],
   "source": [
    "!pip install -U scikit-image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6ee93bdb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the Mask R-CNN model\n",
    "from mrcnn.model import MaskRCNN\n",
    "from mrcnn import utils\n",
    "import mrcnn.model as modellib\n",
    "from mrcnn import visualize\n",
    "from mrcnn.config import Config\n",
    "from mrcnn import model as modellib, utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "323d8101",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/host/Mask-RCNN-TF2\n"
     ]
    }
   ],
   "source": [
    "dir_path = os.getcwd()\n",
    "print(dir_path)\n",
    "MODEL_DIR = dir_path\n",
    "COCO_MODEL_PATH = dir_path + \"/mask_rcnn_coco.h5\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "52fdf4a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MolesConfig(Config):\n",
    "    ''' \n",
    "    MolesConfig:\n",
    "        Contain the configuration for the dataset + those in Config\n",
    "    '''\n",
    "    NAME = \"moles\"\n",
    "    GPU_COUNT = 1 # put 2 or more if you are 1 or more gpu\n",
    "    IMAGES_PER_GPU = 1 # if you are a gpu you are choose how many image to process per gpu\n",
    "    NUM_CLASSES = 1 + 2  # background + (malignant , benign)\n",
    "    IMAGE_MIN_DIM = 128\n",
    "    IMAGE_MAX_DIM = 128\n",
    "\n",
    "    # hyperparameter\n",
    "    LEARNING_RATE = 0.001\n",
    "    STEPS_PER_EPOCH = 75\n",
    "    VALIDATION_STEPS = 5\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f66d7140",
   "metadata": {},
   "outputs": [],
   "source": [
    "class LesionDataset(utils.Dataset):\n",
    "    def load_lesions(self, dataset_dir, subset):\n",
    "        \"\"\"Load a subset of the lesions dataset.\n",
    "\n",
    "        dataset_dir: Root directory of the dataset.\n",
    "        subset: Subset to load: train or val\n",
    "        \"\"\"\n",
    "        # Add classes\n",
    "        self.add_class(\"lesions\", 1, \"lesion\")\n",
    "\n",
    "        # Load annotations\n",
    "        annotations = pd.read_csv(os.path.join(dataset_dir, \"annotation.csv\"))\n",
    "        annotations = annotations[annotations['subset'] == subset]\n",
    "        annotations = annotations.reset_index(drop=True)\n",
    "\n",
    "        # Load image and masks\n",
    "        for i in range(annotations.shape[0]):\n",
    "            image_path = os.path.join(dataset_dir, annotations['ImageIndex'][i])\n",
    "            mask_path = os.path.join(dataset_dir, \"Segmentation\", annotations['ImageIndex'].apply(lambda x: x.split(\".\")[0] + \".png\"))\n",
    "\n",
    "            image = cv2.imread(image_path)\n",
    "            mask = cv2.imread(mask_path, cv2.IMREAD_GRAYSCALE)\n",
    "\n",
    "            self.add_image(\"lesions\", image_id=i, path=image_path,\n",
    "                            width=image.shape[1], height=image.shape[0],\n",
    "                            mask=mask)\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "da945e83",
   "metadata": {},
   "outputs": [],
   "source": [
    "class SkinLesionConfig(Config):\n",
    "    # Give the configuration a recognizable name\n",
    "    NAME = \"skin_lesion_cfg\"\n",
    "    \n",
    "    # Set batch size to 1 since you'll be running inference on one image at a time.\n",
    "    # Batch size = GPU_COUNT * IMAGES_PER_GPU\n",
    "    GPU_COUNT = 1\n",
    "    IMAGES_PER_GPU = 1\n",
    "\n",
    "    # Number of classes (including background)\n",
    "    NUM_CLASSES = 1+1  # Background + skin lesion\n",
    "\n",
    "    # Use small images for faster training. Set the limits of the small side\n",
    "    # the large side, and that determines the image shape.\n",
    "    IMAGE_MIN_DIM = 256\n",
    "    IMAGE_MAX_DIM = 256\n",
    "\n",
    "    # Use smaller anchors because our image and objects are small\n",
    "    RPN_ANCHOR_SCALES = (8, 16, 32, 64, 128)  # anchor side in pixels\n",
    "\n",
    "    # Reduce training ROIs per image because the images are small and have\n",
    "    # few objects. Aim to allow ROI sampling to pick 33% positive ROIs.\n",
    "    TRAIN_ROIS_PER_IMAGE = 32\n",
    "\n",
    "    # Use a small epoch since the data is simple\n",
    "    STEPS_PER_EPOCH = 100\n",
    "\n",
    "    # use small validation steps since the epoch is small\n",
    "    VALIDATION_STEPS = 5\n",
    "    \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3e35cce8",
   "metadata": {},
   "outputs": [],
   "source": [
    "lesion = {\n",
    "    'nv': 'malignant',\n",
    "    'mel': 'malignant',\n",
    "    'bkl': 'benign',\n",
    "    'bcc': 'malignant',\n",
    "    'akiec': 'malignant',\n",
    "    'vasc': 'malignant',\n",
    "    'df': 'malignant'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "575f9e7c",
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "module 'tensorflow' has no attribute 'math'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-11-3df8330249bb>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m model =MaskRCNN(mode=\"inference\", \n\u001b[1;32m      2\u001b[0m                              \u001b[0mconfig\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mSkinLesionConfig\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m                              model_dir='./')\n\u001b[0m",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, mode, config, model_dir)\u001b[0m\n\u001b[1;32m   1835\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_dir\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_dir\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1836\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_log_dir\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1837\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeras_model\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1838\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1839\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mbuild\u001b[0;34m(self, mode, config)\u001b[0m\n\u001b[1;32m   2036\u001b[0m                                      \u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mPOOL_SIZE\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mNUM_CLASSES\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2037\u001b[0m                                      \u001b[0mtrain_bn\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTRAIN_BN\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2038\u001b[0;31m                                      fc_layers_size=config.FPN_CLASSIF_FC_LAYERS_SIZE)\n\u001b[0m\u001b[1;32m   2039\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2040\u001b[0m             \u001b[0;31m# Detections\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mfpn_classifier_graph\u001b[0;34m(rois, feature_maps, image_meta, pool_size, num_classes, train_bn, fc_layers_size)\u001b[0m\n\u001b[1;32m    923\u001b[0m     \u001b[0;31m# Shape: [batch, num_rois, POOL_SIZE, POOL_SIZE, channels]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    924\u001b[0m     x = PyramidROIAlign([pool_size, pool_size],\n\u001b[0;32m--> 925\u001b[0;31m                         name=\"roi_align_classifier\")([rois, image_meta] + feature_maps)\n\u001b[0m\u001b[1;32m    926\u001b[0m     \u001b[0;31m# Two 1024 FC layers (implemented with Conv2D for consistency)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    927\u001b[0m     x = KL.TimeDistributed(KL.Conv2D(fc_layers_size, (pool_size, pool_size), padding=\"valid\"),\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/keras/engine/topology.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs, **kwargs)\u001b[0m\n\u001b[1;32m    617\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    618\u001b[0m             \u001b[0;31m# Actually call the layer, collecting output(s), mask(s), and shape(s).\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 619\u001b[0;31m             \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    620\u001b[0m             \u001b[0moutput_mask\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompute_mask\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mprevious_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    621\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m    388\u001b[0m         \u001b[0;31m# e.g. a 224x224 ROI (in pixels) maps to P4\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    389\u001b[0m         \u001b[0mimage_area\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcast\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimage_shape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mimage_shape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat32\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 390\u001b[0;31m         \u001b[0mroi_level\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlog2_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msqrt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mh\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mw\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;36m224.0\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msqrt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mimage_area\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    391\u001b[0m         roi_level = tf.minimum(5, tf.maximum(\n\u001b[1;32m    392\u001b[0m             2, 4 + tf.cast(tf.round(roi_level), tf.int32)))\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mlog2_graph\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m    339\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mlog2_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    340\u001b[0m     \u001b[0;34m\"\"\"Implementation of Log2. TF doesn't have a native implementation.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 341\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlog\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlog\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m2.0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    342\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    343\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'tensorflow' has no attribute 'math'"
     ]
    }
   ],
   "source": [
    "model =MaskRCNN(mode=\"inference\", \n",
    "                             config=SkinLesionConfig(),\n",
    "                             model_dir='./')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "55f86dc4",
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "module 'tensorflow' has no attribute 'random'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-9-b90ae3dcb3bd>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mconfig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mSkinLesionConfig\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mMaskRCNN\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"training\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'./'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m model.load_weights(\"mask_rcnn_coco.h5\", by_name=True, exclude=[\n\u001b[1;32m      4\u001b[0m     \u001b[0;34m\"mrcnn_class_logits\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"mrcnn_bbox_fc\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m     \"mrcnn_bbox\", \"mrcnn_mask\"])\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, mode, config, model_dir)\u001b[0m\n\u001b[1;32m   1835\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_dir\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_dir\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1836\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_log_dir\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1837\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeras_model\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1838\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1839\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mbuild\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mbuild\u001b[0;34m(self, mode, config)\u001b[0m\n\u001b[1;32m   1988\u001b[0m             \u001b[0mrois\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_class_ids\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_bbox\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_mask\u001b[0m \u001b[0;34m=\u001b[0m\u001b[0;31m\\\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1989\u001b[0m                 DetectionTargetLayer(config, name=\"proposal_targets\")([\n\u001b[0;32m-> 1990\u001b[0;31m                     target_rois, input_gt_class_ids, gt_boxes, input_gt_masks])\n\u001b[0m\u001b[1;32m   1991\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1992\u001b[0m             \u001b[0;31m# Network Heads\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.5/dist-packages/keras/engine/topology.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs, **kwargs)\u001b[0m\n\u001b[1;32m    617\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    618\u001b[0m             \u001b[0;31m# Actually call the layer, collecting output(s), mask(s), and shape(s).\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 619\u001b[0;31m             \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    620\u001b[0m             \u001b[0moutput_mask\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompute_mask\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mprevious_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    621\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m    662\u001b[0m             lambda w, x, y, z: detection_targets_graph(\n\u001b[1;32m    663\u001b[0m                 w, x, y, z, self.config),\n\u001b[0;32m--> 664\u001b[0;31m             self.config.IMAGES_PER_GPU, names=names)\n\u001b[0m\u001b[1;32m    665\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    666\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/utils.py\u001b[0m in \u001b[0;36mbatch_slice\u001b[0;34m(inputs, graph_fn, batch_size, names)\u001b[0m\n\u001b[1;32m    818\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    819\u001b[0m         \u001b[0minputs_slice\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 820\u001b[0;31m         \u001b[0moutput_slice\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgraph_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minputs_slice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    821\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput_slice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mtuple\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    822\u001b[0m             \u001b[0moutput_slice\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0moutput_slice\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36m<lambda>\u001b[0;34m(w, x, y, z)\u001b[0m\n\u001b[1;32m    661\u001b[0m             \u001b[0;34m[\u001b[0m\u001b[0mproposals\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgt_class_ids\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgt_boxes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgt_masks\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    662\u001b[0m             lambda w, x, y, z: detection_targets_graph(\n\u001b[0;32m--> 663\u001b[0;31m                 w, x, y, z, self.config),\n\u001b[0m\u001b[1;32m    664\u001b[0m             self.config.IMAGES_PER_GPU, names=names)\n\u001b[1;32m    665\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/host/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mdetection_targets_graph\u001b[0;34m(proposals, gt_class_ids, gt_boxes, gt_masks, config)\u001b[0m\n\u001b[1;32m    551\u001b[0m     positive_count = int(config.TRAIN_ROIS_PER_IMAGE *\n\u001b[1;32m    552\u001b[0m                          config.ROI_POSITIVE_RATIO)\n\u001b[0;32m--> 553\u001b[0;31m     \u001b[0mpositive_indices\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshuffle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpositive_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mpositive_count\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    554\u001b[0m     \u001b[0mpositive_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpositive_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    555\u001b[0m     \u001b[0;31m# Negative ROIs. Add enough to maintain positive:negative ratio.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAttributeError\u001b[0m: module 'tensorflow' has no attribute 'random'"
     ]
    }
   ],
   "source": [
    "config = SkinLesionConfig()\n",
    "model = MaskRCNN(mode=\"training\", config=config, model_dir='./')\n",
    "model.load_weights(\"mask_rcnn_coco.h5\", by_name=True, exclude=[\n",
    "    \"mrcnn_class_logits\", \"mrcnn_bbox_fc\",\n",
    "    \"mrcnn_bbox\", \"mrcnn_mask\"])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "39e9b07c",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Metadata:\n",
    "    ''' \n",
    "    Metadata:\n",
    "        Contain everything about an image\n",
    "        - Mask\n",
    "        - Image\n",
    "        - Description\n",
    "    '''\n",
    "    def __init__(self, lesion, img, mask):\n",
    "        #self.meta = meta\n",
    "        self.img = img\n",
    "        self.type = lesion\n",
    "        self.mask = mask\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9239c8dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MoleDataset(utils.Dataset):\n",
    "    ''' \n",
    "    MoleDataset:\n",
    "        Used to process the data\n",
    "    '''\n",
    "    def load_shapes(self, dataset, height, width):\n",
    "        ''' Add the 2 class of skin cancer and put the metadata inside the model'''\n",
    "        self.add_class(\"moles\", 1, \"malignant\")\n",
    "        self.add_class(\"moles\", 2, \"benign\")\n",
    "\n",
    "        for i, info in enumerate(dataset):\n",
    "            height, width, channels = info.img.shape\n",
    "            self.add_image(source=\"moles\", image_id=i, path=None,\n",
    "                           width=width, height=height,\n",
    "                           img=info.img, shape=(info.type, channels, (height, width)),\n",
    "                           mask=info.mask, extra=info)\n",
    "\n",
    "    def load_image(self, image_id):\n",
    "        return self.image_info[image_id][\"img\"]\n",
    "\n",
    "    def image_reference(self, image_id):\n",
    "        if self.image_info[image_id][\"source\"] == \"moles\":\n",
    "            return self.image_info[image_id][\"shape\"]\n",
    "        else:\n",
    "            super(self.__class__).image_reference(self, image_id)\n",
    "\n",
    "    def load_mask(self, image_id):\n",
    "        ''' load the mask and return mask and the class of the image '''\n",
    "        info = self.image_info[image_id]\n",
    "        shapes = info[\"shape\"]\n",
    "        mask = info[\"mask\"].astype(np.uint8)\n",
    "        class_ids=np.array([self.class_names.index(shapes[0])])\n",
    "        return mask, class_ids.astype(np.int32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2208b647",
   "metadata": {},
   "outputs": [],
   "source": [
    "metadata = pd.read_csv('HAM10000_metadata.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "44c42b9e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>lesion_id</th>\n",
       "      <th>image_id</th>\n",
       "      <th>dx</th>\n",
       "      <th>dx_type</th>\n",
       "      <th>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>localization</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HAM_0000118</td>\n",
       "      <td>ISIC_0027419</td>\n",
       "      <td>bkl</td>\n",
       "      <td>histo</td>\n",
       "      <td>80.0</td>\n",
       "      <td>male</td>\n",
       "      <td>scalp</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>HAM_0000118</td>\n",
       "      <td>ISIC_0025030</td>\n",
       "      <td>bkl</td>\n",
       "      <td>histo</td>\n",
       "      <td>80.0</td>\n",
       "      <td>male</td>\n",
       "      <td>scalp</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HAM_0002730</td>\n",
       "      <td>ISIC_0026769</td>\n",
       "      <td>bkl</td>\n",
       "      <td>histo</td>\n",
       "      <td>80.0</td>\n",
       "      <td>male</td>\n",
       "      <td>scalp</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>HAM_0002730</td>\n",
       "      <td>ISIC_0025661</td>\n",
       "      <td>bkl</td>\n",
       "      <td>histo</td>\n",
       "      <td>80.0</td>\n",
       "      <td>male</td>\n",
       "      <td>scalp</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>HAM_0001466</td>\n",
       "      <td>ISIC_0031633</td>\n",
       "      <td>bkl</td>\n",
       "      <td>histo</td>\n",
       "      <td>75.0</td>\n",
       "      <td>male</td>\n",
       "      <td>ear</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     lesion_id      image_id   dx dx_type   age   sex localization\n",
       "0  HAM_0000118  ISIC_0027419  bkl   histo  80.0  male        scalp\n",
       "1  HAM_0000118  ISIC_0025030  bkl   histo  80.0  male        scalp\n",
       "2  HAM_0002730  ISIC_0026769  bkl   histo  80.0  male        scalp\n",
       "3  HAM_0002730  ISIC_0025661  bkl   histo  80.0  male        scalp\n",
       "4  HAM_0001466  ISIC_0031633  bkl   histo  75.0  male          ear"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metadata.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "ddc4b4a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "metadata['dx']=metadata['dx'].astype('str')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d740a17f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def data(x):\n",
    "    return metadata[metadata['image_id']==x]\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e544292c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "a7bd6d4e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>lesion_id</th>\n",
       "      <th>image_id</th>\n",
       "      <th>dx</th>\n",
       "      <th>dx_type</th>\n",
       "      <th>age</th>\n",
       "      <th>sex</th>\n",
       "      <th>localization</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8336</th>\n",
       "      <td>HAM_0000025</td>\n",
       "      <td>ISIC_0024349</td>\n",
       "      <td>nv</td>\n",
       "      <td>histo</td>\n",
       "      <td>35.0</td>\n",
       "      <td>male</td>\n",
       "      <td>lower extremity</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        lesion_id      image_id  dx dx_type   age   sex     localization\n",
       "8336  HAM_0000025  ISIC_0024349  nv   histo  35.0  male  lower extremity"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "filename ='ISIC_0024349'\n",
    "data(filename)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7102fd8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b4c640e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8067bbd1",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f1eb21ee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ISIC_0024349.jpg\n",
      "ISIC_0024349\n",
      "[[[165 144 177]\n",
      "  [166 143 181]\n",
      "  [164 139 177]\n",
      "  ...\n",
      "  [167 127 185]\n",
      "  [167 126 184]\n",
      "  [165 125 183]]\n",
      "\n",
      " [[169 144 178]\n",
      "  [168 143 181]\n",
      "  [164 140 182]\n",
      "  ...\n",
      "  [164 126 184]\n",
      "  [166 126 184]\n",
      "  [166 125 186]]\n",
      "\n",
      " [[174 145 184]\n",
      "  [170 143 183]\n",
      "  [166 141 185]\n",
      "  ...\n",
      "  [166 128 186]\n",
      "  [164 126 186]\n",
      "  [164 124 182]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[184 150 197]\n",
      "  [184 150 197]\n",
      "  [186 152 199]\n",
      "  ...\n",
      "  [176 132 203]\n",
      "  [181 137 208]\n",
      "  [177 131 204]]\n",
      "\n",
      " [[184 151 196]\n",
      "  [185 152 197]\n",
      "  [186 152 199]\n",
      "  ...\n",
      "  [169 128 203]\n",
      "  [172 126 202]\n",
      "  [167 121 197]]\n",
      "\n",
      " [[185 150 200]\n",
      "  [186 151 201]\n",
      "  [187 152 202]\n",
      "  ...\n",
      "  [162 120 197]\n",
      "  [158 114 191]\n",
      "  [154 108 185]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024353.jpg\n",
      "ISIC_0024353\n",
      "[[[160 147 231]\n",
      "  [163 151 233]\n",
      "  [163 151 233]\n",
      "  ...\n",
      "  [158 154 236]\n",
      "  [156 154 236]\n",
      "  [154 155 235]]\n",
      "\n",
      " [[160 153 234]\n",
      "  [164 152 234]\n",
      "  [162 150 232]\n",
      "  ...\n",
      "  [153 153 237]\n",
      "  [155 156 237]\n",
      "  [155 156 237]]\n",
      "\n",
      " [[163 153 236]\n",
      "  [164 151 235]\n",
      "  [163 150 236]\n",
      "  ...\n",
      "  [153 153 237]\n",
      "  [157 152 237]\n",
      "  [159 152 237]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[142 151 224]\n",
      "  [143 150 223]\n",
      "  [145 153 223]\n",
      "  ...\n",
      "  [129 147 206]\n",
      "  [132 150 209]\n",
      "  [133 152 209]]\n",
      "\n",
      " [[145 152 225]\n",
      "  [144 152 222]\n",
      "  [146 153 226]\n",
      "  ...\n",
      "  [133 147 213]\n",
      "  [137 150 212]\n",
      "  [134 149 212]]\n",
      "\n",
      " [[147 157 221]\n",
      "  [143 155 219]\n",
      "  [143 156 224]\n",
      "  ...\n",
      "  [134 148 214]\n",
      "  [134 148 214]\n",
      "  [130 147 210]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024313.jpg\n",
      "ISIC_0024313\n",
      "[[[150 140 186]\n",
      "  [151 139 187]\n",
      "  [154 140 188]\n",
      "  ...\n",
      "  [154 138 185]\n",
      "  [153 137 184]\n",
      "  [152 136 183]]\n",
      "\n",
      " [[150 138 186]\n",
      "  [152 139 185]\n",
      "  [153 140 184]\n",
      "  ...\n",
      "  [153 137 184]\n",
      "  [152 136 183]\n",
      "  [150 134 181]]\n",
      "\n",
      " [[155 142 186]\n",
      "  [155 142 186]\n",
      "  [158 146 188]\n",
      "  ...\n",
      "  [153 137 184]\n",
      "  [150 134 181]\n",
      "  [149 133 180]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[165 145 187]\n",
      "  [166 145 190]\n",
      "  [168 148 191]\n",
      "  ...\n",
      "  [168 150 187]\n",
      "  [169 151 188]\n",
      "  [169 151 188]]\n",
      "\n",
      " [[164 144 187]\n",
      "  [167 147 190]\n",
      "  [167 149 190]\n",
      "  ...\n",
      "  [168 150 187]\n",
      "  [169 151 188]\n",
      "  [169 151 188]]\n",
      "\n",
      " [[164 144 187]\n",
      "  [166 146 189]\n",
      "  [167 149 190]\n",
      "  ...\n",
      "  [168 150 187]\n",
      "  [170 152 189]\n",
      "  [168 151 188]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024333.jpg\n",
      "ISIC_0024333\n",
      "[[[102 103 117]\n",
      "  [ 99 100 114]\n",
      "  [100 100 116]\n",
      "  ...\n",
      "  [103  99 111]\n",
      "  [105 100 115]\n",
      "  [107 102 117]]\n",
      "\n",
      " [[101 102 116]\n",
      "  [100 101 115]\n",
      "  [100 100 116]\n",
      "  ...\n",
      "  [106 102 114]\n",
      "  [106 101 116]\n",
      "  [108 103 118]]\n",
      "\n",
      " [[104 103 119]\n",
      "  [103 103 119]\n",
      "  [101 104 119]\n",
      "  ...\n",
      "  [109 105 116]\n",
      "  [108 104 116]\n",
      "  [108 104 116]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 97  97 111]\n",
      "  [ 98  98 112]\n",
      "  [ 98  98 112]\n",
      "  ...\n",
      "  [ 91  79 101]\n",
      "  [ 92  80 102]\n",
      "  [ 94  82 104]]\n",
      "\n",
      " [[ 97  96 112]\n",
      "  [ 97  96 112]\n",
      "  [ 98  98 112]\n",
      "  ...\n",
      "  [ 89  77  99]\n",
      "  [ 90  78 100]\n",
      "  [ 91  79 101]]\n",
      "\n",
      " [[ 95  94 110]\n",
      "  [ 97  96 112]\n",
      "  [ 97  97 111]\n",
      "  ...\n",
      "  [ 87  75  97]\n",
      "  [ 88  76  98]\n",
      "  [ 90  78 100]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024373.jpg\n",
      "ISIC_0024373\n",
      "[[[146 141 220]\n",
      "  [147 141 218]\n",
      "  [147 142 221]\n",
      "  ...\n",
      "  [149 145 220]\n",
      "  [147 146 220]\n",
      "  [147 143 218]]\n",
      "\n",
      " [[147 140 221]\n",
      "  [147 139 222]\n",
      "  [149 139 222]\n",
      "  ...\n",
      "  [145 144 218]\n",
      "  [148 145 217]\n",
      "  [144 146 217]]\n",
      "\n",
      " [[147 141 220]\n",
      "  [146 138 221]\n",
      "  [145 136 226]\n",
      "  ...\n",
      "  [144 143 217]\n",
      "  [142 144 215]\n",
      "  [144 144 214]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[121 108 188]\n",
      "  [117 103 185]\n",
      "  [115 102 186]\n",
      "  ...\n",
      "  [115 124 197]\n",
      "  [110 126 195]\n",
      "  [115 127 197]]\n",
      "\n",
      " [[117 108 189]\n",
      "  [116 106 189]\n",
      "  [117 107 190]\n",
      "  ...\n",
      "  [116 122 195]\n",
      "  [116 122 195]\n",
      "  [123 127 198]]\n",
      "\n",
      " [[118 109 190]\n",
      "  [119 109 192]\n",
      "  [118 109 190]\n",
      "  ...\n",
      "  [112 124 196]\n",
      "  [114 126 198]\n",
      "  [118 127 201]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024306.jpg\n",
      "ISIC_0024306\n",
      "[[[131 131 221]\n",
      "  [126 128 222]\n",
      "  [127 129 223]\n",
      "  ...\n",
      "  [138 134 229]\n",
      "  [138 136 226]\n",
      "  [139 135 230]]\n",
      "\n",
      " [[129 129 223]\n",
      "  [126 130 225]\n",
      "  [133 132 228]\n",
      "  ...\n",
      "  [137 135 225]\n",
      "  [136 136 226]\n",
      "  [135 134 226]]\n",
      "\n",
      " [[126 128 222]\n",
      "  [131 129 229]\n",
      "  [134 130 231]\n",
      "  ...\n",
      "  [132 136 224]\n",
      "  [133 136 227]\n",
      "  [134 137 228]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[125 118 209]\n",
      "  [126 119 210]\n",
      "  [126 121 212]\n",
      "  ...\n",
      "  [128 140 210]\n",
      "  [130 141 209]\n",
      "  [128 139 207]]\n",
      "\n",
      " [[118 116 205]\n",
      "  [121 119 208]\n",
      "  [124 120 209]\n",
      "  ...\n",
      "  [129 142 210]\n",
      "  [131 144 212]\n",
      "  [130 143 211]]\n",
      "\n",
      " [[115 116 207]\n",
      "  [116 120 209]\n",
      "  [118 120 208]\n",
      "  ...\n",
      "  [125 143 208]\n",
      "  [126 146 211]\n",
      "  [124 144 209]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024371.jpg\n",
      "ISIC_0024371\n",
      "[[[  5   6  10]\n",
      "  [  5   6  10]\n",
      "  [  5   6  10]\n",
      "  ...\n",
      "  [  2   2   2]\n",
      "  [  2   2   2]\n",
      "  [  2   2   2]]\n",
      "\n",
      " [[  4   6   7]\n",
      "  [  5   6  10]\n",
      "  [  3   6  10]\n",
      "  ...\n",
      "  [  2   2   2]\n",
      "  [  2   2   2]\n",
      "  [  2   2   2]]\n",
      "\n",
      " [[  5   6  10]\n",
      "  [  6   7  11]\n",
      "  [  7   8  12]\n",
      "  ...\n",
      "  [  2   2   2]\n",
      "  [  2   2   2]\n",
      "  [  2   2   2]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 83  87 115]\n",
      "  [ 88  92 120]\n",
      "  [ 92  96 125]\n",
      "  ...\n",
      "  [ 12  14  24]\n",
      "  [ 11  12  22]\n",
      "  [  9  11  22]]\n",
      "\n",
      " [[ 75  82 109]\n",
      "  [ 79  86 113]\n",
      "  [ 84  91 118]\n",
      "  ...\n",
      "  [ 12  13  23]\n",
      "  [ 12  13  23]\n",
      "  [  9  11  21]]\n",
      "\n",
      " [[ 73  80 107]\n",
      "  [ 78  85 112]\n",
      "  [ 80  89 116]\n",
      "  ...\n",
      "  [ 12  12  24]\n",
      "  [ 11  11  23]\n",
      "  [  9  11  22]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024392.jpg\n",
      "ISIC_0024392\n",
      "[[[177 165 237]\n",
      "  [178 166 238]\n",
      "  [179 167 239]\n",
      "  ...\n",
      "  [180 171 244]\n",
      "  [178 171 244]\n",
      "  [175 171 243]]\n",
      "\n",
      " [[178 164 236]\n",
      "  [180 166 238]\n",
      "  [179 165 237]\n",
      "  ...\n",
      "  [174 170 245]\n",
      "  [173 168 243]\n",
      "  [172 168 240]]\n",
      "\n",
      " [[174 161 237]\n",
      "  [181 166 240]\n",
      "  [181 165 242]\n",
      "  ...\n",
      "  [173 166 239]\n",
      "  [174 165 238]\n",
      "  [178 166 238]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[159 162 230]\n",
      "  [157 163 230]\n",
      "  [154 162 229]\n",
      "  ...\n",
      "  [152 154 212]\n",
      "  [155 156 214]\n",
      "  [153 154 212]]\n",
      "\n",
      " [[155 164 231]\n",
      "  [154 163 230]\n",
      "  [158 164 231]\n",
      "  ...\n",
      "  [155 156 212]\n",
      "  [159 159 213]\n",
      "  [158 158 212]]\n",
      "\n",
      " [[156 165 229]\n",
      "  [155 164 228]\n",
      "  [153 164 226]\n",
      "  ...\n",
      "  [156 158 216]\n",
      "  [157 158 216]\n",
      "  [157 158 216]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024350.jpg\n",
      "ISIC_0024350\n",
      "[[[167 147 236]\n",
      "  [167 149 240]\n",
      "  [165 148 235]\n",
      "  ...\n",
      "  [162 146 234]\n",
      "  [159 141 230]\n",
      "  [162 144 233]]\n",
      "\n",
      " [[165 147 240]\n",
      "  [167 149 240]\n",
      "  [164 147 238]\n",
      "  ...\n",
      "  [161 144 235]\n",
      "  [158 143 235]\n",
      "  [159 144 236]]\n",
      "\n",
      " [[164 148 235]\n",
      "  [166 150 238]\n",
      "  [164 148 235]\n",
      "  ...\n",
      "  [156 142 230]\n",
      "  [158 144 232]\n",
      "  [161 144 235]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[149 143 222]\n",
      "  [150 144 221]\n",
      "  [148 145 225]\n",
      "  ...\n",
      "  [153 145 215]\n",
      "  [148 143 212]\n",
      "  [147 143 215]]\n",
      "\n",
      " [[146 143 222]\n",
      "  [148 147 221]\n",
      "  [149 144 223]\n",
      "  ...\n",
      "  [154 144 214]\n",
      "  [151 144 211]\n",
      "  [151 147 213]]\n",
      "\n",
      " [[150 144 225]\n",
      "  [150 146 222]\n",
      "  [150 146 222]\n",
      "  ...\n",
      "  [153 144 217]\n",
      "  [152 143 216]\n",
      "  [150 143 218]]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024324.jpg\n",
      "ISIC_0024324\n",
      "[[[171 145 168]\n",
      "  [166 145 167]\n",
      "  [167 147 166]\n",
      "  ...\n",
      "  [170 148 166]\n",
      "  [173 149 167]\n",
      "  [172 150 169]]\n",
      "\n",
      " [[171 147 165]\n",
      "  [166 147 164]\n",
      "  [166 147 164]\n",
      "  ...\n",
      "  [171 149 167]\n",
      "  [174 150 168]\n",
      "  [173 148 168]]\n",
      "\n",
      " [[167 145 163]\n",
      "  [165 146 163]\n",
      "  [165 146 163]\n",
      "  ...\n",
      "  [172 150 168]\n",
      "  [172 148 166]\n",
      "  [174 149 169]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[163 141 169]\n",
      "  [162 140 168]\n",
      "  [165 140 166]\n",
      "  ...\n",
      "  [165 146 165]\n",
      "  [166 147 168]\n",
      "  [167 145 169]]\n",
      "\n",
      " [[162 140 164]\n",
      "  [165 144 166]\n",
      "  [163 141 160]\n",
      "  ...\n",
      "  [164 145 164]\n",
      "  [165 146 167]\n",
      "  [166 145 167]]\n",
      "\n",
      " [[166 140 163]\n",
      "  [165 142 164]\n",
      "  [161 140 162]\n",
      "  ...\n",
      "  [162 146 164]\n",
      "  [165 146 167]\n",
      "  [168 145 167]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024364.jpg\n",
      "ISIC_0024364\n",
      "[[[135 124 217]\n",
      "  [134 125 215]\n",
      "  [135 128 213]\n",
      "  ...\n",
      "  [128 126 202]\n",
      "  [132 127 206]\n",
      "  [130 128 210]]\n",
      "\n",
      " [[140 127 219]\n",
      "  [136 127 214]\n",
      "  [132 126 215]\n",
      "  ...\n",
      "  [133 127 204]\n",
      "  [129 128 202]\n",
      "  [126 123 202]]\n",
      "\n",
      " [[138 129 216]\n",
      "  [136 127 214]\n",
      "  [134 127 218]\n",
      "  ...\n",
      "  [130 130 208]\n",
      "  [123 124 204]\n",
      "  [119 116 195]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[122 122 198]\n",
      "  [109 115 198]\n",
      "  [112 106 195]\n",
      "  ...\n",
      "  [114 115 195]\n",
      "  [114 118 196]\n",
      "  [109 117 194]]\n",
      "\n",
      " [[110 122 200]\n",
      "  [115 119 197]\n",
      "  [109 111 195]\n",
      "  ...\n",
      "  [106 111 190]\n",
      "  [105 111 188]\n",
      "  [110 112 190]]\n",
      "\n",
      " [[122 121 201]\n",
      "  [124 118 200]\n",
      "  [115 112 197]\n",
      "  ...\n",
      "  [ 93 101 184]\n",
      "  [ 99 105 188]\n",
      "  [110 111 192]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024310.jpg\n",
      "ISIC_0024310\n",
      "[[[36 32 73]\n",
      "  [37 35 71]\n",
      "  [42 33 73]\n",
      "  ...\n",
      "  [32 29 78]\n",
      "  [32 29 78]\n",
      "  [30 29 73]]\n",
      "\n",
      " [[36 34 70]\n",
      "  [38 34 69]\n",
      "  [37 32 71]\n",
      "  ...\n",
      "  [33 30 80]\n",
      "  [33 29 81]\n",
      "  [32 30 76]]\n",
      "\n",
      " [[37 32 71]\n",
      "  [37 32 71]\n",
      "  [38 33 72]\n",
      "  ...\n",
      "  [32 30 76]\n",
      "  [30 28 74]\n",
      "  [33 32 74]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[37 37 73]\n",
      "  [39 33 74]\n",
      "  [39 35 77]\n",
      "  ...\n",
      "  [43 32 70]\n",
      "  [41 32 72]\n",
      "  [43 32 72]]\n",
      "\n",
      " [[39 35 77]\n",
      "  [38 33 78]\n",
      "  [39 34 79]\n",
      "  ...\n",
      "  [42 33 73]\n",
      "  [42 33 76]\n",
      "  [41 30 76]]\n",
      "\n",
      " [[40 34 75]\n",
      "  [38 34 76]\n",
      "  [39 35 77]\n",
      "  ...\n",
      "  [41 34 69]\n",
      "  [41 33 70]\n",
      "  [43 32 70]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024329.jpg\n",
      "ISIC_0024329\n",
      "[[[128 139 167]\n",
      "  [128 139 167]\n",
      "  [127 139 167]\n",
      "  ...\n",
      "  [149 159 193]\n",
      "  [150 159 196]\n",
      "  [153 163 197]]\n",
      "\n",
      " [[127 139 167]\n",
      "  [127 139 167]\n",
      "  [128 138 168]\n",
      "  ...\n",
      "  [147 157 191]\n",
      "  [146 158 194]\n",
      "  [152 162 196]]\n",
      "\n",
      " [[126 138 166]\n",
      "  [128 140 168]\n",
      "  [129 140 168]\n",
      "  ...\n",
      "  [146 155 189]\n",
      "  [146 156 190]\n",
      "  [149 159 193]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[102 120 151]\n",
      "  [100 120 151]\n",
      "  [101 121 152]\n",
      "  ...\n",
      "  [128 131 162]\n",
      "  [128 131 162]\n",
      "  [128 131 162]]\n",
      "\n",
      " [[100 120 151]\n",
      "  [100 120 151]\n",
      "  [100 120 151]\n",
      "  ...\n",
      "  [129 130 164]\n",
      "  [127 128 162]\n",
      "  [128 130 161]]\n",
      "\n",
      " [[101 121 152]\n",
      "  [ 99 119 150]\n",
      "  [ 99 119 150]\n",
      "  ...\n",
      "  [128 130 164]\n",
      "  [125 127 161]\n",
      "  [127 130 161]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024339.jpg\n",
      "ISIC_0024339\n",
      "[[[158 149 235]\n",
      "  [160 151 237]\n",
      "  [157 148 238]\n",
      "  ...\n",
      "  [182 169 243]\n",
      "  [183 168 242]\n",
      "  [189 169 244]]\n",
      "\n",
      " [[156 148 235]\n",
      "  [159 149 239]\n",
      "  [159 150 237]\n",
      "  ...\n",
      "  [184 171 247]\n",
      "  [181 167 239]\n",
      "  [180 168 240]]\n",
      "\n",
      " [[161 147 235]\n",
      "  [161 148 240]\n",
      "  [165 150 241]\n",
      "  ...\n",
      "  [183 172 244]\n",
      "  [181 168 244]\n",
      "  [181 170 243]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[142 153 221]\n",
      "  [145 153 223]\n",
      "  [145 153 223]\n",
      "  ...\n",
      "  [150 153 221]\n",
      "  [153 154 222]\n",
      "  [148 149 217]]\n",
      "\n",
      " [[146 154 221]\n",
      "  [143 150 219]\n",
      "  [144 153 221]\n",
      "  ...\n",
      "  [147 153 220]\n",
      "  [150 153 221]\n",
      "  [151 151 221]]\n",
      "\n",
      " [[145 155 225]\n",
      "  [145 152 225]\n",
      "  [147 155 225]\n",
      "  ...\n",
      "  [145 150 219]\n",
      "  [144 149 218]\n",
      "  [143 150 219]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024402.jpg\n",
      "ISIC_0024402\n",
      "[[[174 152 177]\n",
      "  [175 155 178]\n",
      "  [173 151 175]\n",
      "  ...\n",
      "  [152 130 148]\n",
      "  [143 121 139]\n",
      "  [137 115 133]]\n",
      "\n",
      " [[176 152 176]\n",
      "  [170 149 171]\n",
      "  [172 150 174]\n",
      "  ...\n",
      "  [153 131 149]\n",
      "  [149 127 145]\n",
      "  [142 120 138]]\n",
      "\n",
      " [[178 156 180]\n",
      "  [174 152 176]\n",
      "  [174 152 176]\n",
      "  ...\n",
      "  [158 132 155]\n",
      "  [154 132 151]\n",
      "  [150 128 146]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[160 136 160]\n",
      "  [161 140 162]\n",
      "  [162 142 165]\n",
      "  ...\n",
      "  [ 87  59  58]\n",
      "  [ 66  45  43]\n",
      "  [ 49  31  32]]\n",
      "\n",
      " [[155 136 157]\n",
      "  [157 136 158]\n",
      "  [161 140 162]\n",
      "  ...\n",
      "  [ 71  47  49]\n",
      "  [ 53  34  37]\n",
      "  [ 40  27  29]]\n",
      "\n",
      " [[150 132 149]\n",
      "  [154 133 155]\n",
      "  [158 139 160]\n",
      "  ...\n",
      "  [ 54  40  42]\n",
      "  [ 43  28  32]\n",
      "  [ 34  24  30]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024345.jpg\n",
      "ISIC_0024345\n",
      "[[[142 148 195]\n",
      "  [137 142 191]\n",
      "  [133 143 191]\n",
      "  ...\n",
      "  [128 132 185]\n",
      "  [131 133 187]\n",
      "  [131 133 187]]\n",
      "\n",
      " [[141 147 194]\n",
      "  [138 144 191]\n",
      "  [129 139 186]\n",
      "  ...\n",
      "  [130 134 187]\n",
      "  [130 134 187]\n",
      "  [130 134 187]]\n",
      "\n",
      " [[139 147 194]\n",
      "  [140 146 193]\n",
      "  [131 141 189]\n",
      "  ...\n",
      "  [130 134 187]\n",
      "  [129 135 188]\n",
      "  [127 133 186]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[145 151 192]\n",
      "  [145 151 192]\n",
      "  [144 150 193]\n",
      "  ...\n",
      "  [135 143 190]\n",
      "  [133 143 190]\n",
      "  [133 143 190]]\n",
      "\n",
      " [[140 146 189]\n",
      "  [141 147 188]\n",
      "  [140 147 190]\n",
      "  ...\n",
      "  [133 143 190]\n",
      "  [135 143 190]\n",
      "  [135 143 190]]\n",
      "\n",
      " [[136 145 189]\n",
      "  [136 145 188]\n",
      "  [136 144 191]\n",
      "  ...\n",
      "  [131 144 190]\n",
      "  [135 143 190]\n",
      "  [137 143 190]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024361.jpg\n",
      "ISIC_0024361\n",
      "[[[179 184 217]\n",
      "  [179 184 217]\n",
      "  [180 185 218]\n",
      "  ...\n",
      "  [100 117 136]\n",
      "  [110 127 146]\n",
      "  [125 140 166]]\n",
      "\n",
      " [[181 186 219]\n",
      "  [182 184 218]\n",
      "  [179 184 217]\n",
      "  ...\n",
      "  [174 185 215]\n",
      "  [170 181 211]\n",
      "  [165 176 208]]\n",
      "\n",
      " [[180 185 218]\n",
      "  [180 185 218]\n",
      "  [180 182 216]\n",
      "  ...\n",
      "  [164 173 210]\n",
      "  [160 168 205]\n",
      "  [154 163 201]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[175 174 216]\n",
      "  [174 172 214]\n",
      "  [178 177 219]\n",
      "  ...\n",
      "  [175 184 222]\n",
      "  [180 185 224]\n",
      "  [183 183 223]]\n",
      "\n",
      " [[179 178 220]\n",
      "  [177 176 218]\n",
      "  [175 174 216]\n",
      "  ...\n",
      "  [177 186 224]\n",
      "  [180 185 224]\n",
      "  [184 184 224]]\n",
      "\n",
      " [[173 175 216]\n",
      "  [176 178 219]\n",
      "  [175 177 218]\n",
      "  ...\n",
      "  [176 185 223]\n",
      "  [183 185 225]\n",
      "  [184 183 223]]]\n",
      "[[[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]]]\n",
      "ISIC_0024341.jpg\n",
      "ISIC_0024341\n",
      "[[[169 137 214]\n",
      "  [168 138 214]\n",
      "  [165 140 214]\n",
      "  ...\n",
      "  [170 151 220]\n",
      "  [170 156 220]\n",
      "  [164 153 216]]\n",
      "\n",
      " [[165 138 212]\n",
      "  [167 140 214]\n",
      "  [168 141 215]\n",
      "  ...\n",
      "  [168 153 221]\n",
      "  [171 157 223]\n",
      "  [164 152 218]]\n",
      "\n",
      " [[162 138 210]\n",
      "  [165 140 214]\n",
      "  [163 139 209]\n",
      "  ...\n",
      "  [172 153 226]\n",
      "  [173 153 226]\n",
      "  [169 149 222]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[130 119 191]\n",
      "  [130 119 191]\n",
      "  [130 119 192]\n",
      "  ...\n",
      "  [152 144 207]\n",
      "  [153 145 208]\n",
      "  [153 144 207]]\n",
      "\n",
      " [[133 120 194]\n",
      "  [128 117 190]\n",
      "  [129 118 191]\n",
      "  ...\n",
      "  [144 145 203]\n",
      "  [148 150 208]\n",
      "  [151 152 210]]\n",
      "\n",
      " [[130 117 195]\n",
      "  [128 116 192]\n",
      "  [133 117 194]\n",
      "  ...\n",
      "  [146 141 202]\n",
      "  [147 147 207]\n",
      "  [151 151 211]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024321.jpg\n",
      "ISIC_0024321\n",
      "[[[107 106 202]\n",
      "  [108 108 202]\n",
      "  [110 114 203]\n",
      "  ...\n",
      "  [128 120 213]\n",
      "  [131 120 217]\n",
      "  [128 121 218]]\n",
      "\n",
      " [[107 106 202]\n",
      "  [108 106 202]\n",
      "  [116 110 205]\n",
      "  ...\n",
      "  [124 121 213]\n",
      "  [128 122 217]\n",
      "  [125 121 216]]\n",
      "\n",
      " [[115 107 202]\n",
      "  [112 106 201]\n",
      "  [114 108 203]\n",
      "  ...\n",
      "  [120 122 210]\n",
      "  [116 123 210]\n",
      "  [122 124 212]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[122 128 201]\n",
      "  [118 127 201]\n",
      "  [114 126 198]\n",
      "  ...\n",
      "  [111 115 196]\n",
      "  [109 116 196]\n",
      "  [111 118 197]]\n",
      "\n",
      " [[120 129 203]\n",
      "  [119 128 202]\n",
      "  [119 127 204]\n",
      "  ...\n",
      "  [116 119 200]\n",
      "  [114 117 198]\n",
      "  [117 119 203]]\n",
      "\n",
      " [[121 128 201]\n",
      "  [121 127 202]\n",
      "  [119 125 200]\n",
      "  ...\n",
      "  [122 121 201]\n",
      "  [121 120 200]\n",
      "  [122 120 202]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024312.jpg\n",
      "ISIC_0024312\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[165 137 167]\n",
      "  [169 136 171]\n",
      "  [169 136 171]\n",
      "  ...\n",
      "  [176 143 178]\n",
      "  [178 145 180]\n",
      "  [180 144 180]]\n",
      "\n",
      " [[164 137 171]\n",
      "  [169 135 172]\n",
      "  [169 135 172]\n",
      "  ...\n",
      "  [176 144 175]\n",
      "  [176 145 176]\n",
      "  [179 147 178]]\n",
      "\n",
      " [[164 133 170]\n",
      "  [169 134 174]\n",
      "  [168 134 171]\n",
      "  ...\n",
      "  [178 147 180]\n",
      "  [177 144 178]\n",
      "  [178 145 179]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[144 116 162]\n",
      "  [143 116 160]\n",
      "  [145 118 162]\n",
      "  ...\n",
      "  [173 141 176]\n",
      "  [172 139 173]\n",
      "  [174 142 173]]\n",
      "\n",
      " [[142 115 159]\n",
      "  [143 116 160]\n",
      "  [142 115 159]\n",
      "  ...\n",
      "  [173 141 176]\n",
      "  [171 138 173]\n",
      "  [173 140 174]]\n",
      "\n",
      " [[142 117 155]\n",
      "  [142 117 155]\n",
      "  [141 116 154]\n",
      "  ...\n",
      "  [173 141 176]\n",
      "  [173 140 175]\n",
      "  [174 139 173]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024354.jpg\n",
      "ISIC_0024354\n",
      "[[[172 154 231]\n",
      "  [172 153 232]\n",
      "  [171 150 229]\n",
      "  ...\n",
      "  [184 167 241]\n",
      "  [182 168 240]\n",
      "  [182 170 242]]\n",
      "\n",
      " [[170 153 234]\n",
      "  [172 153 234]\n",
      "  [172 153 234]\n",
      "  ...\n",
      "  [184 170 242]\n",
      "  [181 167 239]\n",
      "  [178 166 238]]\n",
      "\n",
      " [[170 150 233]\n",
      "  [171 151 234]\n",
      "  [171 151 234]\n",
      "  ...\n",
      "  [183 169 241]\n",
      "  [184 170 241]\n",
      "  [182 170 242]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[162 152 222]\n",
      "  [165 155 225]\n",
      "  [166 156 222]\n",
      "  ...\n",
      "  [169 164 226]\n",
      "  [165 160 222]\n",
      "  [168 163 225]]\n",
      "\n",
      " [[160 153 220]\n",
      "  [159 155 221]\n",
      "  [162 155 222]\n",
      "  ...\n",
      "  [163 161 221]\n",
      "  [163 158 219]\n",
      "  [168 163 225]]\n",
      "\n",
      " [[164 154 224]\n",
      "  [163 155 226]\n",
      "  [161 153 224]\n",
      "  ...\n",
      "  [165 165 225]\n",
      "  [164 162 221]\n",
      "  [165 164 220]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024314.jpg\n",
      "ISIC_0024314\n",
      "[[[213 179 196]\n",
      "  [212 179 194]\n",
      "  [211 177 194]\n",
      "  ...\n",
      "  [201 173 192]\n",
      "  [202 172 191]\n",
      "  [200 172 191]]\n",
      "\n",
      " [[216 179 195]\n",
      "  [214 181 196]\n",
      "  [214 181 196]\n",
      "  ...\n",
      "  [203 172 193]\n",
      "  [203 173 192]\n",
      "  [200 172 191]]\n",
      "\n",
      " [[213 180 195]\n",
      "  [212 181 196]\n",
      "  [212 181 196]\n",
      "  ...\n",
      "  [204 172 191]\n",
      "  [202 172 191]\n",
      "  [203 172 193]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[191 165 183]\n",
      "  [186 164 182]\n",
      "  [186 167 184]\n",
      "  ...\n",
      "  [190 164 182]\n",
      "  [187 162 182]\n",
      "  [188 164 182]]\n",
      "\n",
      " [[190 164 182]\n",
      "  [188 164 182]\n",
      "  [185 166 183]\n",
      "  ...\n",
      "  [193 166 186]\n",
      "  [189 165 183]\n",
      "  [189 166 181]]\n",
      "\n",
      " [[188 165 180]\n",
      "  [186 164 182]\n",
      "  [186 164 182]\n",
      "  ...\n",
      "  [195 169 187]\n",
      "  [194 166 186]\n",
      "  [195 164 185]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024403.jpg\n",
      "ISIC_0024403\n",
      "[[[141 140 180]\n",
      "  [138 138 178]\n",
      "  [142 138 179]\n",
      "  ...\n",
      "  [179 172 207]\n",
      "  [175 170 201]\n",
      "  [173 167 198]]\n",
      "\n",
      " [[142 141 180]\n",
      "  [140 141 179]\n",
      "  [140 139 178]\n",
      "  ...\n",
      "  [180 175 207]\n",
      "  [173 170 202]\n",
      "  [170 167 199]]\n",
      "\n",
      " [[139 140 178]\n",
      "  [139 142 180]\n",
      "  [141 142 180]\n",
      "  ...\n",
      "  [174 170 205]\n",
      "  [170 166 201]\n",
      "  [171 164 201]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 99 102 147]\n",
      "  [ 94  98 147]\n",
      "  [ 94  98 146]\n",
      "  ...\n",
      "  [153 152 186]\n",
      "  [153 152 186]\n",
      "  [158 152 187]]\n",
      "\n",
      " [[ 97 100 145]\n",
      "  [ 95  97 145]\n",
      "  [ 98 100 148]\n",
      "  ...\n",
      "  [151 153 188]\n",
      "  [151 155 184]\n",
      "  [153 155 185]]\n",
      "\n",
      " [[ 98 101 146]\n",
      "  [ 97  99 147]\n",
      "  [ 97  99 147]\n",
      "  ...\n",
      "  [150 152 186]\n",
      "  [152 154 185]\n",
      "  [153 155 185]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024340.jpg\n",
      "ISIC_0024340\n",
      "[[[ 9  7 13]\n",
      "  [ 8  6 12]\n",
      "  [ 9  8 12]\n",
      "  ...\n",
      "  [19 22 36]\n",
      "  [17 20 34]\n",
      "  [16 19 33]]\n",
      "\n",
      " [[ 8  6 12]\n",
      "  [ 8  6 12]\n",
      "  [ 9  6 15]\n",
      "  ...\n",
      "  [21 21 37]\n",
      "  [20 21 35]\n",
      "  [17 18 32]]\n",
      "\n",
      " [[ 6  6 12]\n",
      "  [ 7  7 13]\n",
      "  [11 10 14]\n",
      "  ...\n",
      "  [24 25 39]\n",
      "  [22 22 36]\n",
      "  [20 20 34]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  1]\n",
      "  ...\n",
      "  [ 1  0  4]\n",
      "  [ 1  0  4]\n",
      "  [ 1  0  4]]\n",
      "\n",
      " [[ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  ...\n",
      "  [ 1  0  2]\n",
      "  [ 0  0  1]\n",
      "  [ 1  0  2]]\n",
      "\n",
      " [[ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  ...\n",
      "  [ 1  0  2]\n",
      "  [ 0  0  1]\n",
      "  [ 1  0  2]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024397.jpg\n",
      "ISIC_0024397\n",
      "[[[143 137 214]\n",
      "  [144 138 217]\n",
      "  [145 139 218]\n",
      "  ...\n",
      "  [104  85 172]\n",
      "  [107  88 173]\n",
      "  [111  87 175]]\n",
      "\n",
      " [[143 136 217]\n",
      "  [143 137 216]\n",
      "  [145 139 218]\n",
      "  ...\n",
      "  [ 95  82 158]\n",
      "  [ 94  78 161]\n",
      "  [102  83 168]]\n",
      "\n",
      " [[140 137 217]\n",
      "  [144 141 220]\n",
      "  [144 139 218]\n",
      "  ...\n",
      "  [115 104 172]\n",
      "  [102  93 160]\n",
      "  [101  90 162]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[130 122 187]\n",
      "  [132 121 191]\n",
      "  [134 121 189]\n",
      "  ...\n",
      "  [ 75  81 146]\n",
      "  [ 69  78 142]\n",
      "  [ 68  74 141]]\n",
      "\n",
      " [[121 120 184]\n",
      "  [127 121 186]\n",
      "  [129 121 186]\n",
      "  ...\n",
      "  [106 105 169]\n",
      "  [104 103 169]\n",
      "  [ 99  98 164]]\n",
      "\n",
      " [[125 121 186]\n",
      "  [127 121 186]\n",
      "  [126 118 183]\n",
      "  ...\n",
      "  [107 113 172]\n",
      "  [106 114 174]\n",
      "  [109 114 175]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024343.jpg\n",
      "ISIC_0024343\n",
      "[[[178 189 211]\n",
      "  [174 187 209]\n",
      "  [171 184 206]\n",
      "  ...\n",
      "  [164 177 199]\n",
      "  [162 175 197]\n",
      "  [160 172 196]]\n",
      "\n",
      " [[173 186 208]\n",
      "  [172 186 208]\n",
      "  [174 185 207]\n",
      "  ...\n",
      "  [163 175 199]\n",
      "  [164 175 197]\n",
      "  [162 171 198]]\n",
      "\n",
      " [[172 186 208]\n",
      "  [172 186 208]\n",
      "  [173 186 208]\n",
      "  ...\n",
      "  [160 174 197]\n",
      "  [160 172 196]\n",
      "  [160 168 197]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[127 149 185]\n",
      "  [127 146 184]\n",
      "  [125 145 186]\n",
      "  ...\n",
      "  [154 162 185]\n",
      "  [156 162 185]\n",
      "  [155 161 184]]\n",
      "\n",
      " [[129 146 185]\n",
      "  [127 143 185]\n",
      "  [120 136 179]\n",
      "  ...\n",
      "  [155 163 186]\n",
      "  [154 162 185]\n",
      "  [153 161 184]]\n",
      "\n",
      " [[120 135 181]\n",
      "  [112 126 174]\n",
      "  [104 118 166]\n",
      "  ...\n",
      "  [156 164 187]\n",
      "  [155 163 186]\n",
      "  [154 162 185]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024351.jpg\n",
      "ISIC_0024351\n",
      "[[[178 156 208]\n",
      "  [179 157 209]\n",
      "  [181 158 212]\n",
      "  ...\n",
      "  [186 166 211]\n",
      "  [185 165 210]\n",
      "  [182 165 208]]\n",
      "\n",
      " [[174 157 208]\n",
      "  [174 157 208]\n",
      "  [178 159 210]\n",
      "  ...\n",
      "  [183 168 206]\n",
      "  [181 168 206]\n",
      "  [183 167 208]]\n",
      "\n",
      " [[176 159 210]\n",
      "  [175 158 209]\n",
      "  [177 158 209]\n",
      "  ...\n",
      "  [183 167 208]\n",
      "  [179 168 208]\n",
      "  [182 170 206]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[188 173 211]\n",
      "  [187 172 210]\n",
      "  [188 173 211]\n",
      "  ...\n",
      "  [193 174 213]\n",
      "  [193 174 213]\n",
      "  [193 174 213]]\n",
      "\n",
      " [[188 172 213]\n",
      "  [188 173 211]\n",
      "  [186 173 211]\n",
      "  ...\n",
      "  [192 173 212]\n",
      "  [193 174 213]\n",
      "  [194 175 214]]\n",
      "\n",
      " [[187 172 210]\n",
      "  [188 173 211]\n",
      "  [186 173 211]\n",
      "  ...\n",
      "  [192 173 212]\n",
      "  [192 173 212]\n",
      "  [194 175 214]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024396.jpg\n",
      "ISIC_0024396\n",
      "[[[158 153 192]\n",
      "  [160 158 194]\n",
      "  [163 159 194]\n",
      "  ...\n",
      "  [101 107 142]\n",
      "  [100 105 138]\n",
      "  [104 106 140]]\n",
      "\n",
      " [[162 156 191]\n",
      "  [164 158 193]\n",
      "  [163 157 192]\n",
      "  ...\n",
      "  [100 106 141]\n",
      "  [100 106 141]\n",
      "  [100 106 141]]\n",
      "\n",
      " [[158 151 188]\n",
      "  [158 154 190]\n",
      "  [158 154 189]\n",
      "  ...\n",
      "  [108 113 146]\n",
      "  [100 106 141]\n",
      "  [103 107 142]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 90  91 125]\n",
      "  [ 93  94 128]\n",
      "  [ 92  94 125]\n",
      "  ...\n",
      "  [ 45  47  65]\n",
      "  [ 48  42  61]\n",
      "  [ 48  36  58]]\n",
      "\n",
      " [[ 89  91 122]\n",
      "  [ 89  91 122]\n",
      "  [ 92  94 124]\n",
      "  ...\n",
      "  [ 47  45  57]\n",
      "  [ 48  38  54]\n",
      "  [ 44  34  51]]\n",
      "\n",
      " [[ 81  89 119]\n",
      "  [ 82  88 123]\n",
      "  [ 84  92 121]\n",
      "  ...\n",
      "  [ 48  41  56]\n",
      "  [ 41  34  49]\n",
      "  [ 40  33  48]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024311.jpg\n",
      "ISIC_0024311\n",
      "[[[174 156 245]\n",
      "  [175 159 243]\n",
      "  [178 159 244]\n",
      "  ...\n",
      "  [173 163 240]\n",
      "  [172 162 239]\n",
      "  [173 163 240]]\n",
      "\n",
      " [[173 154 247]\n",
      "  [175 155 244]\n",
      "  [175 155 244]\n",
      "  ...\n",
      "  [177 165 241]\n",
      "  [176 163 241]\n",
      "  [176 163 241]]\n",
      "\n",
      " [[177 154 246]\n",
      "  [177 158 243]\n",
      "  [182 158 246]\n",
      "  ...\n",
      "  [175 163 239]\n",
      "  [175 162 240]\n",
      "  [174 163 243]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[148 142 219]\n",
      "  [149 143 220]\n",
      "  [147 141 218]\n",
      "  ...\n",
      "  [150 159 222]\n",
      "  [151 159 226]\n",
      "  [151 161 225]]\n",
      "\n",
      " [[149 148 222]\n",
      "  [151 148 220]\n",
      "  [145 146 220]\n",
      "  ...\n",
      "  [156 158 223]\n",
      "  [153 157 222]\n",
      "  [154 157 225]]\n",
      "\n",
      " [[155 153 223]\n",
      "  [154 150 222]\n",
      "  [151 148 220]\n",
      "  ...\n",
      "  [154 158 223]\n",
      "  [153 157 222]\n",
      "  [154 157 225]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024327.jpg\n",
      "ISIC_0024327\n",
      "[[[149 146 215]\n",
      "  [156 145 217]\n",
      "  [155 151 217]\n",
      "  ...\n",
      "  [138 132 203]\n",
      "  [128 124 199]\n",
      "  [139 133 204]]\n",
      "\n",
      " [[153 149 215]\n",
      "  [153 147 212]\n",
      "  [164 153 221]\n",
      "  ...\n",
      "  [136 135 201]\n",
      "  [127 128 196]\n",
      "  [137 129 199]]\n",
      "\n",
      " [[160 149 222]\n",
      "  [157 145 217]\n",
      "  [168 154 226]\n",
      "  ...\n",
      "  [139 135 200]\n",
      "  [135 130 199]\n",
      "  [134 125 198]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[142 144 202]\n",
      "  [144 143 205]\n",
      "  [145 146 207]\n",
      "  ...\n",
      "  [142 136 201]\n",
      "  [143 144 200]\n",
      "  [139 140 196]]\n",
      "\n",
      " [[143 142 206]\n",
      "  [147 141 206]\n",
      "  [148 144 203]\n",
      "  ...\n",
      "  [149 144 206]\n",
      "  [149 144 205]\n",
      "  [142 136 195]]\n",
      "\n",
      " [[141 138 207]\n",
      "  [146 142 208]\n",
      "  [150 145 206]\n",
      "  ...\n",
      "  [139 138 200]\n",
      "  [147 144 206]\n",
      "  [142 142 202]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024382.jpg\n",
      "ISIC_0024382\n",
      "[[[ 55  48  91]\n",
      "  [ 58  48  95]\n",
      "  [ 61  49 101]\n",
      "  ...\n",
      "  [ 44  33  66]\n",
      "  [ 40  30  66]\n",
      "  [ 37  32  64]]\n",
      "\n",
      " [[ 56  49  94]\n",
      "  [ 57  49  96]\n",
      "  [ 60  52  99]\n",
      "  ...\n",
      "  [ 45  33  69]\n",
      "  [ 42  35  68]\n",
      "  [ 40  34  69]]\n",
      "\n",
      " [[ 54  50  99]\n",
      "  [ 58  52  99]\n",
      "  [ 59  55  97]\n",
      "  ...\n",
      "  [ 43  33  75]\n",
      "  [ 43  35  72]\n",
      "  [ 42  35  70]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[102  94 141]\n",
      "  [107  95 143]\n",
      "  [108  96 144]\n",
      "  ...\n",
      "  [ 85  79 126]\n",
      "  [ 85  75 122]\n",
      "  [ 87  77 120]]\n",
      "\n",
      " [[100  94 141]\n",
      "  [104  97 142]\n",
      "  [105  95 142]\n",
      "  ...\n",
      "  [ 83  71 119]\n",
      "  [ 84  74 120]\n",
      "  [ 82  74 115]]\n",
      "\n",
      " [[100  94 141]\n",
      "  [105  95 142]\n",
      "  [105  95 141]\n",
      "  ...\n",
      "  [ 83  73 119]\n",
      "  [ 84  75 118]\n",
      "  [ 81  73 113]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ISIC_0024370.jpg\n",
      "ISIC_0024370\n",
      "[[[136 129 204]\n",
      "  [132 128 203]\n",
      "  [143 130 208]\n",
      "  ...\n",
      "  [150 142 203]\n",
      "  [150 144 203]\n",
      "  [152 146 205]]\n",
      "\n",
      " [[141 131 208]\n",
      "  [138 128 204]\n",
      "  [147 131 208]\n",
      "  ...\n",
      "  [146 144 204]\n",
      "  [153 142 204]\n",
      "  [152 144 207]]\n",
      "\n",
      " [[139 132 207]\n",
      "  [141 131 207]\n",
      "  [140 129 209]\n",
      "  ...\n",
      "  [143 144 205]\n",
      "  [143 144 202]\n",
      "  [146 144 204]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[159 151 204]\n",
      "  [160 151 202]\n",
      "  [157 147 200]\n",
      "  ...\n",
      "  [132 122 188]\n",
      "  [133 126 193]\n",
      "  [132 121 191]]\n",
      "\n",
      " [[160 149 205]\n",
      "  [163 152 208]\n",
      "  [164 149 206]\n",
      "  ...\n",
      "  [134 126 191]\n",
      "  [135 128 195]\n",
      "  [128 118 188]]\n",
      "\n",
      " [[160 153 204]\n",
      "  [159 149 202]\n",
      "  [164 151 203]\n",
      "  ...\n",
      "  [131 125 190]\n",
      "  [123 121 187]\n",
      "  [125 120 189]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024331.jpg\n",
      "ISIC_0024331\n",
      "[[[184 172 202]\n",
      "  [186 171 202]\n",
      "  [184 172 202]\n",
      "  ...\n",
      "  [146 152 197]\n",
      "  [146 151 196]\n",
      "  [143 149 194]]\n",
      "\n",
      " [[187 172 203]\n",
      "  [187 172 203]\n",
      "  [187 171 202]\n",
      "  ...\n",
      "  [145 151 196]\n",
      "  [145 150 195]\n",
      "  [145 150 195]]\n",
      "\n",
      " [[189 174 205]\n",
      "  [188 173 204]\n",
      "  [187 172 203]\n",
      "  ...\n",
      "  [145 151 196]\n",
      "  [144 150 195]\n",
      "  [143 149 194]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[134 146 194]\n",
      "  [137 144 194]\n",
      "  [140 147 197]\n",
      "  ...\n",
      "  [121 132 189]\n",
      "  [122 133 190]\n",
      "  [123 132 189]]\n",
      "\n",
      " [[137 142 197]\n",
      "  [135 141 194]\n",
      "  [141 143 197]\n",
      "  ...\n",
      "  [118 131 187]\n",
      "  [121 134 190]\n",
      "  [119 133 191]]\n",
      "\n",
      " [[136 143 198]\n",
      "  [134 140 193]\n",
      "  [138 142 195]\n",
      "  ...\n",
      "  [114 129 185]\n",
      "  [118 133 189]\n",
      "  [116 133 190]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024360.jpg\n",
      "ISIC_0024360\n",
      "[[[34 33 53]\n",
      "  [33 35 53]\n",
      "  [32 36 54]\n",
      "  ...\n",
      "  [33 26 39]\n",
      "  [28 24 36]\n",
      "  [27 23 35]]\n",
      "\n",
      " [[32 33 53]\n",
      "  [33 34 54]\n",
      "  [32 36 55]\n",
      "  ...\n",
      "  [33 26 39]\n",
      "  [29 25 37]\n",
      "  [28 24 36]]\n",
      "\n",
      " [[33 34 54]\n",
      "  [33 34 54]\n",
      "  [31 35 54]\n",
      "  ...\n",
      "  [31 26 41]\n",
      "  [29 23 40]\n",
      "  [29 23 42]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[55 50 77]\n",
      "  [59 54 83]\n",
      "  [61 58 84]\n",
      "  ...\n",
      "  [28 18 34]\n",
      "  [28 18 34]\n",
      "  [27 18 31]]\n",
      "\n",
      " [[54 51 73]\n",
      "  [59 53 78]\n",
      "  [57 52 79]\n",
      "  ...\n",
      "  [29 21 31]\n",
      "  [25 19 30]\n",
      "  [27 17 33]]\n",
      "\n",
      " [[51 50 70]\n",
      "  [53 50 72]\n",
      "  [54 51 76]\n",
      "  ...\n",
      "  [25 19 30]\n",
      "  [23 20 29]\n",
      "  [22 18 30]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024322.jpg\n",
      "ISIC_0024322\n",
      "[[[123  93 194]\n",
      "  [128  95 192]\n",
      "  [125  92 197]\n",
      "  ...\n",
      "  [123  92 197]\n",
      "  [122  88 196]\n",
      "  [126  90 196]]\n",
      "\n",
      " [[124  98 192]\n",
      "  [122  95 191]\n",
      "  [123  92 197]\n",
      "  ...\n",
      "  [120  84 196]\n",
      "  [117  86 195]\n",
      "  [124  94 199]]\n",
      "\n",
      " [[130  99 198]\n",
      "  [122  92 193]\n",
      "  [122  91 196]\n",
      "  ...\n",
      "  [118  88 195]\n",
      "  [114  85 188]\n",
      "  [120  89 198]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[139 146 209]\n",
      "  [140 148 208]\n",
      "  [141 146 207]\n",
      "  ...\n",
      "  [142 119 201]\n",
      "  [130 120 197]\n",
      "  [145 123 205]]\n",
      "\n",
      " [[139 147 207]\n",
      "  [142 151 215]\n",
      "  [143 150 213]\n",
      "  ...\n",
      "  [148 128 205]\n",
      "  [144 123 198]\n",
      "  [144 126 203]]\n",
      "\n",
      " [[138 148 212]\n",
      "  [136 148 212]\n",
      "  [138 149 211]\n",
      "  ...\n",
      "  [139 127 199]\n",
      "  [140 127 195]\n",
      "  [144 128 199]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024323.jpg\n",
      "ISIC_0024323\n",
      "[[[ 77  91 133]\n",
      "  [ 81  93 135]\n",
      "  [ 83  95 137]\n",
      "  ...\n",
      "  [ 65  75 115]\n",
      "  [ 68  77 120]\n",
      "  [ 67  76 119]]\n",
      "\n",
      " [[ 79  92 136]\n",
      "  [ 83  94 138]\n",
      "  [ 85  96 140]\n",
      "  ...\n",
      "  [ 67  76 119]\n",
      "  [ 68  77 120]\n",
      "  [ 68  77 120]]\n",
      "\n",
      " [[ 81  94 138]\n",
      "  [ 85  96 140]\n",
      "  [ 84  95 139]\n",
      "  ...\n",
      "  [ 69  78 121]\n",
      "  [ 69  78 121]\n",
      "  [ 68  78 118]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 73  85 121]\n",
      "  [ 73  84 122]\n",
      "  [ 74  85 123]\n",
      "  ...\n",
      "  [ 66  64 100]\n",
      "  [ 65  63  99]\n",
      "  [ 66  64 100]]\n",
      "\n",
      " [[ 71  82 120]\n",
      "  [ 71  82 120]\n",
      "  [ 71  81 121]\n",
      "  ...\n",
      "  [ 66  64 100]\n",
      "  [ 64  62  98]\n",
      "  [ 64  62  98]]\n",
      "\n",
      " [[ 71  81 121]\n",
      "  [ 71  81 121]\n",
      "  [ 71  81 121]\n",
      "  ...\n",
      "  [ 65  63  99]\n",
      "  [ 63  61  97]\n",
      "  [ 63  61  97]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024386.jpg\n",
      "ISIC_0024386\n",
      "[[[155 120 207]\n",
      "  [162 120 207]\n",
      "  [156 121 208]\n",
      "  ...\n",
      "  [183 149 213]\n",
      "  [178 144 205]\n",
      "  [175 142 203]]\n",
      "\n",
      " [[161 122 207]\n",
      "  [157 121 205]\n",
      "  [157 118 204]\n",
      "  ...\n",
      "  [183 149 214]\n",
      "  [174 145 208]\n",
      "  [177 146 207]]\n",
      "\n",
      " [[163 121 204]\n",
      "  [151 113 201]\n",
      "  [148 110 198]\n",
      "  ...\n",
      "  [174 147 211]\n",
      "  [174 146 212]\n",
      "  [183 149 213]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[157 131 195]\n",
      "  [161 134 198]\n",
      "  [157 134 196]\n",
      "  ...\n",
      "  [149 134 196]\n",
      "  [166 141 203]\n",
      "  [165 140 202]]\n",
      "\n",
      " [[156 128 194]\n",
      "  [167 134 201]\n",
      "  [166 129 197]\n",
      "  ...\n",
      "  [153 144 201]\n",
      "  [157 140 197]\n",
      "  [151 138 194]]\n",
      "\n",
      " [[154 127 193]\n",
      "  [160 129 196]\n",
      "  [166 129 197]\n",
      "  ...\n",
      "  [162 143 200]\n",
      "  [158 138 197]\n",
      "  [160 138 196]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024389.jpg\n",
      "ISIC_0024389\n",
      "[[[167 135 216]\n",
      "  [170 134 218]\n",
      "  [169 134 220]\n",
      "  ...\n",
      "  [106  82 166]\n",
      "  [101  72 157]\n",
      "  [144 111 196]]\n",
      "\n",
      " [[173 134 219]\n",
      "  [170 134 218]\n",
      "  [168 130 218]\n",
      "  ...\n",
      "  [107  84 166]\n",
      "  [100  71 157]\n",
      "  [122  87 177]]\n",
      "\n",
      " [[174 133 224]\n",
      "  [178 135 226]\n",
      "  [171 133 223]\n",
      "  ...\n",
      "  [103  81 160]\n",
      "  [ 99  80 161]\n",
      "  [102  79 161]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[137 111 195]\n",
      "  [134 111 193]\n",
      "  [134 107 193]\n",
      "  ...\n",
      "  [ 62  50 140]\n",
      "  [ 61  50 143]\n",
      "  [ 62  52 142]]\n",
      "\n",
      " [[136 112 196]\n",
      "  [135 111 195]\n",
      "  [135 111 195]\n",
      "  ...\n",
      "  [ 60  48 138]\n",
      "  [ 60  51 137]\n",
      "  [ 59  50 137]]\n",
      "\n",
      " [[135 111 195]\n",
      "  [134 110 194]\n",
      "  [135 111 195]\n",
      "  ...\n",
      "  [ 63  48 139]\n",
      "  [ 59  50 137]\n",
      "  [ 57  48 134]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024387.jpg\n",
      "ISIC_0024387\n",
      "[[[ 0  1 11]\n",
      "  [ 1  0  9]\n",
      "  [ 1  0  9]\n",
      "  ...\n",
      "  [18 23 38]\n",
      "  [20 26 39]\n",
      "  [20 23 37]]\n",
      "\n",
      " [[ 0  2 10]\n",
      "  [ 0  0  8]\n",
      "  [ 1  1  7]\n",
      "  ...\n",
      "  [17 23 36]\n",
      "  [18 24 37]\n",
      "  [19 25 38]]\n",
      "\n",
      " [[ 1  3 11]\n",
      "  [ 2  2  8]\n",
      "  [ 2  1 10]\n",
      "  ...\n",
      "  [16 22 35]\n",
      "  [15 23 36]\n",
      "  [17 23 36]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[15 21 34]\n",
      "  [15 18 32]\n",
      "  [15 18 32]\n",
      "  ...\n",
      "  [15 16 30]\n",
      "  [14 14 30]\n",
      "  [13 15 26]]\n",
      "\n",
      " [[16 22 33]\n",
      "  [16 22 35]\n",
      "  [15 20 35]\n",
      "  ...\n",
      "  [12 16 27]\n",
      "  [12 16 27]\n",
      "  [13 17 28]]\n",
      "\n",
      " [[16 22 33]\n",
      "  [18 24 37]\n",
      "  [16 21 36]\n",
      "  ...\n",
      "  [12 16 27]\n",
      "  [12 16 27]\n",
      "  [13 17 28]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024393.jpg\n",
      "ISIC_0024393\n",
      "[[[151 142 232]\n",
      "  [149 141 228]\n",
      "  [148 139 229]\n",
      "  ...\n",
      "  [151 133 227]\n",
      "  [146 133 225]\n",
      "  [152 134 227]]\n",
      "\n",
      " [[144 139 230]\n",
      "  [145 142 228]\n",
      "  [143 141 230]\n",
      "  ...\n",
      "  [145 131 225]\n",
      "  [148 133 224]\n",
      "  [148 133 225]]\n",
      "\n",
      " [[149 138 234]\n",
      "  [152 142 235]\n",
      "  [151 143 236]\n",
      "  ...\n",
      "  [143 132 225]\n",
      "  [142 133 224]\n",
      "  [141 134 225]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[127 139 211]\n",
      "  [128 140 210]\n",
      "  [134 144 214]\n",
      "  ...\n",
      "  [129 140 208]\n",
      "  [128 139 207]\n",
      "  [133 143 213]]\n",
      "\n",
      " [[123 136 210]\n",
      "  [124 137 211]\n",
      "  [129 141 213]\n",
      "  ...\n",
      "  [129 145 212]\n",
      "  [125 141 208]\n",
      "  [133 142 210]]\n",
      "\n",
      " [[125 139 211]\n",
      "  [126 140 212]\n",
      "  [128 142 214]\n",
      "  ...\n",
      "  [129 145 212]\n",
      "  [128 141 209]\n",
      "  [129 142 210]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024344.jpg\n",
      "ISIC_0024344\n",
      "[[[41 37 66]\n",
      "  [35 39 64]\n",
      "  [42 42 66]\n",
      "  ...\n",
      "  [24 24 48]\n",
      "  [22 24 48]\n",
      "  [21 25 49]]\n",
      "\n",
      " [[44 39 66]\n",
      "  [47 44 70]\n",
      "  [43 47 66]\n",
      "  ...\n",
      "  [20 24 48]\n",
      "  [23 26 47]\n",
      "  [21 24 45]]\n",
      "\n",
      " [[46 41 68]\n",
      "  [47 43 72]\n",
      "  [47 49 73]\n",
      "  ...\n",
      "  [20 25 46]\n",
      "  [19 24 45]\n",
      "  [24 25 46]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[27 26 46]\n",
      "  [28 27 47]\n",
      "  [28 26 46]\n",
      "  ...\n",
      "  [ 0  3 17]\n",
      "  [ 1  5 16]\n",
      "  [ 2  6 17]]\n",
      "\n",
      " [[27 26 46]\n",
      "  [27 26 46]\n",
      "  [30 29 49]\n",
      "  ...\n",
      "  [ 2  7 16]\n",
      "  [ 2  6 17]\n",
      "  [ 1  5 16]]\n",
      "\n",
      " [[26 27 47]\n",
      "  [25 26 46]\n",
      "  [28 29 49]\n",
      "  ...\n",
      "  [ 2  7 16]\n",
      "  [ 2  6 17]\n",
      "  [ 1  5 16]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024325.jpg\n",
      "ISIC_0024325\n",
      "[[[135 126 223]\n",
      "  [140 126 227]\n",
      "  [135 125 225]\n",
      "  ...\n",
      "  [144 130 231]\n",
      "  [138 131 228]\n",
      "  [143 132 229]]\n",
      "\n",
      " [[143 125 226]\n",
      "  [139 125 227]\n",
      "  [137 123 228]\n",
      "  ...\n",
      "  [140 131 228]\n",
      "  [142 131 228]\n",
      "  [143 134 231]]\n",
      "\n",
      " [[134 127 224]\n",
      "  [134 123 225]\n",
      "  [132 124 224]\n",
      "  ...\n",
      "  [136 130 225]\n",
      "  [141 130 226]\n",
      "  [144 133 229]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[119 130 214]\n",
      "  [121 131 215]\n",
      "  [122 131 218]\n",
      "  ...\n",
      "  [134 140 213]\n",
      "  [133 139 214]\n",
      "  [136 142 217]]\n",
      "\n",
      " [[118 131 210]\n",
      "  [118 130 212]\n",
      "  [121 134 213]\n",
      "  ...\n",
      "  [134 136 214]\n",
      "  [133 135 213]\n",
      "  [133 137 216]]\n",
      "\n",
      " [[117 129 211]\n",
      "  [116 130 212]\n",
      "  [120 133 212]\n",
      "  ...\n",
      "  [138 138 214]\n",
      "  [134 134 210]\n",
      "  [137 137 215]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024348.jpg\n",
      "ISIC_0024348\n",
      "[[[107 104 203]\n",
      "  [119 111 206]\n",
      "  [123 115 210]\n",
      "  ...\n",
      "  [117 121 200]\n",
      "  [114 121 200]\n",
      "  [118 122 201]]\n",
      "\n",
      " [[107 105 201]\n",
      "  [122 117 209]\n",
      "  [123 118 210]\n",
      "  ...\n",
      "  [117 124 204]\n",
      "  [115 125 203]\n",
      "  [117 120 201]]\n",
      "\n",
      " [[112 105 209]\n",
      "  [121 113 213]\n",
      "  [119 115 211]\n",
      "  ...\n",
      "  [113 126 202]\n",
      "  [114 124 202]\n",
      "  [117 124 204]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[104 117 193]\n",
      "  [107 114 193]\n",
      "  [103 120 193]\n",
      "  ...\n",
      "  [108 116 193]\n",
      "  [105 113 190]\n",
      "  [102 111 191]]\n",
      "\n",
      " [[101 115 191]\n",
      "  [107 111 190]\n",
      "  [105 118 196]\n",
      "  ...\n",
      "  [112 114 192]\n",
      "  [107 111 189]\n",
      "  [107 111 192]]\n",
      "\n",
      " [[108 118 195]\n",
      "  [112 116 195]\n",
      "  [114 118 197]\n",
      "  ...\n",
      "  [103 113 191]\n",
      "  [100 112 190]\n",
      "  [ 98 114 191]]]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024385.jpg\n",
      "ISIC_0024385\n",
      "[[[155 142 234]\n",
      "  [158 146 234]\n",
      "  [155 143 233]\n",
      "  ...\n",
      "  [171 161 238]\n",
      "  [169 159 235]\n",
      "  [166 159 234]]\n",
      "\n",
      " [[153 144 235]\n",
      "  [155 146 236]\n",
      "  [153 143 233]\n",
      "  ...\n",
      "  [170 159 232]\n",
      "  [166 159 232]\n",
      "  [166 157 230]]\n",
      "\n",
      " [[154 141 233]\n",
      "  [158 143 235]\n",
      "  [156 140 234]\n",
      "  ...\n",
      "  [171 160 232]\n",
      "  [170 161 234]\n",
      "  [168 159 232]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[140 133 206]\n",
      "  [135 131 206]\n",
      "  [137 130 203]\n",
      "  ...\n",
      "  [150 146 212]\n",
      "  [146 145 209]\n",
      "  [145 144 210]]\n",
      "\n",
      " [[134 134 204]\n",
      "  [132 132 202]\n",
      "  [134 130 201]\n",
      "  ...\n",
      "  [154 147 214]\n",
      "  [154 146 211]\n",
      "  [155 145 211]]\n",
      "\n",
      " [[136 131 206]\n",
      "  [134 129 204]\n",
      "  [134 129 204]\n",
      "  ...\n",
      "  [155 147 217]\n",
      "  [155 146 213]\n",
      "  [155 145 215]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024337.jpg\n",
      "ISIC_0024337\n",
      "[[[148 147 215]\n",
      "  [148 147 215]\n",
      "  [141 147 214]\n",
      "  ...\n",
      "  [151 149 208]\n",
      "  [148 148 208]\n",
      "  [145 142 204]]\n",
      "\n",
      " [[147 149 213]\n",
      "  [142 143 211]\n",
      "  [143 147 212]\n",
      "  ...\n",
      "  [149 144 206]\n",
      "  [147 146 210]\n",
      "  [150 144 209]]\n",
      "\n",
      " [[145 148 209]\n",
      "  [142 147 208]\n",
      "  [140 145 206]\n",
      "  ...\n",
      "  [146 141 203]\n",
      "  [143 143 203]\n",
      "  [148 146 206]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[134 138 197]\n",
      "  [134 137 198]\n",
      "  [130 135 198]\n",
      "  ...\n",
      "  [139 140 200]\n",
      "  [135 139 198]\n",
      "  [127 131 190]]\n",
      "\n",
      " [[136 137 197]\n",
      "  [126 136 196]\n",
      "  [121 135 194]\n",
      "  ...\n",
      "  [126 134 187]\n",
      "  [129 134 189]\n",
      "  [132 133 194]]\n",
      "\n",
      " [[135 136 197]\n",
      "  [126 134 194]\n",
      "  [125 133 192]\n",
      "  ...\n",
      "  [135 134 196]\n",
      "  [136 135 197]\n",
      "  [135 135 195]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024346.jpg\n",
      "ISIC_0024346\n",
      "[[[162 156 238]\n",
      "  [161 153 240]\n",
      "  [159 150 240]\n",
      "  ...\n",
      "  [173 149 243]\n",
      "  [173 149 243]\n",
      "  [178 153 245]]\n",
      "\n",
      " [[157 151 238]\n",
      "  [160 152 239]\n",
      "  [156 151 236]\n",
      "  ...\n",
      "  [172 151 244]\n",
      "  [170 149 242]\n",
      "  [172 151 243]]\n",
      "\n",
      " [[163 152 238]\n",
      "  [160 151 238]\n",
      "  [157 149 236]\n",
      "  ...\n",
      "  [167 149 242]\n",
      "  [168 151 242]\n",
      "  [168 151 242]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[140 139 219]\n",
      "  [142 142 218]\n",
      "  [144 142 218]\n",
      "  ...\n",
      "  [161 161 231]\n",
      "  [161 162 230]\n",
      "  [162 160 230]]\n",
      "\n",
      " [[143 142 222]\n",
      "  [141 141 219]\n",
      "  [143 143 221]\n",
      "  ...\n",
      "  [159 159 229]\n",
      "  [161 159 229]\n",
      "  [156 158 228]]\n",
      "\n",
      " [[149 144 223]\n",
      "  [146 145 225]\n",
      "  [148 145 225]\n",
      "  ...\n",
      "  [153 158 227]\n",
      "  [156 158 229]\n",
      "  [158 157 229]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024317.jpg\n",
      "ISIC_0024317\n",
      "[[[222 202 214]\n",
      "  [223 201 213]\n",
      "  [227 203 215]\n",
      "  ...\n",
      "  [217 203 215]\n",
      "  [217 203 214]\n",
      "  [218 204 216]]\n",
      "\n",
      " [[222 200 212]\n",
      "  [220 200 212]\n",
      "  [223 203 215]\n",
      "  ...\n",
      "  [210 198 210]\n",
      "  [211 199 211]\n",
      "  [209 199 211]]\n",
      "\n",
      " [[219 199 211]\n",
      "  [221 201 213]\n",
      "  [224 202 214]\n",
      "  ...\n",
      "  [208 198 210]\n",
      "  [210 200 213]\n",
      "  [208 198 211]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[199 189 206]\n",
      "  [199 192 207]\n",
      "  [197 190 205]\n",
      "  ...\n",
      "  [208 199 212]\n",
      "  [209 200 213]\n",
      "  [208 198 211]]\n",
      "\n",
      " [[200 190 206]\n",
      "  [198 188 205]\n",
      "  [196 184 202]\n",
      "  ...\n",
      "  [208 196 214]\n",
      "  [206 195 211]\n",
      "  [205 195 208]]\n",
      "\n",
      " [[200 189 205]\n",
      "  [199 189 206]\n",
      "  [196 186 203]\n",
      "  ...\n",
      "  [208 196 214]\n",
      "  [206 196 213]\n",
      "  [204 194 210]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024400.jpg\n",
      "ISIC_0024400\n",
      "[[[18 14 26]\n",
      "  [17 11 28]\n",
      "  [18 12 29]\n",
      "  ...\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]]\n",
      "\n",
      " [[18 14 26]\n",
      "  [17 12 27]\n",
      "  [18 13 28]\n",
      "  ...\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]]\n",
      "\n",
      " [[18 13 28]\n",
      "  [18 13 28]\n",
      "  [17 11 28]\n",
      "  ...\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]\n",
      "  [ 0  0  0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[13  9 21]\n",
      "  [13  9 20]\n",
      "  [14  8 19]\n",
      "  ...\n",
      "  [70 61 94]\n",
      "  [72 61 93]\n",
      "  [72 59 91]]\n",
      "\n",
      " [[14 10 21]\n",
      "  [12  8 19]\n",
      "  [13  7 18]\n",
      "  ...\n",
      "  [72 61 94]\n",
      "  [71 59 95]\n",
      "  [70 59 92]]\n",
      "\n",
      " [[13  9 20]\n",
      "  [13  9 20]\n",
      "  [14  8 19]\n",
      "  ...\n",
      "  [70 59 92]\n",
      "  [70 59 92]\n",
      "  [67 56 89]]]\n",
      "[[[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]]]\n",
      "ISIC_0024358.jpg\n",
      "ISIC_0024358\n",
      "[[[140 145 190]\n",
      "  [137 143 186]\n",
      "  [133 139 184]\n",
      "  ...\n",
      "  [164 160 195]\n",
      "  [163 159 194]\n",
      "  [162 158 193]]\n",
      "\n",
      " [[142 145 190]\n",
      "  [137 140 185]\n",
      "  [139 141 189]\n",
      "  ...\n",
      "  [162 158 194]\n",
      "  [162 158 193]\n",
      "  [162 158 194]]\n",
      "\n",
      " [[146 146 192]\n",
      "  [144 144 192]\n",
      "  [143 146 191]\n",
      "  ...\n",
      "  [157 158 192]\n",
      "  [163 157 192]\n",
      "  [162 158 193]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[129 127 187]\n",
      "  [131 129 188]\n",
      "  [130 132 186]\n",
      "  ...\n",
      "  [136 130 177]\n",
      "  [133 127 176]\n",
      "  [131 122 173]]\n",
      "\n",
      " [[129 127 186]\n",
      "  [128 128 188]\n",
      "  [129 131 185]\n",
      "  ...\n",
      "  [135 129 176]\n",
      "  [135 129 178]\n",
      "  [130 123 174]]\n",
      "\n",
      " [[126 127 185]\n",
      "  [127 128 186]\n",
      "  [129 130 186]\n",
      "  ...\n",
      "  [134 128 175]\n",
      "  [135 129 178]\n",
      "  [133 126 177]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024384.jpg\n",
      "ISIC_0024384\n",
      "[[[137 126 218]\n",
      "  [135 126 217]\n",
      "  [134 127 218]\n",
      "  ...\n",
      "  [ 69  75 152]\n",
      "  [ 61  68 148]\n",
      "  [ 57  64 143]]\n",
      "\n",
      " [[133 126 217]\n",
      "  [130 126 215]\n",
      "  [133 127 216]\n",
      "  ...\n",
      "  [ 78  82 163]\n",
      "  [ 70  75 154]\n",
      "  [ 70  71 151]]\n",
      "\n",
      " [[132 128 217]\n",
      "  [132 125 216]\n",
      "  [131 125 214]\n",
      "  ...\n",
      "  [ 83  87 165]\n",
      "  [ 78  81 162]\n",
      "  [ 77  81 164]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[109 119 197]\n",
      "  [113 123 200]\n",
      "  [117 128 202]\n",
      "  ...\n",
      "  [120 134 200]\n",
      "  [117 131 197]\n",
      "  [121 133 199]]\n",
      "\n",
      " [[116 124 201]\n",
      "  [112 122 199]\n",
      "  [114 127 201]\n",
      "  ...\n",
      "  [122 135 197]\n",
      "  [120 133 195]\n",
      "  [125 136 198]]\n",
      "\n",
      " [[115 128 202]\n",
      "  [113 126 200]\n",
      "  [115 129 201]\n",
      "  ...\n",
      "  [124 136 202]\n",
      "  [124 136 202]\n",
      "  [125 134 201]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024394.jpg\n",
      "ISIC_0024394\n",
      "[[[197 191 246]\n",
      "  [196 190 245]\n",
      "  [199 193 246]\n",
      "  ...\n",
      "  [172 172 226]\n",
      "  [172 172 226]\n",
      "  [171 171 225]]\n",
      "\n",
      " [[199 193 248]\n",
      "  [198 192 245]\n",
      "  [196 190 243]\n",
      "  ...\n",
      "  [171 171 225]\n",
      "  [171 171 225]\n",
      "  [172 172 226]]\n",
      "\n",
      " [[199 191 244]\n",
      "  [196 190 243]\n",
      "  [195 189 242]\n",
      "  ...\n",
      "  [169 169 223]\n",
      "  [170 170 224]\n",
      "  [170 170 224]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[169 171 211]\n",
      "  [174 174 214]\n",
      "  [181 177 218]\n",
      "  ...\n",
      "  [195 181 223]\n",
      "  [196 179 222]\n",
      "  [194 175 218]]\n",
      "\n",
      " [[167 176 214]\n",
      "  [173 175 215]\n",
      "  [182 176 217]\n",
      "  ...\n",
      "  [203 184 227]\n",
      "  [202 183 226]\n",
      "  [197 178 221]]\n",
      "\n",
      " [[168 179 217]\n",
      "  [173 178 217]\n",
      "  [185 179 220]\n",
      "  ...\n",
      "  [203 184 227]\n",
      "  [204 184 227]\n",
      "  [199 179 222]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024334.jpg\n",
      "ISIC_0024334\n",
      "[[[156 151 226]\n",
      "  [159 153 230]\n",
      "  [161 152 232]\n",
      "  ...\n",
      "  [160 153 228]\n",
      "  [161 152 232]\n",
      "  [161 151 228]]\n",
      "\n",
      " [[156 149 230]\n",
      "  [163 152 232]\n",
      "  [161 153 230]\n",
      "  ...\n",
      "  [161 152 232]\n",
      "  [160 154 233]\n",
      "  [158 149 229]]\n",
      "\n",
      " [[156 148 231]\n",
      "  [162 150 232]\n",
      "  [163 150 236]\n",
      "  ...\n",
      "  [162 151 231]\n",
      "  [159 150 230]\n",
      "  [159 151 228]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[135 137 215]\n",
      "  [131 131 209]\n",
      "  [128 130 208]\n",
      "  ...\n",
      "  [151 147 212]\n",
      "  [150 146 211]\n",
      "  [152 145 212]]\n",
      "\n",
      " [[134 139 214]\n",
      "  [133 133 209]\n",
      "  [130 131 211]\n",
      "  ...\n",
      "  [149 141 204]\n",
      "  [152 144 209]\n",
      "  [151 147 212]]\n",
      "\n",
      " [[136 139 214]\n",
      "  [135 135 211]\n",
      "  [130 130 208]\n",
      "  ...\n",
      "  [143 138 207]\n",
      "  [146 140 211]\n",
      "  [148 147 215]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024359.jpg\n",
      "ISIC_0024359\n",
      "[[[112 113 204]\n",
      "  [119 116 208]\n",
      "  [121 119 209]\n",
      "  ...\n",
      "  [126 125 217]\n",
      "  [127 120 217]\n",
      "  [126 120 215]]\n",
      "\n",
      " [[117 113 208]\n",
      "  [117 116 210]\n",
      "  [119 118 212]\n",
      "  ...\n",
      "  [123 123 213]\n",
      "  [124 121 213]\n",
      "  [123 119 214]]\n",
      "\n",
      " [[123 115 210]\n",
      "  [124 118 213]\n",
      "  [121 118 210]\n",
      "  ...\n",
      "  [117 117 207]\n",
      "  [120 120 210]\n",
      "  [122 123 211]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[103 107 195]\n",
      "  [ 96 103 190]\n",
      "  [ 97 103 192]\n",
      "  ...\n",
      "  [101 113 191]\n",
      "  [ 98 112 188]\n",
      "  [ 98 115 188]]\n",
      "\n",
      " [[ 98 107 194]\n",
      "  [ 96 104 193]\n",
      "  [100 101 192]\n",
      "  ...\n",
      "  [101 114 192]\n",
      "  [ 96 109 187]\n",
      "  [ 98 112 188]]\n",
      "\n",
      " [[ 97 108 192]\n",
      "  [ 98 107 194]\n",
      "  [106 110 199]\n",
      "  ...\n",
      "  [100 121 189]\n",
      "  [101 119 190]\n",
      "  [101 119 190]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024365.jpg\n",
      "ISIC_0024365\n",
      "[[[117 130 216]\n",
      "  [120 134 217]\n",
      "  [122 130 219]\n",
      "  ...\n",
      "  [121 136 209]\n",
      "  [115 131 208]\n",
      "  [125 135 213]]\n",
      "\n",
      " [[119 129 217]\n",
      "  [115 129 211]\n",
      "  [116 127 211]\n",
      "  ...\n",
      "  [121 137 204]\n",
      "  [123 139 206]\n",
      "  [121 138 211]]\n",
      "\n",
      " [[126 131 216]\n",
      "  [118 126 215]\n",
      "  [120 126 215]\n",
      "  ...\n",
      "  [127 138 206]\n",
      "  [135 142 215]\n",
      "  [135 142 211]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[126 132 203]\n",
      "  [124 132 202]\n",
      "  [122 133 201]\n",
      "  ...\n",
      "  [120 129 197]\n",
      "  [113 128 197]\n",
      "  [113 125 197]]\n",
      "\n",
      " [[122 135 203]\n",
      "  [120 136 202]\n",
      "  [120 134 200]\n",
      "  ...\n",
      "  [114 121 194]\n",
      "  [118 124 201]\n",
      "  [112 121 195]]\n",
      "\n",
      " [[121 130 203]\n",
      "  [126 140 206]\n",
      "  [129 144 207]\n",
      "  ...\n",
      "  [108 123 196]\n",
      "  [105 122 195]\n",
      "  [105 120 193]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024368.jpg\n",
      "ISIC_0024368\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[ 89  94 196]\n",
      "  [ 85  90 192]\n",
      "  [ 79  89 190]\n",
      "  ...\n",
      "  [142 126 197]\n",
      "  [137 125 191]\n",
      "  [134 120 191]]\n",
      "\n",
      " [[ 75  88 186]\n",
      "  [ 82  93 191]\n",
      "  [ 85  97 191]\n",
      "  ...\n",
      "  [135 124 194]\n",
      "  [135 124 196]\n",
      "  [133 125 196]]\n",
      "\n",
      " [[ 77  89 185]\n",
      "  [ 87  94 191]\n",
      "  [ 90  98 188]\n",
      "  ...\n",
      "  [136 121 195]\n",
      "  [135 119 196]\n",
      "  [128 115 191]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 65  51 109]\n",
      "  [ 74  60 118]\n",
      "  [ 82  68 126]\n",
      "  ...\n",
      "  [120 113 174]\n",
      "  [114 112 172]\n",
      "  [117 112 173]]\n",
      "\n",
      " [[ 74  66 119]\n",
      "  [ 82  73 130]\n",
      "  [ 94  82 142]\n",
      "  ...\n",
      "  [121 110 173]\n",
      "  [118 110 171]\n",
      "  [120 112 173]]\n",
      "\n",
      " [[ 91  82 145]\n",
      "  [102  93 156]\n",
      "  [117 105 171]\n",
      "  ...\n",
      "  [120 112 173]\n",
      "  [115 108 169]\n",
      "  [116 108 171]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024330.jpg\n",
      "ISIC_0024330\n",
      "[[[185 157 180]\n",
      "  [190 155 181]\n",
      "  [189 155 185]\n",
      "  ...\n",
      "  [178 146 177]\n",
      "  [178 147 180]\n",
      "  [178 147 174]]\n",
      "\n",
      " [[186 157 182]\n",
      "  [189 157 182]\n",
      "  [190 155 181]\n",
      "  ...\n",
      "  [174 146 176]\n",
      "  [175 146 179]\n",
      "  [177 147 176]]\n",
      "\n",
      " [[187 156 183]\n",
      "  [188 155 182]\n",
      "  [189 156 183]\n",
      "  ...\n",
      "  [173 145 175]\n",
      "  [174 144 179]\n",
      "  [174 146 175]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[165 137 160]\n",
      "  [167 138 163]\n",
      "  [163 136 162]\n",
      "  ...\n",
      "  [159 137 161]\n",
      "  [159 137 161]\n",
      "  [159 137 162]]\n",
      "\n",
      " [[161 135 159]\n",
      "  [162 136 160]\n",
      "  [163 137 161]\n",
      "  ...\n",
      "  [159 136 164]\n",
      "  [160 138 162]\n",
      "  [160 138 162]]\n",
      "\n",
      " [[163 134 159]\n",
      "  [164 135 160]\n",
      "  [165 136 161]\n",
      "  ...\n",
      "  [157 135 159]\n",
      "  [159 137 161]\n",
      "  [157 135 160]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024391.jpg\n",
      "ISIC_0024391\n",
      "[[[111 102 213]\n",
      "  [117 103 215]\n",
      "  [119 107 213]\n",
      "  ...\n",
      "  [ 98  92 201]\n",
      "  [101  89 201]\n",
      "  [105  91 203]]\n",
      "\n",
      " [[114 102 214]\n",
      "  [117 104 212]\n",
      "  [121 107 213]\n",
      "  ...\n",
      "  [ 97  91 200]\n",
      "  [101  90 200]\n",
      "  [104  91 199]]\n",
      "\n",
      " [[115 101 213]\n",
      "  [119 106 214]\n",
      "  [118 111 216]\n",
      "  ...\n",
      "  [ 99  91 198]\n",
      "  [ 99  91 198]\n",
      "  [ 96  90 197]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 89  82 186]\n",
      "  [ 93  81 187]\n",
      "  [ 93  82 186]\n",
      "  ...\n",
      "  [119 120 206]\n",
      "  [114 120 203]\n",
      "  [121 125 208]]\n",
      "\n",
      " [[ 98  84 189]\n",
      "  [ 98  80 187]\n",
      "  [ 99  81 188]\n",
      "  ...\n",
      "  [113 123 201]\n",
      "  [116 125 205]\n",
      "  [121 128 208]]\n",
      "\n",
      " [[ 97  82 187]\n",
      "  [ 98  82 189]\n",
      "  [ 98  84 190]\n",
      "  ...\n",
      "  [116 121 200]\n",
      "  [117 121 202]\n",
      "  [119 123 204]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024399.jpg\n",
      "ISIC_0024399\n",
      "[[[113 116 214]\n",
      "  [105 111 208]\n",
      "  [108 112 207]\n",
      "  ...\n",
      "  [137 127 217]\n",
      "  [134 125 212]\n",
      "  [134 125 215]]\n",
      "\n",
      " [[108 119 211]\n",
      "  [104 117 209]\n",
      "  [109 116 213]\n",
      "  ...\n",
      "  [131 126 217]\n",
      "  [131 124 215]\n",
      "  [135 125 218]]\n",
      "\n",
      " [[107 118 210]\n",
      "  [109 120 212]\n",
      "  [113 121 215]\n",
      "  ...\n",
      "  [132 125 216]\n",
      "  [131 126 217]\n",
      "  [129 130 218]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[109 114 199]\n",
      "  [108 115 200]\n",
      "  [110 121 203]\n",
      "  ...\n",
      "  [100 116 193]\n",
      "  [104 116 194]\n",
      "  [107 120 196]]\n",
      "\n",
      " [[109 115 198]\n",
      "  [106 116 200]\n",
      "  [108 121 207]\n",
      "  ...\n",
      "  [104 117 196]\n",
      "  [104 118 194]\n",
      "  [104 118 194]]\n",
      "\n",
      " [[112 114 202]\n",
      "  [113 116 201]\n",
      "  [109 119 203]\n",
      "  ...\n",
      "  [103 116 194]\n",
      "  [105 118 196]\n",
      "  [105 118 197]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024352.jpg\n",
      "ISIC_0024352\n",
      "[[[147 135 165]\n",
      "  [149 137 167]\n",
      "  [150 138 166]\n",
      "  ...\n",
      "  [166 147 172]\n",
      "  [165 149 173]\n",
      "  [161 147 171]]\n",
      "\n",
      " [[147 135 163]\n",
      "  [148 136 166]\n",
      "  [149 137 165]\n",
      "  ...\n",
      "  [164 148 172]\n",
      "  [163 147 171]\n",
      "  [160 144 168]]\n",
      "\n",
      " [[149 137 167]\n",
      "  [148 136 166]\n",
      "  [148 136 164]\n",
      "  ...\n",
      "  [163 147 171]\n",
      "  [161 145 169]\n",
      "  [160 144 168]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[156 143 167]\n",
      "  [156 143 167]\n",
      "  [155 142 168]\n",
      "  ...\n",
      "  [155 141 165]\n",
      "  [155 141 165]\n",
      "  [156 142 166]]\n",
      "\n",
      " [[156 142 166]\n",
      "  [156 142 166]\n",
      "  [156 142 166]\n",
      "  ...\n",
      "  [154 140 164]\n",
      "  [155 141 165]\n",
      "  [155 141 165]]\n",
      "\n",
      " [[156 142 166]\n",
      "  [156 142 166]\n",
      "  [157 143 167]\n",
      "  ...\n",
      "  [154 140 164]\n",
      "  [155 141 165]\n",
      "  [155 141 165]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024398.jpg\n",
      "ISIC_0024398\n",
      "[[[158 142 226]\n",
      "  [161 144 231]\n",
      "  [162 145 232]\n",
      "  ...\n",
      "  [175 159 237]\n",
      "  [169 153 230]\n",
      "  [167 153 231]]\n",
      "\n",
      " [[158 142 225]\n",
      "  [159 143 226]\n",
      "  [159 145 227]\n",
      "  ...\n",
      "  [178 159 240]\n",
      "  [175 154 233]\n",
      "  [173 152 231]]\n",
      "\n",
      " [[162 144 227]\n",
      "  [159 140 225]\n",
      "  [155 138 225]\n",
      "  ...\n",
      "  [176 157 236]\n",
      "  [176 158 235]\n",
      "  [171 155 232]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[113 128 207]\n",
      "  [111 125 208]\n",
      "  [108 124 207]\n",
      "  ...\n",
      "  [149 143 214]\n",
      "  [146 142 213]\n",
      "  [149 146 218]]\n",
      "\n",
      " [[108 128 205]\n",
      "  [106 123 204]\n",
      "  [103 120 201]\n",
      "  ...\n",
      "  [152 141 213]\n",
      "  [147 141 212]\n",
      "  [154 147 222]]\n",
      "\n",
      " [[103 126 202]\n",
      "  [ 96 118 196]\n",
      "  [ 92 113 194]\n",
      "  ...\n",
      "  [153 145 216]\n",
      "  [148 140 211]\n",
      "  [154 145 218]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024381.jpg\n",
      "ISIC_0024381\n",
      "[[[105 100 121]\n",
      "  [107 100 121]\n",
      "  [104 102 122]\n",
      "  ...\n",
      "  [127 113 125]\n",
      "  [127 113 125]\n",
      "  [127 113 125]]\n",
      "\n",
      " [[106 100 119]\n",
      "  [108 101 122]\n",
      "  [105 100 121]\n",
      "  ...\n",
      "  [126 114 126]\n",
      "  [127 113 125]\n",
      "  [126 112 124]]\n",
      "\n",
      " [[105 100 121]\n",
      "  [106  98 121]\n",
      "  [103  98 119]\n",
      "  ...\n",
      "  [127 113 125]\n",
      "  [129 112 125]\n",
      "  [128 111 124]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[102 102 120]\n",
      "  [104 102 121]\n",
      "  [102 100 120]\n",
      "  ...\n",
      "  [116 102 120]\n",
      "  [118 102 119]\n",
      "  [115 102 118]]\n",
      "\n",
      " [[107 106 122]\n",
      "  [104 104 122]\n",
      "  [105 103 123]\n",
      "  ...\n",
      "  [114 101 117]\n",
      "  [113 101 119]\n",
      "  [113 101 119]]\n",
      "\n",
      " [[113 104 124]\n",
      "  [109 105 124]\n",
      "  [109 105 124]\n",
      "  ...\n",
      "  [117 104 120]\n",
      "  [112 100 118]\n",
      "  [113 101 119]]]\n",
      "[[[255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[255 255 255]\n",
      "  [255 255 255]\n",
      "  [255 255 255]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]\n",
      "\n",
      " [[  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  ...\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]\n",
      "  [  0   0   0]]]\n",
      "ISIC_0024401.jpg\n",
      "ISIC_0024401\n",
      "[[[172 149 234]\n",
      "  [168 146 234]\n",
      "  [166 147 234]\n",
      "  ...\n",
      "  [173 144 224]\n",
      "  [174 148 226]\n",
      "  [171 146 226]]\n",
      "\n",
      " [[172 147 237]\n",
      "  [171 149 238]\n",
      "  [168 149 236]\n",
      "  ...\n",
      "  [172 146 224]\n",
      "  [175 146 225]\n",
      "  [174 145 225]]\n",
      "\n",
      " [[172 146 236]\n",
      "  [172 147 239]\n",
      "  [172 147 237]\n",
      "  ...\n",
      "  [170 145 225]\n",
      "  [171 144 224]\n",
      "  [169 141 224]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[158 149 216]\n",
      "  [158 147 219]\n",
      "  [160 150 220]\n",
      "  ...\n",
      "  [142 133 206]\n",
      "  [143 135 206]\n",
      "  [148 139 212]]\n",
      "\n",
      " [[153 150 219]\n",
      "  [156 151 220]\n",
      "  [156 152 218]\n",
      "  ...\n",
      "  [145 134 204]\n",
      "  [147 136 206]\n",
      "  [150 139 211]]\n",
      "\n",
      " [[158 152 217]\n",
      "  [159 151 216]\n",
      "  [159 153 218]\n",
      "  ...\n",
      "  [143 134 201]\n",
      "  [145 136 203]\n",
      "  [145 139 204]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024395.jpg\n",
      "ISIC_0024395\n",
      "[[[193 161 242]\n",
      "  [193 163 239]\n",
      "  [191 163 239]\n",
      "  ...\n",
      "  [196 175 243]\n",
      "  [195 173 238]\n",
      "  [195 171 241]]\n",
      "\n",
      " [[187 163 235]\n",
      "  [192 165 238]\n",
      "  [193 163 238]\n",
      "  ...\n",
      "  [202 174 243]\n",
      "  [198 173 241]\n",
      "  [195 170 238]]\n",
      "\n",
      " [[192 165 239]\n",
      "  [190 165 239]\n",
      "  [189 166 240]\n",
      "  ...\n",
      "  [201 176 244]\n",
      "  [198 173 241]\n",
      "  [195 170 238]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[173 161 221]\n",
      "  [173 160 222]\n",
      "  [173 160 222]\n",
      "  ...\n",
      "  [172 155 222]\n",
      "  [173 156 220]\n",
      "  [174 157 221]]\n",
      "\n",
      " [[170 158 224]\n",
      "  [170 158 224]\n",
      "  [169 156 224]\n",
      "  ...\n",
      "  [171 155 219]\n",
      "  [174 156 223]\n",
      "  [179 159 224]]\n",
      "\n",
      " [[172 161 224]\n",
      "  [168 159 222]\n",
      "  [168 158 224]\n",
      "  ...\n",
      "  [171 155 219]\n",
      "  [172 155 222]\n",
      "  [171 157 221]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024366.jpg\n",
      "ISIC_0024366\n",
      "[[[230 211 228]\n",
      "  [229 208 230]\n",
      "  [230 208 232]\n",
      "  ...\n",
      "  [226 201 229]\n",
      "  [226 201 227]\n",
      "  [224 199 225]]\n",
      "\n",
      " [[236 208 231]\n",
      "  [234 208 231]\n",
      "  [233 207 230]\n",
      "  ...\n",
      "  [225 199 229]\n",
      "  [226 201 229]\n",
      "  [224 202 227]]\n",
      "\n",
      " [[237 208 231]\n",
      "  [235 206 231]\n",
      "  [235 207 230]\n",
      "  ...\n",
      "  [224 198 228]\n",
      "  [226 201 229]\n",
      "  [226 201 227]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[208 189 214]\n",
      "  [208 192 216]\n",
      "  [208 192 216]\n",
      "  ...\n",
      "  [203 188 216]\n",
      "  [200 188 218]\n",
      "  [200 185 217]]\n",
      "\n",
      " [[209 191 214]\n",
      "  [209 193 217]\n",
      "  [209 193 217]\n",
      "  ...\n",
      "  [215 192 220]\n",
      "  [206 192 220]\n",
      "  [201 186 217]]\n",
      "\n",
      " [[208 192 215]\n",
      "  [207 191 214]\n",
      "  [208 195 217]\n",
      "  ...\n",
      "  [218 188 217]\n",
      "  [206 191 219]\n",
      "  [203 188 219]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024328.jpg\n",
      "ISIC_0024328\n",
      "[[[139 180 219]\n",
      "  [138 176 211]\n",
      "  [142 180 212]\n",
      "  ...\n",
      "  [171 191 238]\n",
      "  [168 188 235]\n",
      "  [166 184 231]]\n",
      "\n",
      " [[144 156 180]\n",
      "  [138 153 179]\n",
      "  [145 169 193]\n",
      "  ...\n",
      "  [173 183 231]\n",
      "  [177 185 232]\n",
      "  [174 182 229]]\n",
      "\n",
      " [[173 180 197]\n",
      "  [165 177 201]\n",
      "  [157 180 206]\n",
      "  ...\n",
      "  [177 185 232]\n",
      "  [185 191 238]\n",
      "  [186 190 238]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[173 190 223]\n",
      "  [173 190 223]\n",
      "  [168 185 218]\n",
      "  ...\n",
      "  [168 181 219]\n",
      "  [163 181 218]\n",
      "  [167 186 223]]\n",
      "\n",
      " [[174 191 224]\n",
      "  [173 190 223]\n",
      "  [168 185 218]\n",
      "  ...\n",
      "  [171 183 219]\n",
      "  [166 182 219]\n",
      "  [161 179 216]]\n",
      "\n",
      " [[173 190 223]\n",
      "  [172 189 222]\n",
      "  [167 184 217]\n",
      "  ...\n",
      "  [175 184 221]\n",
      "  [170 186 223]\n",
      "  [168 186 223]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024390.jpg\n",
      "ISIC_0024390\n",
      "[[[139 142 233]\n",
      "  [141 145 233]\n",
      "  [138 144 233]\n",
      "  ...\n",
      "  [150 153 228]\n",
      "  [147 151 230]\n",
      "  [152 150 232]]\n",
      "\n",
      " [[138 141 232]\n",
      "  [141 142 233]\n",
      "  [137 139 227]\n",
      "  ...\n",
      "  [152 153 234]\n",
      "  [149 152 233]\n",
      "  [150 150 234]]\n",
      "\n",
      " [[141 141 235]\n",
      "  [137 142 235]\n",
      "  [132 137 228]\n",
      "  ...\n",
      "  [152 152 236]\n",
      "  [152 152 230]\n",
      "  [158 152 233]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 93 120 200]\n",
      "  [ 94 117 197]\n",
      "  [ 95 118 198]\n",
      "  ...\n",
      "  [127 140 208]\n",
      "  [125 141 208]\n",
      "  [129 141 207]]\n",
      "\n",
      " [[ 96 118 200]\n",
      "  [ 96 118 200]\n",
      "  [ 93 114 199]\n",
      "  ...\n",
      "  [128 138 208]\n",
      "  [124 136 208]\n",
      "  [128 137 210]]\n",
      "\n",
      " [[101 122 203]\n",
      "  [ 97 117 202]\n",
      "  [ 93 113 198]\n",
      "  ...\n",
      "  [130 138 208]\n",
      "  [127 136 209]\n",
      "  [126 135 208]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024362.jpg\n",
      "ISIC_0024362\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[139 136 221]\n",
      "  [143 137 224]\n",
      "  [138 135 221]\n",
      "  ...\n",
      "  [140 137 217]\n",
      "  [140 134 215]\n",
      "  [134 129 208]]\n",
      "\n",
      " [[142 133 223]\n",
      "  [141 135 222]\n",
      "  [137 134 220]\n",
      "  ...\n",
      "  [140 141 215]\n",
      "  [137 137 213]\n",
      "  [131 132 212]]\n",
      "\n",
      " [[139 133 220]\n",
      "  [135 134 220]\n",
      "  [133 132 218]\n",
      "  ...\n",
      "  [145 139 221]\n",
      "  [141 135 216]\n",
      "  [144 138 219]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[117 119 203]\n",
      "  [115 119 198]\n",
      "  [116 124 201]\n",
      "  ...\n",
      "  [125 130 205]\n",
      "  [124 130 201]\n",
      "  [124 130 201]]\n",
      "\n",
      " [[115 120 199]\n",
      "  [115 122 201]\n",
      "  [115 122 201]\n",
      "  ...\n",
      "  [131 131 209]\n",
      "  [130 129 203]\n",
      "  [130 133 207]]\n",
      "\n",
      " [[118 122 200]\n",
      "  [116 120 198]\n",
      "  [115 119 197]\n",
      "  ...\n",
      "  [125 129 207]\n",
      "  [127 129 207]\n",
      "  [126 130 208]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024335.jpg\n",
      "ISIC_0024335\n",
      "[[[141 141 217]\n",
      "  [139 145 216]\n",
      "  [136 142 215]\n",
      "  ...\n",
      "  [158 143 224]\n",
      "  [159 143 221]\n",
      "  [158 142 220]]\n",
      "\n",
      " [[146 142 218]\n",
      "  [139 142 217]\n",
      "  [138 143 218]\n",
      "  ...\n",
      "  [162 145 226]\n",
      "  [162 148 226]\n",
      "  [159 146 224]]\n",
      "\n",
      " [[143 139 214]\n",
      "  [142 139 218]\n",
      "  [142 139 218]\n",
      "  ...\n",
      "  [158 148 225]\n",
      "  [158 146 228]\n",
      "  [160 147 225]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[107 121 197]\n",
      "  [110 125 198]\n",
      "  [108 126 197]\n",
      "  ...\n",
      "  [132 143 203]\n",
      "  [131 142 202]\n",
      "  [136 142 201]]\n",
      "\n",
      " [[109 119 196]\n",
      "  [110 121 195]\n",
      "  [107 121 197]\n",
      "  ...\n",
      "  [129 142 198]\n",
      "  [127 141 194]\n",
      "  [128 143 199]]\n",
      "\n",
      " [[108 121 195]\n",
      "  [105 118 192]\n",
      "  [105 119 195]\n",
      "  ...\n",
      "  [130 141 203]\n",
      "  [128 139 199]\n",
      "  [128 142 201]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024369.jpg\n",
      "ISIC_0024369\n",
      "[[[135 141 212]\n",
      "  [141 143 214]\n",
      "  [152 150 220]\n",
      "  ...\n",
      "  [138 144 209]\n",
      "  [139 146 209]\n",
      "  [139 147 207]]\n",
      "\n",
      " [[141 143 213]\n",
      "  [143 142 214]\n",
      "  [147 146 220]\n",
      "  ...\n",
      "  [142 147 208]\n",
      "  [146 150 209]\n",
      "  [140 146 211]]\n",
      "\n",
      " [[147 146 218]\n",
      "  [138 144 211]\n",
      "  [133 136 210]\n",
      "  ...\n",
      "  [145 147 211]\n",
      "  [148 152 211]\n",
      "  [133 140 203]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[122 133 195]\n",
      "  [121 132 194]\n",
      "  [116 131 193]\n",
      "  ...\n",
      "  [149 152 203]\n",
      "  [148 152 205]\n",
      "  [145 151 204]]\n",
      "\n",
      " [[124 133 197]\n",
      "  [124 134 198]\n",
      "  [120 135 198]\n",
      "  ...\n",
      "  [144 149 204]\n",
      "  [141 148 205]\n",
      "  [143 148 203]]\n",
      "\n",
      " [[131 139 199]\n",
      "  [132 142 202]\n",
      "  [128 139 199]\n",
      "  ...\n",
      "  [146 152 205]\n",
      "  [143 150 205]\n",
      "  [134 148 200]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024315.jpg\n",
      "ISIC_0024315\n",
      "[[[138 119 146]\n",
      "  [141 122 149]\n",
      "  [145 125 148]\n",
      "  ...\n",
      "  [ 88  68  91]\n",
      "  [ 84  66  89]\n",
      "  [ 83  64  89]]\n",
      "\n",
      " [[142 121 146]\n",
      "  [141 122 147]\n",
      "  [147 127 150]\n",
      "  ...\n",
      "  [ 88  69  94]\n",
      "  [ 87  69  92]\n",
      "  [ 85  66  91]]\n",
      "\n",
      " [[143 123 146]\n",
      "  [144 127 148]\n",
      "  [148 127 152]\n",
      "  ...\n",
      "  [ 89  70  95]\n",
      "  [ 86  70  94]\n",
      "  [ 83  68  89]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 29  18  26]\n",
      "  [ 29  18  26]\n",
      "  [ 35  22  30]\n",
      "  ...\n",
      "  [ 26  18  29]\n",
      "  [ 25  19  30]\n",
      "  [ 26  18  29]]\n",
      "\n",
      " [[ 26  18  25]\n",
      "  [ 27  19  26]\n",
      "  [ 31  20  28]\n",
      "  ...\n",
      "  [ 22  15  28]\n",
      "  [ 24  18  29]\n",
      "  [ 23  17  28]]\n",
      "\n",
      " [[ 26  18  25]\n",
      "  [ 27  19  26]\n",
      "  [ 30  19  27]\n",
      "  ...\n",
      "  [ 22  15  28]\n",
      "  [ 24  18  29]\n",
      "  [ 24  18  29]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024388.jpg\n",
      "ISIC_0024388\n",
      "[[[149 138 230]\n",
      "  [155 142 234]\n",
      "  [150 140 230]\n",
      "  ...\n",
      "  [153 137 231]\n",
      "  [148 134 228]\n",
      "  [148 135 227]]\n",
      "\n",
      " [[148 138 228]\n",
      "  [154 144 234]\n",
      "  [152 142 232]\n",
      "  ...\n",
      "  [152 139 231]\n",
      "  [147 136 228]\n",
      "  [146 136 226]]\n",
      "\n",
      " [[149 139 229]\n",
      "  [155 143 233]\n",
      "  [155 143 233]\n",
      "  ...\n",
      "  [153 138 229]\n",
      "  [147 136 228]\n",
      "  [145 136 226]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[125 130 209]\n",
      "  [122 129 208]\n",
      "  [120 130 208]\n",
      "  ...\n",
      "  [122 120 209]\n",
      "  [128 122 211]\n",
      "  [127 118 204]]\n",
      "\n",
      " [[118 132 208]\n",
      "  [114 129 202]\n",
      "  [119 132 208]\n",
      "  ...\n",
      "  [125 119 206]\n",
      "  [128 123 208]\n",
      "  [130 122 205]]\n",
      "\n",
      " [[123 133 211]\n",
      "  [120 130 207]\n",
      "  [124 131 210]\n",
      "  ...\n",
      "  [126 119 204]\n",
      "  [126 119 204]\n",
      "  [129 122 207]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024383.jpg\n",
      "ISIC_0024383\n",
      "[[[186 161 221]\n",
      "  [193 165 225]\n",
      "  [191 167 225]\n",
      "  ...\n",
      "  [170 161 212]\n",
      "  [174 160 218]\n",
      "  [168 161 218]]\n",
      "\n",
      " [[193 165 224]\n",
      "  [192 164 223]\n",
      "  [196 166 225]\n",
      "  ...\n",
      "  [177 164 220]\n",
      "  [172 162 215]\n",
      "  [172 161 217]]\n",
      "\n",
      " [[198 166 225]\n",
      "  [191 163 222]\n",
      "  [189 161 221]\n",
      "  ...\n",
      "  [176 162 220]\n",
      "  [168 159 215]\n",
      "  [168 159 215]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[178 165 219]\n",
      "  [174 158 212]\n",
      "  [170 156 214]\n",
      "  ...\n",
      "  [179 169 211]\n",
      "  [181 172 215]\n",
      "  [178 172 213]]\n",
      "\n",
      " [[176 163 219]\n",
      "  [175 160 217]\n",
      "  [174 156 215]\n",
      "  ...\n",
      "  [177 170 213]\n",
      "  [182 170 218]\n",
      "  [180 171 214]]\n",
      "\n",
      " [[180 163 220]\n",
      "  [182 163 220]\n",
      "  [180 158 216]\n",
      "  ...\n",
      "  [173 170 209]\n",
      "  [181 173 214]\n",
      "  [177 170 215]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024342.jpg\n",
      "ISIC_0024342\n",
      "[[[160 152 217]\n",
      "  [163 155 218]\n",
      "  [162 155 216]\n",
      "  ...\n",
      "  [163 159 211]\n",
      "  [165 162 212]\n",
      "  [166 160 213]]\n",
      "\n",
      " [[158 150 213]\n",
      "  [162 153 216]\n",
      "  [162 155 216]\n",
      "  ...\n",
      "  [167 157 217]\n",
      "  [164 157 214]\n",
      "  [164 157 214]]\n",
      "\n",
      " [[163 155 218]\n",
      "  [163 155 218]\n",
      "  [166 154 220]\n",
      "  ...\n",
      "  [165 156 213]\n",
      "  [166 157 213]\n",
      "  [169 159 212]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[103  93 163]\n",
      "  [ 94  88 153]\n",
      "  [ 94  92 152]\n",
      "  ...\n",
      "  [142 142 190]\n",
      "  [144 141 190]\n",
      "  [147 143 192]]\n",
      "\n",
      " [[ 91  91 161]\n",
      "  [ 91  90 162]\n",
      "  [ 86  84 150]\n",
      "  ...\n",
      "  [141 136 187]\n",
      "  [147 141 188]\n",
      "  [153 147 188]]\n",
      "\n",
      " [[ 96  92 163]\n",
      "  [ 96  92 164]\n",
      "  [100  92 163]\n",
      "  ...\n",
      "  [143 135 188]\n",
      "  [145 138 189]\n",
      "  [150 144 193]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024319.jpg\n",
      "ISIC_0024319\n",
      "[[[158 175 202]\n",
      "  [160 177 204]\n",
      "  [154 174 199]\n",
      "  ...\n",
      "  [105 129 147]\n",
      "  [106 131 151]\n",
      "  [ 98 122 140]]\n",
      "\n",
      " [[151 171 202]\n",
      "  [153 173 204]\n",
      "  [152 174 202]\n",
      "  ...\n",
      "  [157 176 209]\n",
      "  [157 176 209]\n",
      "  [158 177 210]]\n",
      "\n",
      " [[149 171 206]\n",
      "  [147 169 204]\n",
      "  [148 173 207]\n",
      "  ...\n",
      "  [145 164 201]\n",
      "  [146 164 201]\n",
      "  [151 169 206]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[140 165 205]\n",
      "  [139 165 205]\n",
      "  [139 164 204]\n",
      "  ...\n",
      "  [153 178 218]\n",
      "  [149 175 212]\n",
      "  [146 173 210]]\n",
      "\n",
      " [[144 169 209]\n",
      "  [144 169 209]\n",
      "  [142 167 207]\n",
      "  ...\n",
      "  [153 179 216]\n",
      "  [149 175 212]\n",
      "  [149 172 210]]\n",
      "\n",
      " [[151 171 212]\n",
      "  [151 171 212]\n",
      "  [148 170 211]\n",
      "  ...\n",
      "  [153 179 216]\n",
      "  [151 177 214]\n",
      "  [152 175 213]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024363.jpg\n",
      "ISIC_0024363\n",
      "[[[ 78  62  79]\n",
      "  [ 80  62  79]\n",
      "  [ 78  62  79]\n",
      "  ...\n",
      "  [ 98  83  87]\n",
      "  [ 96  81  85]\n",
      "  [ 95  80  84]]\n",
      "\n",
      " [[ 80  64  82]\n",
      "  [ 82  64  81]\n",
      "  [ 80  64  81]\n",
      "  ...\n",
      "  [ 99  84  88]\n",
      "  [ 98  83  87]\n",
      "  [ 96  81  85]]\n",
      "\n",
      " [[ 81  65  82]\n",
      "  [ 83  65  82]\n",
      "  [ 82  66  83]\n",
      "  ...\n",
      "  [100  85  89]\n",
      "  [ 97  82  86]\n",
      "  [ 97  82  86]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 87  71  78]\n",
      "  [ 89  73  80]\n",
      "  [ 89  73  80]\n",
      "  ...\n",
      "  [ 81  66  70]\n",
      "  [ 78  64  68]\n",
      "  [ 78  64  68]]\n",
      "\n",
      " [[ 87  71  78]\n",
      "  [ 87  71  78]\n",
      "  [ 88  72  79]\n",
      "  ...\n",
      "  [ 81  66  70]\n",
      "  [ 79  64  68]\n",
      "  [ 79  64  68]]\n",
      "\n",
      " [[ 86  70  77]\n",
      "  [ 88  72  79]\n",
      "  [ 87  71  78]\n",
      "  ...\n",
      "  [ 79  64  68]\n",
      "  [ 79  64  68]\n",
      "  [ 78  63  67]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024338.jpg\n",
      "ISIC_0024338\n",
      "[[[103  95 126]\n",
      "  [106  98 129]\n",
      "  [107  99 130]\n",
      "  ...\n",
      "  [ 98  96 126]\n",
      "  [101  95 126]\n",
      "  [102  94 125]]\n",
      "\n",
      " [[103  95 126]\n",
      "  [106  98 129]\n",
      "  [107  99 130]\n",
      "  ...\n",
      "  [ 99  97 127]\n",
      "  [102  96 127]\n",
      "  [103  95 126]]\n",
      "\n",
      " [[103  95 126]\n",
      "  [105  97 128]\n",
      "  [107  99 130]\n",
      "  ...\n",
      "  [100  98 128]\n",
      "  [102  96 127]\n",
      "  [103  95 126]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[ 91  89 119]\n",
      "  [ 91  89 119]\n",
      "  [ 92  90 120]\n",
      "  ...\n",
      "  [ 83  71 101]\n",
      "  [ 81  72  99]\n",
      "  [ 81  69  97]]\n",
      "\n",
      " [[ 89  89 119]\n",
      "  [ 92  90 120]\n",
      "  [ 90  90 120]\n",
      "  ...\n",
      "  [ 83  71  99]\n",
      "  [ 79  70  97]\n",
      "  [ 77  68  95]]\n",
      "\n",
      " [[ 91  89 119]\n",
      "  [ 93  88 119]\n",
      "  [ 91  89 119]\n",
      "  ...\n",
      "  [ 80  68  96]\n",
      "  [ 76  67  94]\n",
      "  [ 74  65  92]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024379.jpg\n",
      "ISIC_0024379\n",
      "[[[250 215 241]\n",
      "  [252 217 243]\n",
      "  [251 219 244]\n",
      "  ...\n",
      "  [248 198 216]\n",
      "  [247 197 215]\n",
      "  [251 200 214]]\n",
      "\n",
      " [[250 214 238]\n",
      "  [249 215 245]\n",
      "  [249 218 245]\n",
      "  ...\n",
      "  [251 202 216]\n",
      "  [240 197 210]\n",
      "  [242 200 211]]\n",
      "\n",
      " [[249 214 240]\n",
      "  [250 215 242]\n",
      "  [253 216 242]\n",
      "  ...\n",
      "  [249 198 212]\n",
      "  [243 196 212]\n",
      "  [242 197 213]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[234 201 208]\n",
      "  [230 194 206]\n",
      "  [231 197 208]\n",
      "  ...\n",
      "  [230 199 206]\n",
      "  [228 195 202]\n",
      "  [230 196 203]]\n",
      "\n",
      " [[228 197 206]\n",
      "  [232 201 210]\n",
      "  [226 199 208]\n",
      "  ...\n",
      "  [237 195 206]\n",
      "  [237 193 210]\n",
      "  [234 195 210]]\n",
      "\n",
      " [[225 201 209]\n",
      "  [227 203 211]\n",
      "  [228 200 213]\n",
      "  ...\n",
      "  [221 195 208]\n",
      "  [224 197 207]\n",
      "  [229 193 209]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024357.jpg\n",
      "ISIC_0024357\n",
      "[[[167 165 231]\n",
      "  [173 163 233]\n",
      "  [165 158 225]\n",
      "  ...\n",
      "  [155 154 220]\n",
      "  [158 154 225]\n",
      "  [158 152 223]]\n",
      "\n",
      " [[176 168 233]\n",
      "  [173 168 237]\n",
      "  [162 161 229]\n",
      "  ...\n",
      "  [156 157 225]\n",
      "  [160 159 227]\n",
      "  [154 154 224]]\n",
      "\n",
      " [[169 168 232]\n",
      "  [166 168 232]\n",
      "  [164 163 229]\n",
      "  ...\n",
      "  [155 164 228]\n",
      "  [154 160 225]\n",
      "  [152 157 220]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[147 148 208]\n",
      "  [146 144 203]\n",
      "  [148 147 209]\n",
      "  ...\n",
      "  [136 136 196]\n",
      "  [140 135 197]\n",
      "  [133 133 193]]\n",
      "\n",
      " [[138 145 208]\n",
      "  [140 143 204]\n",
      "  [146 145 207]\n",
      "  ...\n",
      "  [141 136 198]\n",
      "  [134 130 196]\n",
      "  [136 134 194]]\n",
      "\n",
      " [[135 141 206]\n",
      "  [138 140 205]\n",
      "  [145 143 209]\n",
      "  ...\n",
      "  [147 139 202]\n",
      "  [135 136 197]\n",
      "  [137 135 195]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024309.jpg\n",
      "ISIC_0024309\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[131 119 207]\n",
      "  [140 123 214]\n",
      "  [135 119 207]\n",
      "  ...\n",
      "  [144 126 209]\n",
      "  [144 125 210]\n",
      "  [145 131 209]]\n",
      "\n",
      " [[132 116 204]\n",
      "  [135 122 208]\n",
      "  [132 119 205]\n",
      "  ...\n",
      "  [147 128 213]\n",
      "  [147 128 213]\n",
      "  [147 131 215]]\n",
      "\n",
      " [[128 110 201]\n",
      "  [131 114 205]\n",
      "  [133 117 205]\n",
      "  ...\n",
      "  [141 128 206]\n",
      "  [144 132 208]\n",
      "  [146 132 210]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[122 120 190]\n",
      "  [123 120 192]\n",
      "  [119 118 190]\n",
      "  ...\n",
      "  [132 123 190]\n",
      "  [132 121 189]\n",
      "  [132 124 189]]\n",
      "\n",
      " [[117 116 188]\n",
      "  [121 120 192]\n",
      "  [123 118 193]\n",
      "  ...\n",
      "  [129 121 191]\n",
      "  [130 121 188]\n",
      "  [135 126 193]]\n",
      "\n",
      " [[111 110 190]\n",
      "  [117 117 193]\n",
      "  [119 113 190]\n",
      "  ...\n",
      "  [127 125 191]\n",
      "  [131 127 192]\n",
      "  [130 126 191]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024380.jpg\n",
      "ISIC_0024380\n",
      "[[[130 100 143]\n",
      "  [132 102 145]\n",
      "  [132 102 145]\n",
      "  ...\n",
      "  [154 130 160]\n",
      "  [155 128 161]\n",
      "  [153 128 162]]\n",
      "\n",
      " [[130 100 143]\n",
      "  [131 101 144]\n",
      "  [132 102 145]\n",
      "  ...\n",
      "  [154 130 160]\n",
      "  [153 129 161]\n",
      "  [153 128 162]]\n",
      "\n",
      " [[131 101 144]\n",
      "  [133 103 146]\n",
      "  [133 103 146]\n",
      "  ...\n",
      "  [153 129 159]\n",
      "  [156 129 162]\n",
      "  [154 129 163]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[127 101 141]\n",
      "  [127 101 141]\n",
      "  [127 101 141]\n",
      "  ...\n",
      "  [156 131 165]\n",
      "  [154 127 166]\n",
      "  [151 124 163]]\n",
      "\n",
      " [[126  99 139]\n",
      "  [125  98 138]\n",
      "  [127 101 141]\n",
      "  ...\n",
      "  [154 129 163]\n",
      "  [153 127 163]\n",
      "  [152 125 164]]\n",
      "\n",
      " [[125  98 138]\n",
      "  [124  97 137]\n",
      "  [126 100 140]\n",
      "  ...\n",
      "  [154 129 163]\n",
      "  [152 126 162]\n",
      "  [150 123 162]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024336.jpg\n",
      "ISIC_0024336\n",
      "[[[141 130 150]\n",
      "  [142 131 151]\n",
      "  [142 131 151]\n",
      "  ...\n",
      "  [134 119 140]\n",
      "  [137 121 139]\n",
      "  [138 122 139]]\n",
      "\n",
      " [[142 131 151]\n",
      "  [143 132 152]\n",
      "  [142 131 151]\n",
      "  ...\n",
      "  [135 121 139]\n",
      "  [137 121 139]\n",
      "  [136 119 140]]\n",
      "\n",
      " [[143 132 152]\n",
      "  [141 130 150]\n",
      "  [142 131 151]\n",
      "  ...\n",
      "  [137 121 139]\n",
      "  [137 121 139]\n",
      "  [137 120 141]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[131 119 141]\n",
      "  [129 119 142]\n",
      "  [132 119 141]\n",
      "  ...\n",
      "  [117  96 121]\n",
      "  [114  95 120]\n",
      "  [112  93 118]]\n",
      "\n",
      " [[131 118 140]\n",
      "  [129 120 141]\n",
      "  [132 120 142]\n",
      "  ...\n",
      "  [115  96 121]\n",
      "  [114  95 120]\n",
      "  [113  94 119]]\n",
      "\n",
      " [[130 117 139]\n",
      "  [129 120 141]\n",
      "  [132 120 142]\n",
      "  ...\n",
      "  [114  95 120]\n",
      "  [114  95 120]\n",
      "  [114  95 120]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024316.jpg\n",
      "ISIC_0024316\n",
      "[[[140 141 227]\n",
      "  [139 137 226]\n",
      "  [137 138 226]\n",
      "  ...\n",
      "  [146 153 233]\n",
      "  [145 151 234]\n",
      "  [148 152 233]]\n",
      "\n",
      " [[141 142 232]\n",
      "  [139 137 227]\n",
      "  [140 141 229]\n",
      "  ...\n",
      "  [146 150 233]\n",
      "  [145 152 231]\n",
      "  [145 153 230]]\n",
      "\n",
      " [[135 139 228]\n",
      "  [134 139 230]\n",
      "  [140 141 231]\n",
      "  ...\n",
      "  [142 151 231]\n",
      "  [146 153 233]\n",
      "  [145 152 232]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[113 127 209]\n",
      "  [113 128 207]\n",
      "  [110 130 208]\n",
      "  ...\n",
      "  [135 137 208]\n",
      "  [125 131 202]\n",
      "  [131 137 208]]\n",
      "\n",
      " [[111 121 198]\n",
      "  [113 123 200]\n",
      "  [118 130 212]\n",
      "  ...\n",
      "  [133 139 210]\n",
      "  [128 135 208]\n",
      "  [125 135 205]]\n",
      "\n",
      " [[112 129 202]\n",
      "  [114 131 204]\n",
      "  [114 133 208]\n",
      "  ...\n",
      "  [133 140 209]\n",
      "  [128 134 205]\n",
      "  [123 134 202]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024377.jpg\n",
      "ISIC_0024377\n",
      "[[[127 134 161]\n",
      "  [130 137 162]\n",
      "  [134 141 166]\n",
      "  ...\n",
      "  [158 159 179]\n",
      "  [158 159 179]\n",
      "  [157 158 178]]\n",
      "\n",
      " [[130 137 162]\n",
      "  [132 139 164]\n",
      "  [134 141 166]\n",
      "  ...\n",
      "  [158 159 179]\n",
      "  [158 159 179]\n",
      "  [157 158 178]]\n",
      "\n",
      " [[133 140 165]\n",
      "  [135 142 167]\n",
      "  [136 143 168]\n",
      "  ...\n",
      "  [158 159 179]\n",
      "  [158 159 179]\n",
      "  [156 157 177]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[145 149 173]\n",
      "  [145 150 171]\n",
      "  [149 151 175]\n",
      "  ...\n",
      "  [169 169 183]\n",
      "  [169 169 183]\n",
      "  [169 170 184]]\n",
      "\n",
      " [[145 150 171]\n",
      "  [146 151 172]\n",
      "  [149 151 175]\n",
      "  ...\n",
      "  [169 169 183]\n",
      "  [168 168 182]\n",
      "  [167 167 181]]\n",
      "\n",
      " [[145 150 171]\n",
      "  [146 151 172]\n",
      "  [149 151 175]\n",
      "  ...\n",
      "  [169 169 183]\n",
      "  [168 168 182]\n",
      "  [167 167 181]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024378.jpg\n",
      "ISIC_0024378\n",
      "[[[113  99 127]\n",
      "  [111  96 128]\n",
      "  [112  97 129]\n",
      "  ...\n",
      "  [125 115 138]\n",
      "  [124 114 137]\n",
      "  [124 117 138]]\n",
      "\n",
      " [[113  97 128]\n",
      "  [112  95 128]\n",
      "  [114  99 130]\n",
      "  ...\n",
      "  [125 115 138]\n",
      "  [125 113 141]\n",
      "  [125 116 143]]\n",
      "\n",
      " [[113  97 128]\n",
      "  [114  97 131]\n",
      "  [115  98 131]\n",
      "  ...\n",
      "  [127 117 140]\n",
      "  [124 114 137]\n",
      "  [122 115 136]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[118 108 125]\n",
      "  [120 108 126]\n",
      "  [122 111 127]\n",
      "  ...\n",
      "  [ 80  74  99]\n",
      "  [ 80  74  99]\n",
      "  [ 81  75 100]]\n",
      "\n",
      " [[118 107 123]\n",
      "  [119 108 124]\n",
      "  [119 108 124]\n",
      "  ...\n",
      "  [ 78  74  99]\n",
      "  [ 76  72  97]\n",
      "  [ 76  72  97]]\n",
      "\n",
      " [[116 104 122]\n",
      "  [116 104 122]\n",
      "  [118 106 124]\n",
      "  ...\n",
      "  [ 78  74  99]\n",
      "  [ 76  72  97]\n",
      "  [ 76  72  97]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024355.jpg\n",
      "ISIC_0024355\n",
      "[[[195 171 213]\n",
      "  [196 176 218]\n",
      "  [192 175 218]\n",
      "  ...\n",
      "  [178 165 219]\n",
      "  [175 161 215]\n",
      "  [177 164 216]]\n",
      "\n",
      " [[194 170 212]\n",
      "  [192 170 212]\n",
      "  [189 169 212]\n",
      "  ...\n",
      "  [175 165 218]\n",
      "  [175 162 216]\n",
      "  [181 165 219]]\n",
      "\n",
      " [[200 173 217]\n",
      "  [197 170 213]\n",
      "  [191 166 210]\n",
      "  ...\n",
      "  [171 161 214]\n",
      "  [172 162 215]\n",
      "  [177 164 218]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[152 151 195]\n",
      "  [151 148 193]\n",
      "  [152 149 194]\n",
      "  ...\n",
      "  [175 164 214]\n",
      "  [174 162 210]\n",
      "  [179 166 212]]\n",
      "\n",
      " [[152 153 197]\n",
      "  [151 152 196]\n",
      "  [155 152 197]\n",
      "  ...\n",
      "  [174 162 212]\n",
      "  [176 161 212]\n",
      "  [181 162 211]]\n",
      "\n",
      " [[151 157 200]\n",
      "  [150 156 199]\n",
      "  [152 153 197]\n",
      "  ...\n",
      "  [173 158 209]\n",
      "  [177 158 207]\n",
      "  [181 159 208]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024367.jpg\n",
      "ISIC_0024367\n",
      "[[[38 40 51]\n",
      "  [41 40 56]\n",
      "  [45 42 58]\n",
      "  ...\n",
      "  [32 24 41]\n",
      "  [30 23 36]\n",
      "  [28 21 34]]\n",
      "\n",
      " [[38 38 54]\n",
      "  [43 41 60]\n",
      "  [46 41 62]\n",
      "  ...\n",
      "  [35 27 44]\n",
      "  [32 25 38]\n",
      "  [31 22 35]]\n",
      "\n",
      " [[39 38 58]\n",
      "  [45 39 62]\n",
      "  [47 39 62]\n",
      "  ...\n",
      "  [35 28 43]\n",
      "  [33 25 36]\n",
      "  [33 24 37]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[24 24 42]\n",
      "  [24 25 45]\n",
      "  [26 23 45]\n",
      "  ...\n",
      "  [ 4  3 12]\n",
      "  [ 1  3 11]\n",
      "  [ 1  4  9]]\n",
      "\n",
      " [[26 25 45]\n",
      "  [26 24 46]\n",
      "  [28 27 43]\n",
      "  ...\n",
      "  [ 0  1 11]\n",
      "  [ 2  3 13]\n",
      "  [ 1  3 11]]\n",
      "\n",
      " [[27 25 45]\n",
      "  [27 25 44]\n",
      "  [31 26 41]\n",
      "  ...\n",
      "  [ 0  1 11]\n",
      "  [ 2  3 13]\n",
      "  [ 1  3 11]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024326.jpg\n",
      "ISIC_0024326\n",
      "[[[117 125 215]\n",
      "  [116 124 213]\n",
      "  [124 125 216]\n",
      "  ...\n",
      "  [117 118 216]\n",
      "  [118 119 217]\n",
      "  [121 120 218]]\n",
      "\n",
      " [[123 126 217]\n",
      "  [124 123 215]\n",
      "  [126 125 217]\n",
      "  ...\n",
      "  [113 116 214]\n",
      "  [113 119 216]\n",
      "  [114 120 217]]\n",
      "\n",
      " [[127 126 222]\n",
      "  [126 124 220]\n",
      "  [126 125 219]\n",
      "  ...\n",
      "  [111 115 210]\n",
      "  [112 118 213]\n",
      "  [112 120 214]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[104 110 199]\n",
      "  [106 110 198]\n",
      "  [108 110 198]\n",
      "  ...\n",
      "  [103 114 194]\n",
      "  [103 114 194]\n",
      "  [102 111 191]]\n",
      "\n",
      " [[106 110 199]\n",
      "  [103 109 198]\n",
      "  [102 110 199]\n",
      "  ...\n",
      "  [103 113 191]\n",
      "  [102 114 192]\n",
      "  [ 99 112 190]]\n",
      "\n",
      " [[105 111 200]\n",
      "  [101 109 198]\n",
      "  [101 111 199]\n",
      "  ...\n",
      "  [108 117 197]\n",
      "  [107 114 194]\n",
      "  [101 110 190]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024332.jpg\n",
      "ISIC_0024332\n",
      "[[[127 132 177]\n",
      "  [129 132 177]\n",
      "  [127 131 179]\n",
      "  ...\n",
      "  [146 141 186]\n",
      "  [145 142 187]\n",
      "  [146 141 186]]\n",
      "\n",
      " [[128 132 180]\n",
      "  [129 133 181]\n",
      "  [128 132 180]\n",
      "  ...\n",
      "  [144 141 186]\n",
      "  [145 142 187]\n",
      "  [143 143 183]]\n",
      "\n",
      " [[130 135 180]\n",
      "  [130 135 180]\n",
      "  [131 136 181]\n",
      "  ...\n",
      "  [145 141 190]\n",
      "  [147 141 188]\n",
      "  [146 141 186]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[137 137 177]\n",
      "  [138 137 177]\n",
      "  [139 138 178]\n",
      "  ...\n",
      "  [155 148 191]\n",
      "  [155 148 191]\n",
      "  [154 147 190]]\n",
      "\n",
      " [[138 137 176]\n",
      "  [139 138 177]\n",
      "  [142 141 180]\n",
      "  ...\n",
      "  [154 147 190]\n",
      "  [153 146 189]\n",
      "  [155 148 191]]\n",
      "\n",
      " [[138 137 177]\n",
      "  [138 136 178]\n",
      "  [141 140 180]\n",
      "  ...\n",
      "  [152 145 188]\n",
      "  [153 146 189]\n",
      "  [156 147 190]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024308.jpg\n",
      "ISIC_0024308\n",
      "[[[178 169 242]\n",
      "  [182 173 246]\n",
      "  [182 173 246]\n",
      "  ...\n",
      "  [181 169 245]\n",
      "  [183 171 247]\n",
      "  [183 170 248]]\n",
      "\n",
      " [[179 172 247]\n",
      "  [183 173 249]\n",
      "  [183 172 245]\n",
      "  ...\n",
      "  [181 169 245]\n",
      "  [178 167 247]\n",
      "  [183 170 248]]\n",
      "\n",
      " [[179 172 247]\n",
      "  [181 172 245]\n",
      "  [181 172 245]\n",
      "  ...\n",
      "  [181 168 242]\n",
      "  [180 164 247]\n",
      "  [184 170 248]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[160 170 234]\n",
      "  [162 169 232]\n",
      "  [167 172 235]\n",
      "  ...\n",
      "  [156 144 216]\n",
      "  [155 140 214]\n",
      "  [153 138 212]]\n",
      "\n",
      " [[162 172 236]\n",
      "  [166 172 237]\n",
      "  [168 169 237]\n",
      "  ...\n",
      "  [157 146 216]\n",
      "  [157 143 215]\n",
      "  [156 142 213]]\n",
      "\n",
      " [[165 174 237]\n",
      "  [169 174 237]\n",
      "  [169 171 236]\n",
      "  ...\n",
      "  [154 146 216]\n",
      "  [155 144 216]\n",
      "  [156 145 217]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024307.jpg\n",
      "ISIC_0024307\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[135 123 213]\n",
      "  [133 122 215]\n",
      "  [134 123 219]\n",
      "  ...\n",
      "  [156 142 224]\n",
      "  [159 147 229]\n",
      "  [156 144 226]]\n",
      "\n",
      " [[133 123 213]\n",
      "  [134 125 216]\n",
      "  [133 124 215]\n",
      "  ...\n",
      "  [156 144 226]\n",
      "  [157 145 227]\n",
      "  [157 144 228]]\n",
      "\n",
      " [[133 121 211]\n",
      "  [133 122 215]\n",
      "  [132 121 214]\n",
      "  ...\n",
      "  [158 143 227]\n",
      "  [157 142 226]\n",
      "  [157 142 226]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[117 121 199]\n",
      "  [118 127 200]\n",
      "  [120 125 200]\n",
      "  ...\n",
      "  [141 135 206]\n",
      "  [142 135 208]\n",
      "  [143 137 208]]\n",
      "\n",
      " [[120 121 202]\n",
      "  [120 121 201]\n",
      "  [122 123 203]\n",
      "  ...\n",
      "  [140 134 205]\n",
      "  [140 138 208]\n",
      "  [144 143 209]]\n",
      "\n",
      " [[116 122 199]\n",
      "  [117 122 197]\n",
      "  [121 125 203]\n",
      "  ...\n",
      "  [143 132 204]\n",
      "  [144 137 210]\n",
      "  [143 139 211]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024318.jpg\n",
      "ISIC_0024318\n",
      "[[[164 177 233]\n",
      "  [164 177 233]\n",
      "  [167 176 226]\n",
      "  ...\n",
      "  [159 173 231]\n",
      "  [159 176 233]\n",
      "  [156 175 232]]\n",
      "\n",
      " [[165 177 235]\n",
      "  [165 178 230]\n",
      "  [167 176 226]\n",
      "  ...\n",
      "  [164 176 234]\n",
      "  [161 175 233]\n",
      "  [161 175 233]]\n",
      "\n",
      " [[171 177 230]\n",
      "  [172 177 226]\n",
      "  [172 173 229]\n",
      "  ...\n",
      "  [165 178 234]\n",
      "  [164 177 233]\n",
      "  [164 177 233]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[132 153 205]\n",
      "  [132 152 207]\n",
      "  [137 151 204]\n",
      "  ...\n",
      "  [149 168 225]\n",
      "  [154 169 225]\n",
      "  [155 170 226]]\n",
      "\n",
      " [[129 153 207]\n",
      "  [132 150 203]\n",
      "  [133 151 204]\n",
      "  ...\n",
      "  [148 166 227]\n",
      "  [154 169 225]\n",
      "  [153 170 226]]\n",
      "\n",
      " [[131 151 208]\n",
      "  [130 150 205]\n",
      "  [129 150 205]\n",
      "  ...\n",
      "  [152 165 227]\n",
      "  [153 170 227]\n",
      "  [150 170 225]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024320.jpg\n",
      "ISIC_0024320\n",
      "[[[114 106 201]\n",
      "  [116 105 201]\n",
      "  [117 107 200]\n",
      "  ...\n",
      "  [136 130 211]\n",
      "  [137 131 213]\n",
      "  [143 132 218]]\n",
      "\n",
      " [[101  89 189]\n",
      "  [103  89 190]\n",
      "  [104  92 188]\n",
      "  ...\n",
      "  [132 131 211]\n",
      "  [132 131 211]\n",
      "  [132 130 212]]\n",
      "\n",
      " [[ 86  75 171]\n",
      "  [ 87  74 172]\n",
      "  [ 85  74 170]\n",
      "  ...\n",
      "  [132 129 209]\n",
      "  [129 128 208]\n",
      "  [129 128 208]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[101 104 185]\n",
      "  [103 108 187]\n",
      "  [103 111 188]\n",
      "  ...\n",
      "  [103 118 187]\n",
      "  [100 115 184]\n",
      "  [105 120 189]]\n",
      "\n",
      " [[ 97 104 184]\n",
      "  [100 108 185]\n",
      "  [103 112 186]\n",
      "  ...\n",
      "  [110 119 187]\n",
      "  [104 118 184]\n",
      "  [111 123 189]]\n",
      "\n",
      " [[ 94 104 181]\n",
      "  [101 110 184]\n",
      "  [101 110 183]\n",
      "  ...\n",
      "  [108 119 187]\n",
      "  [105 121 187]\n",
      "  [108 122 188]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024372.jpg\n",
      "ISIC_0024372\n",
      "[[[100  89 145]\n",
      "  [105  92 148]\n",
      "  [106  93 147]\n",
      "  ...\n",
      "  [159 142 185]\n",
      "  [160 143 186]\n",
      "  [160 141 184]]\n",
      "\n",
      " [[ 95  86 143]\n",
      "  [103  90 146]\n",
      "  [107  94 146]\n",
      "  ...\n",
      "  [161 141 184]\n",
      "  [160 142 183]\n",
      "  [159 139 182]]\n",
      "\n",
      " [[ 92  83 140]\n",
      "  [ 98  86 144]\n",
      "  [106  91 148]\n",
      "  ...\n",
      "  [159 139 184]\n",
      "  [157 140 183]\n",
      "  [157 140 183]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[139 122 166]\n",
      "  [141 121 166]\n",
      "  [142 122 167]\n",
      "  ...\n",
      "  [157 149 190]\n",
      "  [160 150 193]\n",
      "  [160 150 192]]\n",
      "\n",
      " [[139 122 166]\n",
      "  [142 122 167]\n",
      "  [145 124 169]\n",
      "  ...\n",
      "  [158 150 187]\n",
      "  [161 149 191]\n",
      "  [161 150 190]]\n",
      "\n",
      " [[143 123 168]\n",
      "  [144 124 169]\n",
      "  [146 126 171]\n",
      "  ...\n",
      "  [158 147 187]\n",
      "  [155 146 186]\n",
      "  [156 147 187]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024375.jpg\n",
      "ISIC_0024375\n",
      "[[[159 153 212]\n",
      "  [160 150 210]\n",
      "  [158 149 206]\n",
      "  ...\n",
      "  [152 146 203]\n",
      "  [150 147 203]\n",
      "  [150 143 204]]\n",
      "\n",
      " [[157 153 212]\n",
      "  [153 150 206]\n",
      "  [156 150 209]\n",
      "  ...\n",
      "  [157 145 205]\n",
      "  [157 148 205]\n",
      "  [154 145 201]]\n",
      "\n",
      " [[156 148 213]\n",
      "  [160 149 212]\n",
      "  [166 151 213]\n",
      "  ...\n",
      "  [168 155 211]\n",
      "  [163 152 208]\n",
      "  [154 148 203]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[143 147 200]\n",
      "  [144 146 204]\n",
      "  [141 145 203]\n",
      "  ...\n",
      "  [150 144 197]\n",
      "  [149 147 199]\n",
      "  [140 142 196]]\n",
      "\n",
      " [[136 143 198]\n",
      "  [147 149 207]\n",
      "  [144 148 199]\n",
      "  ...\n",
      "  [152 148 200]\n",
      "  [151 147 199]\n",
      "  [139 143 194]]\n",
      "\n",
      " [[136 145 195]\n",
      "  [145 149 200]\n",
      "  [145 147 201]\n",
      "  ...\n",
      "  [144 146 200]\n",
      "  [144 142 194]\n",
      "  [137 141 192]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024376.jpg\n",
      "ISIC_0024376\n",
      "[[[150 128 217]\n",
      "  [158 134 222]\n",
      "  [161 136 222]\n",
      "  ...\n",
      "  [133 121 211]\n",
      "  [128 116 206]\n",
      "  [123 114 205]]\n",
      "\n",
      " [[142 126 214]\n",
      "  [151 129 218]\n",
      "  [156 138 221]\n",
      "  ...\n",
      "  [138 123 215]\n",
      "  [128 116 212]\n",
      "  [118 112 207]]\n",
      "\n",
      " [[144 131 217]\n",
      "  [149 130 217]\n",
      "  [148 132 215]\n",
      "  ...\n",
      "  [139 121 215]\n",
      "  [131 118 210]\n",
      "  [121 109 205]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[123 122 202]\n",
      "  [117 117 201]\n",
      "  [115 115 199]\n",
      "  ...\n",
      "  [114 109 194]\n",
      "  [105 105 189]\n",
      "  [102 106 189]]\n",
      "\n",
      " [[130 124 206]\n",
      "  [115 120 199]\n",
      "  [111 118 197]\n",
      "  ...\n",
      "  [111 107 196]\n",
      "  [104 104 188]\n",
      "  [104 107 188]]\n",
      "\n",
      " [[126 122 204]\n",
      "  [123 117 198]\n",
      "  [123 117 198]\n",
      "  ...\n",
      "  [114 106 193]\n",
      "  [109 107 189]\n",
      "  [109 108 188]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024356.jpg\n",
      "ISIC_0024356\n",
      "[[[ 88  85 101]\n",
      "  [113 114 135]\n",
      "  [106 109 137]\n",
      "  ...\n",
      "  [  5   4   6]\n",
      "  [  0   1   2]\n",
      "  [  8   8  14]]\n",
      "\n",
      " [[113 113 137]\n",
      "  [149 151 185]\n",
      "  [138 146 183]\n",
      "  ...\n",
      "  [ 11  10  20]\n",
      "  [ 42  48  71]\n",
      "  [ 92  99 126]]\n",
      "\n",
      " [[106 107 135]\n",
      "  [137 142 181]\n",
      "  [130 138 178]\n",
      "  ...\n",
      "  [101 105 133]\n",
      "  [118 134 171]\n",
      "  [121 139 176]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[124 127 141]\n",
      "  [162 168 191]\n",
      "  [151 159 188]\n",
      "  ...\n",
      "  [158 165 184]\n",
      "  [158 165 184]\n",
      "  [158 165 184]]\n",
      "\n",
      " [[121 124 139]\n",
      "  [160 165 190]\n",
      "  [150 158 187]\n",
      "  ...\n",
      "  [158 165 184]\n",
      "  [158 165 184]\n",
      "  [158 165 184]]\n",
      "\n",
      " [[123 125 143]\n",
      "  [162 166 194]\n",
      "  [151 159 189]\n",
      "  ...\n",
      "  [159 166 185]\n",
      "  [158 165 184]\n",
      "  [158 165 184]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024374.jpg\n",
      "ISIC_0024374\n",
      "[[[178 172 231]\n",
      "  [184 177 234]\n",
      "  [171 170 226]\n",
      "  ...\n",
      "  [156 160 211]\n",
      "  [164 165 216]\n",
      "  [168 169 220]]\n",
      "\n",
      " [[172 175 226]\n",
      "  [173 172 228]\n",
      "  [161 168 223]\n",
      "  ...\n",
      "  [154 162 215]\n",
      "  [156 161 216]\n",
      "  [159 163 221]]\n",
      "\n",
      " [[166 170 228]\n",
      "  [171 169 229]\n",
      "  [167 164 219]\n",
      "  ...\n",
      "  [158 160 214]\n",
      "  [152 158 211]\n",
      "  [170 171 222]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[139 152 208]\n",
      "  [145 156 208]\n",
      "  [154 160 213]\n",
      "  ...\n",
      "  [160 167 210]\n",
      "  [163 169 214]\n",
      "  [160 167 210]]\n",
      "\n",
      " [[142 153 205]\n",
      "  [145 160 209]\n",
      "  [143 156 202]\n",
      "  ...\n",
      "  [159 168 211]\n",
      "  [160 169 212]\n",
      "  [158 169 207]]\n",
      "\n",
      " [[142 153 203]\n",
      "  [144 156 204]\n",
      "  [151 159 206]\n",
      "  ...\n",
      "  [158 170 210]\n",
      "  [157 169 209]\n",
      "  [152 171 206]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n",
      "ISIC_0024347.jpg\n",
      "ISIC_0024347\n",
      "[[[179 168 231]\n",
      "  [178 166 232]\n",
      "  [177 165 231]\n",
      "  ...\n",
      "  [175 165 231]\n",
      "  [174 164 230]\n",
      "  [173 163 229]]\n",
      "\n",
      " [[177 165 231]\n",
      "  [180 165 233]\n",
      "  [178 164 235]\n",
      "  ...\n",
      "  [177 164 232]\n",
      "  [178 164 230]\n",
      "  [171 163 228]]\n",
      "\n",
      " [[179 166 234]\n",
      "  [180 167 235]\n",
      "  [178 165 233]\n",
      "  ...\n",
      "  [172 164 234]\n",
      "  [169 161 231]\n",
      "  [169 157 229]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[160 156 208]\n",
      "  [159 158 214]\n",
      "  [163 162 218]\n",
      "  ...\n",
      "  [152 150 210]\n",
      "  [150 145 207]\n",
      "  [155 148 209]]\n",
      "\n",
      " [[167 161 218]\n",
      "  [161 158 214]\n",
      "  [166 159 220]\n",
      "  ...\n",
      "  [151 149 209]\n",
      "  [149 147 207]\n",
      "  [152 150 209]]\n",
      "\n",
      " [[167 164 219]\n",
      "  [166 161 216]\n",
      "  [162 158 217]\n",
      "  ...\n",
      "  [155 149 208]\n",
      "  [155 149 208]\n",
      "  [157 151 210]]]\n",
      "[[[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " ...\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]\n",
      "\n",
      " [[0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  ...\n",
      "  [0 0 0]\n",
      "  [0 0 0]\n",
      "  [0 0 0]]]\n"
     ]
    }
   ],
   "source": [
    "config = MolesConfig()\n",
    "all_info = []\n",
    "\n",
    "# path of Data that contain Descriptions and Images\n",
    "path_data = dir_path\n",
    "if not os.path.exists(path_data):\n",
    "    raise Exception(path_data + \" Does not exists\")\n",
    "\n",
    "warning = True\n",
    "\n",
    "# Load all the images, mask of the Dataset\n",
    "\n",
    "for filename in os.listdir(path_data+\"/ISIC-images/Images/\"):\n",
    "    print(filename)\n",
    "    if len(filename) > 16:\n",
    "        if warning:\n",
    "            print(\"Maybe the filename is wrong, should be something like: ISIC_0000000 , not ISIC_0000000.json or something else\")\n",
    "            print(\"Now the filename is \"+ filename[:12]+ \" check that is correct\")\n",
    "            warning = False\n",
    "        filename = filename[:12]\n",
    "\n",
    "    filename = filename.split('.')[0]\n",
    "    print(filename)\n",
    "\n",
    "    #data = json.load(open(path_data+\"/ISIC-images/Description/\"+filename+\".json\"))\n",
    "    #print(data)\n",
    "    img = cv2.imread(path_data+\"/ISIC-images/Images/\"+filename+\".jpg\")\n",
    "    print(img)\n",
    "    \n",
    "    img = cv2.resize(img, (128, 128))\n",
    "    mask = cv2.imread(path_data+\"/ISIC-images/Segmentation/\"+filename+\"_segmentation.png\")\n",
    "    print(mask)\n",
    "    if mask.all() == 1:\n",
    "        break\n",
    "        continue\n",
    "    mask = cv2.resize(mask, (128, 128))\n",
    "    \n",
    "    info = Metadata(lesion[data(filename).iloc[0][2]],img, mask)\n",
    "    all_info.append(info)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "07e27c6c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(128, 128, 3)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7db6524e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# split the data into train and test\n",
    "percentual = (len(all_info)*30)//100\n",
    "np.random.shuffle(all_info)\n",
    "train_data = all_info[:-percentual]\n",
    "val_data = all_info[percentual+1:]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "16d9937d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# processing the data\n",
    "dataset_train = MoleDataset()\n",
    "dataset_train.load_shapes(train_data, config.IMAGE_SHAPE[0], config.IMAGE_SHAPE[1])\n",
    "dataset_train.prepare()\n",
    "dataset_val = MoleDataset()\n",
    "dataset_val.load_shapes(val_data, config.IMAGE_SHAPE[0], config.IMAGE_SHAPE[1])\n",
    "dataset_val.prepare()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b4b0e742",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxAAAACWCAYAAABO+G6lAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAACBTUlEQVR4nOz9e7RtW3bWh/1672PMtfY+59xbtx4qqSSEJcDhlQAOBsxTxCjGBmMar5AEpxE/4sQxbgSD8YNmwsPYODaNYIztFoiCgx38bMJBChDAPCQskIVJCw+LRwC9SkilqnvvOWfvteYYvff80ee5Kgppnwuue2+pavxaO+3uvfbac84192ztjj56/75PMpPFYrFYLBaLxWKxeDvoe30Bi8VisVgsFovF4vsOq4BYLBaLxWKxWCwWb5tVQCwWi8VisVgsFou3zSogFovFYrFYLBaLxdtmFRCLxWKxWCwWi8XibbMKiMVisVgsFovFYvG2ec8LCBH5u0TkD37Ka3/57+A4Xy0iP+r4+h8SkU+IiBzf/xsi8o++jWP8OhH56598PSLyo0Tka0Xkj4nIHxaRLz1e/9LjtT8iIv+ViHzRA8f9ASLyDSLyTER+4ie9/ptF5OuOf//CJ73+L4rI14vInxKRX/a3ey8W3zd48ZyLyI8UkV/xLpzvn32nz7FYLBaLxeKzn/e8gPg08jXATzi+/gnANwA/7JO+/+Nv4xi/Dfipn/LaR4Gfnpk/Gfg3gV9zvP5PA78jM78M+J3AL3nguB8Fvhz4zz7l9X8nM38c8OOBf+QoNJ4A/xjw4vX/rYg8ehvXvvg+Smb+mcz8P70Lp1oFxGKxWCwWi//efJ8pIETkt4nI/0pEVER+v4j82E95y9cAL3b3fwTw7wI/UUROwIcz86+97ByZ+VEgPuW1b8/Mp8e3V2AeX/854H3H168B3yEiJxH5GhH5wSLy+UcH4bXMvMvMj38P5/tLx3/jOK4D98C3ATfHv3tgvOzaF+8dRxftG0Tkd4nInxWRf1JEfqeI/Omjm/RTjy7VHxeR3yMi50/5/S8Tkd/+SV//GRH5L0XkPxWRX3y8/pdF5DeKyB8Vkd99vPZBEflDRxfsa0Xk7z5e/7+JyP9FRL7q6G59noj8L4AvPN77L7+7d2ixWCwWi8VnE+29voCD/7GI/JGXvOeXAX+Y6ib8ocz8k5/y8z8F/F9FpANJdRz+TeDPAl8PICJ/H/CvfQ/H/rWZ+YcfOvnRBfj1wD9+vPQHgd8vIv84cAJ+TGZej++/AngD+KWZ+YmXfC5E5H8J/P9eFDki8tXAN1IF3q/PzP1lx1i853w+9Wy+D/jrwPcHPkb9HX9LZv5UABH5jcAvAP6D7+U4/xbwDwPfAvz+T3q9Af+PzPyVIvIHROSHH8f+BzNzF5F/EPgXqO4VwJ/LzH9SRP4l4Bdk5m8VkV97dMwWi8VisVgs/o75TCkgviEzf9qLb74nDURmXkTkK4B/A/iC7+Xn3wH8HOC/zczvEJHPp7oSX3O8578Gvuxv9+KOouQ/Bn5jZv754+XfCPyqzPwvROR/DvwG4H+fmd8oIn8VeH9m/om3ceyfBvyvqUUjxy7yzwW+lCog/qiIfGVmfuvf7nUv3lX+u8y8AN8uIt+Smd8OICL3wA8XkV9HFZofBt584DhPMvObj9/9U5/0+szMP3N8/U3AB6hi5d85nvMNePpJ7/+GT3rvD/jv88EWn5uIyD8D/DzgL2fmP/FeX8/ic5P1HC7ea9Yz+D3zfWmE6Quo3f9fRy3Wvye+Bvjnga89vv824Odz6B9E5O87Rjg+9d//5IHzKvC7gK/MzK/85B9RO8wA3wG8/3j/lwMd+JiI/KyXfKYfe3yen5eZ95903KeZeT1euwKPHzrO4jOC/F6+BvhXgF+dmT8F+C+pv/H3xrNPEuT/6AfeJ8Avoorlnwz82k85bn7KewHm8TwvFi8lM39rZn7Z+h/m4r1kPYeL95r1DH7PfKZ0IB7kWPR8BTUS9HUi8rtF5B/KzK/+lLd+DfDPAV93fP+1wD9CjTG9tANxVJm/EPghUk5M/xTwo4CfAXxYRH4R8P/NzF9CjTP9+yIyqYLhnxKRzwP+VeAfoDQNf1BE/jS14/xfAD8U+GEi8tWZ+auB33Gc+iulDKP+ucz8hkM78XXUwu+/ysxv/Du4bYvPHH438DtE5Bup0baHOhC/HPi9IvKtwAV4aHztDwD/kYj8ZEqT8zL+M+CrROT/lZm/5e1d+mKxWCwWi8XfjGR+6mbpYrF4rxCRnplDqqL8fcC/nJn/zXt9XYvFYrFYLBYvWOMMi8VnFv+AiPxRyhTgz6/iYbFYLBaLxWcaqwOxWCwWi8VisVgs3jarA7FYLBaLxWKxWCzeNquAWCwWi8VisVgsFm+bB12Y/pNf/5dz0w1RwWOybYJaIxuoCEgjgZyBZNKsIy7MMZg+ka6oKYaiJjx/tkMkrQuQdDXkLPSuQCcJwmFcdtq5Ee6QigKRyciJ0GA4keAOGUlOiAisKafWyOmYGq0ZCUwfxD7pfUNUcQK3xGiYAQJqHVUhHcISBHpX0iBHoAkTYcwBDdyTj7/5BnG959Unr4EYU+D+GoBiAvt1B4G0RmRABjMSkw49+e/+4l/gtScf4n2vPMK0cR33fPEXfQHX/Rmf+NgzzqL0c4cMwoNt61hXJAWRBAlMjSTqPAKGklGfCQncJ0hiCdIaoiAk170Ct2cK1/Emf/Ib/hy/7T//RQ/Zi75n3Pyof2bN2X0Ocf/f/tbPuOdwPYOfW3wmPoOwnsPPNdZzuPhM4Ht7Dh8sICKc+3mhW6eZcX935dHjhmSZzMs1iR6MGOBQxjHJ9bojCKeToaaoNITg/Kgz58TDsRQCQSdccDQDtU5Op/UGIUygSa2MQwbqSk5n4jCVTRtpwlBHAnAYY5KR1BpayTgW2RJkOFMdBAxDmqObcrkMmilnMaI7hCEZjJloGKrGnJPMKpzu9yvP7p6z9eDJKx+GCHZPIkCRWsj3hrkzrxOVQNMJVUyTXQaNxvf/yBfx7R/9Dp48eYzoBFE++tHvQENRhfsIdB+MhG4K6fhIIpzt1JAQIkEUPB01IxJSjz/QBFJIFFoCQiakJGbJ7k5msDUDG5/eJ26xWCwWi8Vi8VnJwzkQDhPHZ7BhdNsYV8eyQxOSiT8L9GQEzvNnO2aKacMl2UkMR0hEqR3zBNNOkEyHzKTtCRhzHzRVaDAJGsrHn77O+x6/SqaAJi4CblyuAxewJogJiCAGqoaqYCZgQgzqd0+Ge6B1IZxPnfaqcBlXttlIT1y9dvZzIipIJh4JKaglY5/sEdzfX3j1duO03ZCiIAa7k3uiPUESz8lp66RXt0R7R3JgqZxFyBn0bSOzpsiqGApid9wD2ZSTJPdzgiY9G+6GiKMIvgetNYJAXBAMSWibIZqMfTA9j06E46mIBxlH50IEHztoQ0RRWRsKi8VisVgsFouX86AGQgwyFYlGTpgjagc7JuOyEwJ6a6hADNjvJ9fngzkcEahlrSGR+B5cL87IHVUQTbTBvk/2+yAzSBP2FGYMkElI8OT2luETyWQEeO2ns50M3Wp0aR7jOGLgCZd7Z+yzPmAzIgNrDW0NF2G645nsl7quRJg5GGNnevUurmOA5NFtca4e7Dgf+/jrPL7d6KcbEAGvESrEMBNMQAVMYfqOdkHMCUnElIgk0wmcDLgf93zX6x/jzef3jBEMB1Xl6RvPcVdSFAljHMVM9VZAAnx3fB/4nGiAJOxXZ9wlTEW1Rp3kCCKOdHYm2oyJczqdQYMI+OIv/n7v2EO2WCwWi8Visfjs4cEORO8biR/6goY4SFeuOWmtowpqNUcfKkQTwoOmgqLgSaiTQi1mj938PWuEST3Z2oZaLeINmAozApsbakmTxp6DETXb71HdDBElqJEpHPAgpXH1nfRAvZHXGm1qvWGWOEqfAecOCrkHqh3PRPLoYlCb9j6EYQ4oBMx0Pv76J/j8D79GhpAzCRFcgumTRAjAM5CoBbuoEu6oCkG+pV3wkeyxE658wUc+xEe/7WO8+sormAaXi5PhPNlu2a8XTo9PTK8uw50PztOQHoRSVxoJMYkWNKSKDIckwQJpNa6VM0EcTWXsAzU5/i5ACh947f3v4GO2WCwWi8Visfhs4WENhMB2bmSCR9A3YUQAxogrSGc4TAd3R1VBhDAhMpCADEObAlGL1lmjNENrvKgppAqeWUJlGVUcpJdgWpNNNuZ0kgmRDIck2LaNSYAFiTGGY9rILZg4ORNTZRwi6YxgH4OtNcIV60ZTQTYnPIgJlz3YeqKaEMK9XxGSN56+zhd86POY16R1w8O5xAVtDbJE3pkJlniAeKIeGFJjTulIgjYlM+hp3Mfg8c1jmn0Mk2CfiUugJEyhmRBjVmGVcN60tBCixExEwNQIAp9BqCCapFB/FFEYSkgy5gQtbYV2rcKDIAhMO5afkVqtxWKxWCwWi8VnGA8WEOITbY1IZbNGZpKeqMBmJybJ3JPhTjMlyBJNa1kb1W58lJiXJCJpooQE6TWK45KYGajgqTRqQSwNyGCM/RBBCypKE8d77eaHTFISFSUE3INwp20NzXIcgqRJ4Om0rmy3J+buqAS5C/0EagoKl3kFV7wrYsF1Dz72Xa/zyvs2Pvi+DzA8kQ1GOtaEjc4+vcTTouVWlUJmVlMEoXXBmjEn7MOPIkPBg603ZOzA5I1PvMH7XnsfaXA2Bc0akcpa62+qSBwCbc8aPnNID6wLokK8ULdTHYqccRxCSokyJ6aKSC99iCgiBqJYe1gOs1gsFovFYrFYwEs1EA1oEGURCqVrEBFmJNe5c80JCTlqBz4ot6OYNZsf04lw9r0sWQVqlGeAWCImzJl4Ji6lY2jHzL4GNOuMjENAXU5DZopaw9QwUdLr2uxU1q2SiVoiqjCAWbv6IkprjXbaaF0xM/aYXOckXbFNcfHSfWzK0/tv44s//4M8Ob9CShUGEYkeu/6SQlfDjntT+oZEDbCJGjRRNBPRoG1w6nBKoWty7uXW9AWf/2Fef/omp668cnui3RhqgokiCk3q86hq6Si60MxQhZCsaxOpsSwtPQlS91ZMmJlMDyKTSGVOZ0YCikirbo/EO/F8LRaLxWKxWCw+y3hw23mS9AQQMmpkpmllGrgLTU5ogxmT4YPIRMVQpDbQrexDQwMRQZFyZJrAoWcwLXFvuOA4j0/t0A4kM7LEx6lkJN1OZET9fBxjQ6G0TSqPgkSaoJJVZIRgZ8W6lj5hznIdojogbgFZ2o7oVRCoNkZObnTj/e//YkQbEkfnICciRuQs1yJxBD1GmqIKHBSnrkutnKMwEAcTZRwZFlvvTL+iqjw63XI+NVKEzWr6SLeORRLTQYWMstVNO3IxwkuLkcEco0aX6k+FCIQEkvW+OZOU0p9MnWxiZEyEcoHKLM+rxWKxWCwWi8XiZTy4akxXUoPwhIB+MogkR836p9fud+uKbB0VxUeSI8hWdkQxAzElqTA2DQUDa0Z4MvadhAqo2yfRGmLlMhRW759zoGqk16IYF1pTfDhzzOo0qNCbgR0XL4IoWN84JpSIa+DDkQ7zGuQop6etGaQhWTapSUImZh0xsDDEg+lGhiCWh+i7lQaDPBburUqkDJrUaw5YClvv7HPSWuN+7HVfRLHekbZjGlzv77l59REaUi5WCi0Mn1XADAfRqL+aB5oCUl0fRhVkYVEuVyRa/SBmAlF6kIwg0crFyIGPo8CRpYFYLBaLxWKxWLycl247K9BOVhqHhGyJZjkizX1nkjQ626kjx6hRao081bi/lV4hlTEDJyCc7eaENIXhGJA+6doJ6ntTQTEmQTcrm9cAnwGmaAR5dbrWaJGQZIBqQ6T0F/miuxHUYtuU9CQ8SRUUMBMiHEaiKGadVNh9Vpq1NEKjFt9kHTeTTBArt6mIylVAAo8gZlQXxgQ/pAwmwqaGH69d76GLwHGeD73vNV5//RN86LXHNCt3p9iD4U4XZZ8QPplDShOSQFRnBy9thKjU5+Owb5USv3uChIIkLctmtmx2DbP5VndpsVgsFovFYrF4GQ9qIKwJaloiXKFGaUSwrbIV8tAfjOsgJ5gZZkbbOgbH6E+NOMkLy9BQ0BJkezinUycpW1Q7dtYzamzJI8qpVCEVaELrhwZCBNkqPE4OFyQRIdNxn8jRFQhq9n9maSgCJ0KIEEglXfE9jjyIJLWKkRgQnkQGiDDHrIC3CMhEebHiPoQbUpoKAeacTC9hhmkVKfbCocqDR6fG6VTfmwmJ8crjxyByFCaAZOVIWFViKZWTMcPxiMrlmIHHrHTv43wSVVAhipqRWrkXaWXraq1hTSEbWervOu8783wtFovFYrFYLD7LeLAD4TMYkiitEpq1VRKy8dYYjJ7KEnROp/eGmZGSFYJGLcodkMPGVRA8HHdqxKncUkl/K4GhxMAvRoOOTIXWGjNmOTxlkk3JPQl3RGs5nyFVBFBjVurKtlUydZJkCmGOpR4uTlE5F5mkj3IsCmr8SZKIwRjVDZnTiahRKbEs21oUXmRcpDBj4nNiUhax1QWQEjszaSZIKInSrQoE1NBwRBv79cr9mNw2w6KC4DSS1EA8iSxtSaYQ6WTUiFIJQKr7IKZHeFzdzm1rb+lDVOq++pGfEVldFxEt4cVisVgsFovFYvESHs6BGIOBsrXKUCBrVMc2rWA2rcXqC3eisZdVapJEgmc5A0mtsyHBusBUfATWjuA19OhugKrVWBCHVWkGZkLvrboEUfP6eWgEJI4v8oULFMCxa0/iydG+4LA4VaTVTv9bi/8wjqqIPBbXL8Z69jEwUaZn7f7vgeNVmFSDpMatanV+6Bq0rFWBaljUYl0iaWY4gWhy2hozBJEgs3N76rz5ide5+eD7kRSadjIn2WoULEOwai6AKXNWQrccWdMZx3wSx/0WRTU4bY0xZ2k7eGE7a0Q4MmocK1cLYrFYLBaLxWLxNnhwhKkWo0kwjgV9kJKMZ45ljeaISWkeMtmHcxmjdACZxAzYAw3ICTmF2AERHNgDZpRDkqni1wR9sSNeo0lIHu5NL3QGlR2RBCFec/9Z3YqsagGO1GdFGNO5jMnuTkQetqUCLUqfoZ00PearBCcqT2IG4UGmHCF2yfTJmE4clrWXfTJGsA/HY4IIaq2uCeCwvBUpAbOkIBqYJr0b22lj26SC9iL40Ic/jzffeF4CBU3KWVUwAVWlt+N+Z33+PAqp0MM+VxKfjnuUE9SRsG2qmNbYV2TdV6nBMeYx8pXLxnWxWCwWi8Vi8TZ42IUpBcnE05FMjNI+VCiEYZJoixpD2sutyVJIE/AkPZieiBm4Y1Zz+xlKatbPUEyDkUGTjsZRkEjCSNpmpVXIFyFpoAh5jAeJVvdhXp30xE5W1qlaOQfupcFQE2jV3TAqPE6ldBbVzaiBq03AAzzLJlUiUFWy11jU8EkTIQI8jrA8gUhh6yVkJhJxJRX0KIDiGOFC6/rbUeCkNBoTt+R0XLvPRHVWsvdU8MqTyFZ6EBFhXis8TkVJqS5CHAJvSSVC8CgdiJJoypHlcYyJiRwdiUqvZhUQi8VisVgsFou3wcMuTPPF7j/4dGJeac3Q3skZ6KFhMMqRiAzSkznyCEHTw1JUoFUCdYSgEW+lRI8x0a3yIbJNIlq5CzUjLhN1rQV81CLZDytUpXIVMCHimPKfjrRKfi7NQ1DabynHI44uxJHqPGUeYu4JHKNZqUirhXXMSXoQKUhls5E5cVOmV9CbR407XaO0HtumlTDNiwV8dSGMoyiCIy8D8EQ12UzxTSGNm1tlutOliqTW7HBV4kjrrrEoa8p3p19kdSKO+x+ZaFRBpAkzAxXQVKIFYwbdvMbIrPQaq4BYLBaLxWKxWLwdHhZR75PT7bksW4eT1x19fEPix+59dQWaKGG1+x5zlrGPQsBbu+Ql+qWsYDMwBEl9y1LUVEEdD4OZNIwU4XJ1NgNptVRGyx1JtByS/EWgmgh2UkTlLRFxIEfgGzStoR2Rw1+WOITRwnV3brethMhIhdchJUjOGg2KGdiLcirqXxJvaRAIGNdJ5FE4SRIBog1TyuKWSuFOUSLH4bokqDQ2g2zB7ZONZ8+f88HXXiVikBJYtyoAMNwTJ3H97i5ChpQrFTDnlTKRPZUmohvhQUSgKOlxFGJBhhBaxZL7KiAWi8VisVgsFi/n4QJilBbgdN4Ig3HnxAxmOCaCYKVPzkSUciXKVotZFSQm4RPpnfTKW7A0wo+ciPC3hM0RkxOnt8TG7n4sjoPhgWV1MIzGGIPWFZMqHCYJLWnWSwidiZrSKI9XicPilHq/vtAUZDAQTBJnINlwqiCaVFcipcTTLwTSzXrt+784j5bOoOTXwRhJxM7WE9WO+otAvMqyCBylVxfBFD3cqvbrPdI2Hj96hW/6zo/x2vuelBA6KiBPDAirnAx9IcouMbRZ5T1kAM0ghCAZnsQEiXrvPNKpVYWRSVep0ajDxWmxWCwWi8VisXgZD2sgJNmfXdn6hoqhrTG9XJOmOxmJqhwL+RJOSxpWXqOlocDwCZJJt0bMmrnfr6NsXT2Yu9MUvOUR7DYRUVQaLrN0Be5HnoIQqpDHoln0LWtXkQpLyzwyHo6RoWZWo/9RQz+i323DqoegGSBGpWY7CVFFTbgQUovwiHyri5FRnYckkVY/JxOfR7hbKuHOzOoyNIUXCuaUvfQkIyq/QSscbgZs7VxGqxqIGOqlQ7He8RloCER1Z7AqctQEZnUW7BBxu0d1YlxpZqROYgYtatRMtFLEUyvNWq2/s0/aYrFYLBaLxeKzggddmLbzmTTlOnbm/V5uQejRbTCUZErlNgiGSiszowRmZTiYNZpqpSJ7Il2xzehbw9TKsWkvYfa4v9a4jQQxkhkTpNKjHeGyO/sssXTZoxp+2JNGwvRDNKxKOMQ4BAOSiBiiFeymh62r6iEKV63PMDiKhNJ3oEeC80hyT8asMLkZh6uSJOKlun5hINskaQKmDZGyo00/xqWyBOQeXpawGUd3o5ynMh1VxWOi2sp+1ozpid8fORkacNRDJoc/lSeq0FqNMkUcKRoBzAAEaxsmRrhXsRWKSlQDKLMSwheLxWKxWCwWi5fwcA6EOkllIORIzAIVqZwGA2udRtKakhzmRyrg1M768b3Y4Yi0O9ODdmpoN8Z+JY4RprkfFqxSNrDndgJ62bUOp21lYzpjHB0NqfwIlBwJASk7nlZajAisld5hRqISCIkeDlAhR4FyhMEJkB0giCiZQ+uN3Qd+N4gRXCPo51PpBVrpCSKytBGtugxqjaaGSWkpKsCtgubcK8X6fNqYx8gUGbiPqg2yRNJ4MHc4narjEynInGgz1Bpjn/jwI7uhiiFRKxF7OJn1d7K0Sn6Ydd9NgaaI1TVBXVeGIroKiMVisVgsFovFy3mwA9FapThfxyDE2a9eYzczSx9xOAChtSXe+pFTYA69RntCIJsimyFNmR6M6yBm6RRUqwy4jh1USa9xp+f3d1zHlTkq12AfV1CpcSStuf0gCEAb9LPSbzrWDLJsVmckKQpSKdFzOmMOqlFR40eKoI2jQ2GYNOywjQ0PNButndluDAh8XOHqMAIVg/TqcngJwaUdqXlZoms7bFaDCqZT8xqtUsGZhAjf9Ylvh4R9TBqdV195zLd89FuIJqjB1hTMQbxck5oead9RAX+ShxtUdU6adZo1olV2hXvgUeLrvlF1GiU4RyrMb/rSQCwWi8VisVgsXs6DBcTcJyJC00aMKhrGDNIFbQ1tlS3gPolRmgA1OJ37kbicTDmC2TKhCbYl2sDDiSn0bthJ6L2zj8EMZQRcLpN4DsN37n2yT2G/VnbE1huZwhhR3YajkIgpeAQjd8SE1hTTPFyeXnRK4lBAQ3gwY5TWAYCBRhU+5kJekzkntgHnzvmms/XqcMxrICls23aEsAUxgrge7QuH9PJDamrMDGYEUI5U4oJMxZrw7NlTPB1mxfa9+ugJ99d7jIl1RVoDq7/BnI4YaLdyU0qYUSNSHoC0En5noKKoGCIvdCDJcDv0DtUhyghyJHf313f4UVssFovFYrFYfDbw4AiTWee671gY+3DMFR4bU/YSOI8aY5oOks7wGueJOGxCJcqpaQapihkQyuWyoxYojX0/AtOOlOT9frDdbOT9zmXcs20b4ck+J3mFR49u2GOWDmBTUoLcHTEjdq9xpFYWq5ghXekpJUg2oZkQEyDQHozLYSErA8GYI1BLJEu7QCYRk2YNOXdwwCfXeSV35XSz0briMxhRlq9IibvNtFKep1awXFbSdkYyxuCaJZT+4Ps/gHvSD6H1zc0ZUcXDsFZOTyZ2pD4cRZAJdqOM3TGs9A4aiMzK7BAtvYeVcHyGVIFHJX+rdMIHKgEibC+JBFksFovFYrFYLOBlGggZ0Du44JegqdYcvgqilYoQl0kabDeNzCNdWo3QQP1wSxoBzComjrwC8Rpp2sxITwZJ741LDjaFmyc37Jcd2ysILpqSO1wuF0SN882J0CDcCWoB71ndEQtl6zUa1dRIDSQEyUAM3APVspttrcLmIsElyebg5eYUOtmiMyOIWXqQUi8rW3bIRAfYaUN0EDmJFGIml3mlW0M3o1llU6QdDZ+UKm7uE79eOJ0fo7ox/YpodR0i65p6gLRATBBpzFl5Gip1/5oaPipkbxKECdqEOSYjFG2KJ/gYmCqnVm5RHLkSkQORTli+w4/aYrFYLBaLxeKzgQcLCDFBruUSdHN7w5xOS6kU5MNWFFGEKF2BNtIdmU5KI3KiQGiSowqH1GDrjflsMmMgQxg6mb6j0jmfGsMvtNNGywpBs5tetqOaRJT1qd4LokeK8uGu1NSYVwd3ajv/sF6NsocVpILvulFWUULfWmkdEJzqYIxnV6xbyRu2Ss2WKew5qkBR4RKVV4Ek2oRt64yZZEzmmEhWSJtF6QtSS6gcCfu1XJgAeus0uWHMe9I61oWrX3Af3I17ujYyb6p4wbGtcbdfP8nStca2QuII6ZP6fKpcrwPNBDF8OK6Oa8cjObdW3Zv663FEgy8Wi8VisVgsFg/yYAEx9mTbGnOWGNessYcTzx1Vp50M6TBncn16BYHz7UY3JfeduU8yk9YqbyDCK2ztmtgmzJy18ufIIyAZ12DrhorCpkgqacmIAE1u2xnxyfPves72uNGiczq36npkIFtlUpgJGY57Yq1ViJtSC+6rY5KkBWYNs0qPHlN4vj+jbUbMJAboFkCWK9Nhy8qAbWtlHyulk8hJjS3J4QKVkC7UFNREzXHpRAh906NgapUp0YN0ZV4miWFz48mTx7z+N97gAz/gw4g6SLlJRU5UjXF/6DYiICujo2eUXS4Dw9AG93NyN/bqZghcNblpSYoT0xGrTs1kf8cftsVisVgsFovF930e7kBM4crApNG0MY95ezt3JKIyFBBUlI2kb50mrWxIMVort6NxZDcwaxHcFDxGLYoZqHdMOmdr7BkQwRxOP23EGOz7oOtWacq70x9t1QU5kqpjlO5AVSqITkobISYVeJdyCKmr8yBe3QEaeDoC+HXQ1djahk6t1XY44hDsRATTHZMT2KRTidp7lquSIRhCitZ9ccdFK/k6DQkBnVwvDtlp25kYUW5PMzBTns+dzQ0afOiDH+Av/pVv5gfmRzCBmTVq5TUNRmsNkcR3iHSadUQbTiBRhZ5IuVadMvFw1JQtoPcN0hGxSgMX2HQFyS0Wi8VisVgsXs7DylkVmnSu+0BqkolLTM5suNacvU6IDLoZY16Z4mzN0BqMwWd1Fbg6IoFbhbXtY6Bitas/B9fdiVOgmzImpCZyHWhX2m4V6CZC20BwrAkzAyaMMdFWzlBNDDzxmqlCs5XdKuW4mpE1SkSiE6RXZkQEXHJQXqtCU0EfKde7vfIsgHPruECO6ioMSucheuRTZKCtQypjDyQPx6m546r0frhTXcq1Sa1yJs6WzJG4OK8/vauF/mbcnm9xB7EG4kQ4OhUfXmF+TVBtxAyQQFqFw/kUclSHJ4FTa0RW8J0qSDjZBAMQKycmWTaui8VisVgsFouX8+Dge6TQTsLNo0aidDP2PbjsV5pILbYRem9gBlYLWE1BqIW0jyghtSZuQpOGDkWvAnsgnszd2RCYDsPZx05cBx4l9CWNPXYyJk4yJjCFE42mUjkL08kIXCbDXgiWBXDEHDQwEcSUPLIk9iHsz51xcSJnCZF9Ejh7lOi7bUJGJV9Dr1yHKYQLPY2uWrv4hzXs9AFQI1RzMoYzskaa5mUiAtINxbCpbHkmPCuFzwW/OOcunBqA8+bznXFRmJ0MPYTSiaoiWZkZooAJkUlJK4TeFOsNUcFJTJRNlK0pqkZHaaaoCdMrv2OxWCwWi8VisXgZD3YgLnMnnnfa1rDuyNZ4tFmNzrhzblZ5CFGFhmonSHZ3lBqhAUhPmtXvSebRhajMg4aStNIRBOSctGbQKlH5+nxHJbBev48nHoMIpZ8bJlIJ2NJKxE0g+BEgV45Q0wXRZI/KPgicMZymjcSqiJmV/9C0M+ZArgGPD7ckcfrWGfeD3mt0aI6JNIEjpTuAkUFrwoydjIRjsU4EpkpmkNOICLRB6EC1o7kxr/fcbidmODoF5on7/R4TuF6fMq4NtsBoiFSInolCCzQVz2BQidSSRwdHq7h7rMqMnSaGi9CbHWZSiacAk7EvEfVisVgsFovF4uU8LKK+DDgHoY6Flp0pwrkZ0huK0G+S2HdyOK6BmBJzHuLlSm8WvGxUTZmSeAQJXGcg4ph2XBrDd9RAUUSMGPW7WENHMBKmBDTFcHJCNGOOQYwrt+dbiMo8UJQeUiJli6OLcOQ6iHN1uHtz8OjJI/IoODatgqepET3BQbuifcPHIKU0GNoVFWWORD2Ze6U9J8negr5tlbhwWN2qGI5jR8NHLfEZiCbSJpfhtN4QEx7JRs7gGjvhgzfeeM4HXntEiTEEqEyNathM1PTIfxA2Uwwh3JnaUOv4nJV/cdjOGsKL6L1EKv26bzUOtlgsFovFYrFYvISHOxCe2ABORrZKmpbadCfurkQTNhMiEu2dlMm+7zW6tAfj6qiBiVRWwwhiQowSJ4sHIw3RgSjkTKbAzESG01rgEUAnCa7XvULWRhUSIko36OcT0QfTE5Way4pwhkilNmdialh27u+v2Hljy+Ber1z8QkZyMmOfja6w75AkMoMIQVJwDGnB9fmgh6GSqAUZAZsQVxjhxDUY+4XeG+ebm2OsqnQfEmUda12QVpqEeYTf+TiKl7My75wbO/GRD34+f+Nj38EXfP4PrEIohTnKplaaEu0YuZqVa0HCFCEd9jlAk8wAaTQt0bnMxEJxL6vdJkm2xHwVEIvFYrFYLBaLl/NgAfH4yUbrVo4/c3I6dy7XCy0Fvd2Y+6wFe0+uz+5KfNwbYobdGniQs4TROsGj5vVdApegmZEkKhWEJhlkWCVcd3AXujbmZUc3pVvlTLgPdDNmTnQXtHf8KDpCoG9HwNqskafsyXAgnbRg7DvuEDGxaYQETy+D3hpxkpIzOMybhPvAehC7kFm7/SOD9CADpjsxkhiBAiqKiALKvk9OvaHNiMuETRAV8CirWDkcpFxIDa5X55F22k3DZ/Chz/sg3/XxZ8xRydbnXhkPMYydwX6dqHXAuHpCOF2S6YMZJbJuVt0aMa0CzcG00TsEiasiHiUXWSwWi8VisVgsXsKDBcSjR7fk3DGC2ZQ5B9vWKmG5K2INQ3A3+o3WKM8IxIV2UqQp07NcklJBa8d8OghKaqIpaJTQV1XRTBrBZThqRrbKVhhPr/RzZ7/C3d097//Qq5goksK8OKebG3zMslu9JMIAVXQrMXU4pS/QEkVnBNu2cb1eUVHyAtc2yOz0FFJh369lsborfTN8d3w6MZMRDh0YULNVk5kltraWiE9UYe7G6eZMRukXIit/IczJBtJgXgNz5WYzxjW4PRnPxl6jY6fjXO4MbWRAa4ao0rQTI3AtW1cTCPfSSWiQWXkXmlruUwZkMtkRVcKzdPD9GHFaLBaLxWKxWCxewoMFxNPLHedzI65BNuEiwQ3GPieajvVegW9WmgBUiL3ckHAlIvER5ISRkwwnPVCByz5RFZoJ0gz2JDJo5xP7cDajNAdeQW7TAtGgP+k8fvQIV2GjCoRIuF52rAuaUr94ZFR4ONoEEUUuUWnRWlqGdjL6qXO9uzJmgBpxuXI3Bjc3t2gz0pNr7syxlW5herkdRWJXYb+b+HQ8Dh2EJLYp/VQhceNuMO6Dftpwv3J61BGDDIFZ4Xl9sxKcS5AZ7BNuzjfcXe5LhyGGW+BRGRPpAZEQgZT5FXMOZoKqYiY1smQl4BahRplUiK06JzLr/hACBvmwoe9isVgsFovFYgG8pIAIDwhq5xrnvLVy+DFDA7qCR9BMGQkthcnhutSONOY00pSIYL+/Z85JkEyZRDindqJZR09gzwWfVyICOW9Il7JFDWPLhiBYaDkd4cwdJCu3QZrguyMkqlKdhkO8nbN0GvN+cr1U1kNoggbtdGa6c/XJvN+RACxxf047dU5Si/txHURATMfUSILpWfanqcw92a8DrALrxGrkKQR0VN6EGRilT2j9ROvGBngKMWcVLFm5GsxE0gG47pPeFd8HgjOPYi3TMW34ERQnIvSjWJCs4iABVPBrYGFs2nFxmCB2ZHWM6v4sFovFYrFYLBYv48EC4rxtEEnrglgvkW4k3ShRrwt9ExKhWWNmolEJ0ClZv3sWwoMmSqZxuZQD02lrIKAqTA9UkmjG5f7K1gUhsQZmHVLII80tpcaIzBopztwDPRmZVkJpTTzrHEJ1IzLAh7PjRBdElXkZpAr7eM4ewf1lkE1wF9pI5jaI+52bzbh9dAYJPKME5NMr+yEUEakFv9bYU3igQ5jqzBloHCnSIjQ7Yb1xcxJCpCxorTHuLthWmg1RJe+D3BKh0azx8e96nY98wWuEJ2pa780kZqucDA26VcenPnmWoNyhda0U7hI9lFYEiHbclwnM6jAt3mOsYV/6I9+yPwaYf/Hr38MLWiwWi8VisfhbebCAMD0C0fxIbB5BSiAjmNT8vmVDshatWzMkgzEnGVJjTXIs5B36zcZ2ceZ0sB3UaKn41dlnltvRM9DXhLSs8aae7PcTNqsuwji6IlsiXcmmiBpE4mOg517J05JwaCuyO2HQTx25Bld3UoT7uyv3c7DvwX516J3nz3ZCgtYmp74hsRFyX8VJwNiTjGTTJHIQXiNUgtJODZ+TlMRDjqA3wBO5TuS8Medk2wxVgQhcd+xkSICkI2qEVq6EmXJzc+Ljr3+cz/+C91UnqEkVCSKoJhCoVvdneDJmFXzdlPRyf7IowbSYEVmWvDmOESaF3SvXY/Ee8P4vBDVe+f5fwqvvf8If/JVfhll1gzKTH/0rH0HCG3/1r8B3ffN7fLGLxWKxWCwWL8uBmEEAm274nnRruDt7OL1VTkMqhDgpld+gZmhIjdEI5cyEMveBiHI6n5D7C8/mFbOg9SeoGUagody+tiEShFNuTgrar7Q8E3vlSgQlhp6Z2Cxdxen2RKgQs3IaEghJMhP20hdAdQNiBvu+c3ed3O/OdTo5k4jBcx/MfWIqvHIKmgV5ASWxJqgImcllJCnQJBEJpAMl1wBNXKuIwsrlKGeNccVUppbuIYGGgkHEwE0RFaRLFWUpvPrKDR/7+CfYZIOeiCmIMJmICeLATNJL9uFkJWibQgYiSaQi7og2pHXcR+VhaFSidSvb3cW7y6s/+sv4BT/jh/G+G+OX/5QfyNb+1jGyv/bbfh4eya/6fd/Iv/erf+t7cJWLxWKxWCwWfzMPFxDuhyUpqB5BbFQgXEplSM99EuZgAlYWqGRpEyIrOwIBuhLDsWakgGEV2JaVmGyUboHWIGflJqTQW+P86BWeP534ndPaRm/K+XHj7m4gYvSzkeF1rihLVcxAITxrsUwQGexjMofz/G5yHXAh2SekT8YMnt1fsYBrBDkdlyQfdW76RlyD6c7JpO4DWpoQSSITEnpTwiZTr8RQIjf6uSPNyIi3iggmqAi0REjaTcevgZrgw8k0WofmHRNj5qD1BgaaQtdOUFa07kEkYIJpJXyHKK2XHS2zRNcRXjazqUyCzTZCBl2FveU7+6Qt/iYe/cifyJ/8138mH371/NL3mgq/+st/EL/3j/8MvuUPfhXcPOFLvuyn8vTNez72tf/vd+FqF4vFYrFYLL6bh21czxvNBAIiE81KTB57WYS6VldCyx+0HH8ArXH76lB41kiGVO8gZHKNK49efYzJRu6z5vkVxnVUInQzTt1KAJ3Jpje8efcGjiMpKBs+ktNmjBnMQe3Gq4Al6UJE0E4NbYpfR2kosrIQ7vbJJ958zj5Bzg3fg9TkjafPiTTu7y4gWRqGG6MPQTyxSGYkLol14dQ2QgQwMqB3A51EXBjzHlUDaYhUkaRoFVjAuJ+0rdF7Q0UY4bSuBKWb6GcDS2RKdThEKc+mBBUUwdNL09AVda/PLFU42aG7SEmSCqFTKUtdiaNronUxTYXzElG/q/yuX/7384HH29t+/7kbX/d//HJ++vtu+MIPP+Z3/+IfzV/41jf58auAWCwWi8Vi8S7zkg7ERFBUKpEZ0bJhtZqtH8MRVbS24cs9SSBRUCUliHDSE7QWvj6Dx49PZKvk5O2mM66T+6vTbozcqdF80yPbTNkvINO4PTV2BlhyuduxfuQzjJocql9IZjoekDtspw0azKszRjAjefPNC0/vdy77Fe6FEcKj245rlK2pJNcxUBX2ffLsCFw7iXHdo4obd+YpOakSs4Lrtk4t+qNBNPZ9R21nO59wDeZ0DFAF6V73FCVoJYCWSY6jE9MrdA8DFyflEEOLkRmElBAbVQiv4L09yBBMq+grN9sSa0sTcpZmAk+OVgWmgkdU92bxrvGRV841Zva3waNT44//yi976/tXbzs/4n/28/nrf/VjvP7t3wnf9Gc/zVe5WCwWi8Vi8bfyYAEhDns4TQRCcUBNyk61CWLGpJyXTLSEulkL0QgvJyapwuL+csWOERuzXk5IDpihM2jWQAM9aS2QSSIbOYS7fUe3Bgin1uCk5Yg0BltvbJvhkUcwWhUepgkSzNhJqTXzdUbpHi4797tzdxkkE/fkzbvkyStnLpcLibD7Tu6T+ETy1Izbs/Hq6YbLxdk2o4miV6d12FrnUYNUQzH66cy5CU8n0AxrhiLHyJFj6SWV6BUStzVDRbjcT3wKt6etMhtmEkO4fXTLx994xud96EkVSVFjSl0VWtnMOhNzI8Yh3qZEuEJpVOTFuNQ18T3YHknNN6WW0luWBuL7Gh957Yav+iU/gW/8tqd8w994nf/zV35/vvUPfdV7fVmLxWKxWCw+y3mwgNi2jjMZd0FvUZoHV0yUpolZ7fyP6bz5/BkJmCoiWpoAIBzSB2M65/OJrjXznQlNlX0vQfTWO3vsaBNU+zGzD2N47bSXSps5yp3pdHNmj7uyPRVBLJnXK2ZbFRFWAuoyNhU8YHjy9LJzCWd6cL3uzBgVwgbs+2DMwSYNE2fsDgOsnarjcB1oNK7DMCl9RohzvonSH2zCuTWExpZKjgvSFFXBRMvyVsDMvtupyZPrLEtZlY6cgrSqeCKC3oTXXn3CX/+mb+IDH/ihqCZqwn4NiEk/9dKVpCNN6z5JZXfELF2HmrJpva6dI4lbytVJFGtB+OpAvBv8yF/48/nBX/warz16++NLD/Ho1Ph7vuQ1/p4veY2f/MUf5BeY8E1/4Pd+Wo69WCwWi8Vi8T3xYAHRVTCMULCt5uhjeC2EFRQ57ES1xm1E6z1vzc/UccSEV2+fVJKy5+FkBOlOhOMpbJtie6uiRCqvQFVLwzBrnr+sUROJQFM4395UYJwcNqRa+RSCMj1IAqYwZnDdJ3f3O3fXwf2+c50X9nkhPXAFHc4+J55e7lBmbxVB934P09ibopbYmJgI5+6EVPr0zY2ye4XFdcqW1RAkS8eRHTiBoczpNIXwul/OQNxo1mjyovAKWjfUNoZ39uuV9EEcn08QpnslY2s5M1VhAIkwQ8kw9uFIJmoctq+CbUq4E9EhJ0IlVy/eef71n/0/5Mf+gPe/I8f+H3zkCf+7n/F38y/+gXfk8IvFYrFYLBbASwqILO0z1gWRSp22ZsQxtlRiCMdMefLKq2/lMWREuf9kOQ2FVHZCjFlhcNRCf5+Vv5AecHOqgsQqydo139JTmApoLY4lQRuIBKqNZopHAIq1XvEPlAD6RaDc2Cf7vrNfd8Y+GGNyGTuOH9rvgST4nCDCkHJsCjquMH3n+XRkF7bTYxpGaw3ziVoVK9fh3O8TUZBNMBO204aokVkBcJH12XJ42bVmsI+Bng01J6YQ0lDlrZEi1capdR7fnMkMfHdCskoIScZ0VGu0rMTSiXvpOMgkVfERDBVaP7oQre5nWeUmmWt86d3ggz/hy/nAp6nz8L3x93/p5/GD/uGfzV/6f37lO3qexWKxWCwWn7s8uO0chzi6tUpvbqr0ppU07fmWmDejdA7p5cIkUuNGqlV4iB6BcplE1jx+AtODu2dXIsDdcQ+mJx7gUQnV7vLWmE7fGnYWXJx93BNMMoU5sroUWuf1PNKoE2JWsN0+Hfdk7jvX630VPggeTo77cjQSJzPY9+dcr8+57s95fnkTj527+ze5vzxj3+9wScbcuYwLYwzm1Xn+bHD/fLBfnes+CYGbR2dSarxJVZGpzKujb63vky6CpqCRYImnE1T9MGeNUUkzVJ0IIxPmmMyo93k4sU/G1REUTPEX2RStxqxaV2ZWMnZmEiZIU8QOkfuRX714Z3n6iacMf2eLtR/0+Y/5WT/+i9/RcywWi8Visfjc5uG5lYgaf3kxw9+3cuvJRDn0DS/WQ5GExzEa4+XpI5WRIAQqQjfDUHJU8Jkey1bREjmnJD5KE3C9Dq7Xnet1Z/fB/qK4cMEnkGUne91HJVvHCwF3FS0cy+JKWSgb2iSIOUkPYg4yBpE7M3ZmPEcSpg+m3zH8jn2+wWV/g+v1TUZemHNwd3mdfX/OmBculzvu7+/Yx+C6Dy6XydgrK2LExLbOiKPrctipmpZg2vcaLWqnRoojWhatwYCIivl2GPeDdKX3zrOn95U+oVXEVYcma1yrbiAqggESUloQrURrQcqFygVxOXI6vO5pZtnDLt5Rrn/+6/jP/8K38+wy39Hz/KIf8YX80J/zc9/RcywWi8Visfjc5cECwt0Z6YQ7vifjbhzhaUa64GMeXYda1NKO8RjAvSxd7RA1RCbaGq1ZZUUcC9zt3LGm+HVWR2MmuCIzIYTrmOzD8TGZcyci8UGN6YSUq5FJCZMTMut1k8qFmASZgakyo6xkmxhkMqcTfk8kzBmYGZFX0neaBtu2M/wZOe4wqjuQHlwvT8GvzLxn+D0eXgXEdEYI7sJ1Dy5jElYdGFFFpK5T1VCx+iwkemRejLtJZt2762Uw9kr4JoT3vfKYb/not5cuw2oECREkhBmHYDzLico9ySydgx66EEKr+HLwGRBUJ0STSQmuF+88v+m3fw3f9on7d/Qcf9eHHvEf/hM/hh/8s3/OO3qexWKxWCwWn5u8VDmrGOKGX52+dawrmOBUZAJQM/Q1PXMUDZXQ7CMID2piv04190DQ6kDIi4VwOSntM7m4c70faBp+zcqhUCE0Ccm3ipUIL7H2sTjXI6wuLRGD1g0I3Ocx9uOYav0uDaQC3ARDI7FspZuICli7vz7ncn+P58AJ5rgy/Tn7eM51/wR3+1PQnb5Nbs4Vkjdjsu8DH4HvZZeq0koDUvHcSDNGeKV5p9Rif1RGBQ4xEvcoD10JtJVt7s3tYy6XC+6BR+L7YVt7FCYjSig+p5MBPubxnqAdOpJNDJPqHM29ujYvHoJcORDvCvlX/jT/01/3+/mWj7/zRcQP+dJ3Rqy9WCwWi8Xic5uHOxAhBF679jeGtVqci0Lb9NA2KHG4MaGKZzkIITVE5OEQEyFJLS2FALbBaTNuzycM4fa2Y5ZMH+zzyvSBWmBbjSVVotqxm9+EbhtGdUKYR9KzKsaLXANQUU5bo5vRmiJNOZ9vuLnpqDXAECkdeWudLoGQeExmJNdJJTTnZJ87KsH55PgY7Jdn3D19xvP759yNeyC5XK48u7/nMidkVo6DC5fLjs9kRjBi0nrDpe6ZdcWkVwfBwMchQg8w7eWc1AzRxFrjMuCyB+MakIlYohIwk5j1NyOEuTu+O+PeGWOSkZCBTEpjMp0IxThGnnQVEO8Wb3z9H+Hu+s53fH7u/+jDnH7oj3vHz7NYLBaLxeJziwcLiL61GgsyYKtddo+oBegeqIPvju8Dvzox43Bhqp10hmJx7LQfTkptM/pNR5swrpOuwtYNE+XmtHGyRjtviAq9dW56o6sgkfRzR3tZvkpXhif7mEzKcQmBGc6MEgX3bvS+kd3Qs3E6C30zXnnymN4boiW87l3xeM6IgaqTaSCBcYS5pRA62f3C5TrYM9h9cj/hOit0jqPO8Qjux+D59UJk0tV4/ROvc39/wWcwh7PHILW6ER4JYvTWQB3TKsDGyLqfWU5XEg0MpFfGgysVnjecGV6fPwLJOL5PfIJfgnGB69WZ0/EIeqtW0bhEFWBZQu7FZxc/9Qd+iB/5937Je30Zi8VisVgsPst4sIBIn6TAuDgxJmr51iJVJTAVmiatAdR73GscJwCXEhSTZfsqHkyZTCaecH600c6dtint2AE/P9q4udlop8bpcae1hmCVfK1G145YQywqg6IZ2RoqVrkRkkAyfJA4Ikm3Ei43UbZN0R586EOv8SU/8COHuNlQnBRFstfYFkIymFnWtLd2IlPZZxIxgNIcTL/y7Pk91+GINUKEfRxjWwrb1ujnG5IK5qt4jNp9HjmJEHIkZUOrRBuVqUHiOKpaI0sZWAZP33yKNqFtHesNT2FOCBTMwIR+UlqvP+2YwfVukKPcqsY1ePZ8sgP3Y/LsMom97tfi3WM/HLHeSW5Pjf/wF/+9SwuxWCwWi8Xi08rDBcShHZjU3L30BgF9A+mC56xpIQdNxbICyc43HVMhI3CppTBRdk0tlRiBHSNPrTdEpQLcuqHAK6czr73vMTTImIy507uCC6LHuJJ0yHKHak1rBEdqYX3qna6NdugcWqM6H1tj68rWDFHhem+c+y0ZRpodlrMdNatsBSotW3RjzJ2WJSxvYkBZ0mYY4+qMmQQdMEw7qRszBRXl/a/ckhpcrheaGmqtOiW7VxJ2OqGBiNK3E9aVbTMitToLnhBwfnLDd37nxxEP4nB3OnXj0dY5bVp5HaWMLqG2VRmURyaGaDKvk7h34mJkduIKT1+/cHljiajfTX7SP/0VfPT1yzt+ng883vh9/4efxPf78p/5jp9rsVgsFovF5wYPFhAzJr0lGROZMO8HLtCkI13RbqQmbEZ2EKvdeiXYtkbrjdYE1Agx1IzWOqqtgtp8QnqJqSnr0ZyBnUBNubm5QTbl5nRGspU9a1ilO0eQORFzRBM3cC2NwSUqsVm10Von02hS4W/nc+d0Ph15Fo1Hr32A1jYkNxRHpNHEEDYilZEDkUFkYnaqTomUw5GQ5dqkzv28cM0dacbMylfYTGnnxun2xBtvvIk0pZ/KgenFyJCkETiOlyjdARXEpAoCQLvRe+O0PSLTsYQm0E3R3tDNEIOZZWm7706Qde+l5NGmgYtw8fLe9enkdAiIUO7vV5jcu8q3/2V+9r/9tbz+fH/HT/XqbedP/Cs/jQ//lJ/+jp9rsVgsFovFZz8PFhBb62zbRtcarwGImDg71oW2tdIYRNBMaSejnTuYERJgMCskmt6re4CVcFgUbCvBc2bFHsCL3API5tCDx688IlpCL5HBnpPLHOyRdGuoVleEOSEDTaGFIZHssXOJC/SJdsWOTsT51Hjl5szj287nvfoqrzx+hdYbiCE2mYdjEpZsG6TdEwojnMij28IhfI4kpjOuzxmXu8rA6J1pwh3BlEDE+KKPfCGpilhiZvV5FeYcgHK9OHNcoSlokKJ4JhElrraTcrIOpYVG9UidTiciIRVTA6WyJIgSu58UvVFmKPvuILCns1/uS/d+FvLIoFi8u/ylr/69XMa7U7g9Pje27cHg+cVisVgsFou3xcMuTBKEwPlJo51qVsms4QF5rYVsAOfThprgGUSU7iEyyHAyByIwBa77RMTpm6BdCaHGgHB0q6Lj9L4zqolixAzmdUIKgrKdrKxLA25OnXbuCJX50GzDpJWrUT9251PREOaeZVkqSu8Naw1ryrkrY165vXmVV86vcNPOGEJvG2ZnbvvG6XyDZyd10HpHm+EyyJaoGaKG9Y3H5xtMlLt9B5JXbs/0VoFx+9xJnXzHR7+VfS8tiZrSTDn3DWdw7g2ic7nelyVrOOmUnmSf4HB7Y2Q67orsCR5YQCM5CdyIsXXBNkGaIA2sKy4T92DcT3zUqNK+z8qUyHpva/YuPG6Lvwmf/JCf9Wv41nfY0nWxWCwWi8Xi08mDBYR4MvdZu/EqZDPCr3ANxt2V2MtqFU3UhK61aBeEUEgT+lFc6JE5EAgYWBPUBLLhqSgdUG5vNhAYcQWCSZIT9jEIhWtOXJMxg3GkXs8ju6G25/WtlGXJhKhrIivr4Pn9BRXn5qbz5Hzm0emEPmpst09wg4iGauf25gZH8bHz2uPGqSv4BesCG8hWXZmtn2ltq5C6hEfbhomw+2RrxnlrnNuJjY0v/KIv4Pn9Hdc5sQbWG9qTzTqOYw02OyEhGEFr5SS1Wed8uqG1RuK8ef+0Av5GIEcgHAqDoNF41M6cTFGJMoeaLzIoKr1aJJDTxhRnxgX1qLC5xbvP/VN++M/8VTy9XyL2xWKxWCwW3zd4sIC43ycqJaRObSRGUDvb2moMKcUJnXiUpiE1K0yOSqKeHngm2vKtFOrMyjAYO8x9ohmoCs0aRNLOirigIiSBnmuy5/4yeNLPzLvExyRiMvwIq/M8Ag4Cs4aqEbljOlGqM/J467xye8PN6cRGcvvKifNt53ZrbKdHnNstbWucbm6JKZie2FMZ8Zjt9ITb9z3mxl7jsb7CeXtM62dsu0XaCdk2To82bm9PvHK70QUkk74J/Wz0G+V0OvH6m2+SriQV6ta100xRbUiTur/aOPWN1EbbGlgSDMScJ6/e8J3f8Z0kVRA4s4TWBicTGpCeNQaWhx5DkzknZsrtqdHptHRkBLkLw4Mx1wjTe8a4vuMDZH/9Y3c8feP5O3yWxWKxWCwWnws8OBR9MuWtrOIZSAotG6igZwGHnJU5kOZkKhC1250Ns0qbVpFKjj6E0006nlKZB4B2iO5Yb+zXSRsNNWfOQFIxE7RtxP1O9uS8VZrzyToZVUBcYsfE6L2uWUzJ2AjKscgCrtdB3xTThhDEdXLqDUnl5vGJu/0JYcr18jpmN/RQUp/X54xbLB+jNxtQY1m939CtMivaqdN7BwHr8OSmsTWlMvUEsjoCX/SRz2Nrgh5WrWnKHE4wUFEyKhRu6ET0RHhyOm9cLpNTe8xrj51vfvM76ed2dBQ6GYJYkjaZOdn3YCK0ViF1TZy7684rj27YzLiOK5tZuWO5oAnjfrkwfbbym//YX+HX/FtfDd/2je/1pSwWi8Visfgs4MECwnoiKCZCurOTnHuHE9ieeIMmjXkZ5J7EZkgmmkLkrFwFMQSlNSvRcTYikj0GimOnE6dTRzS4pld+gcPdBMlAUHRCP0E7txIGn89wd8H3gfWGSSPC2cdOTMM2JaUsVEktwfJwmlUgnZoTBtcMugqv3G7cXSePbhvKLdYa1/s7kuQ2hMmkWwmXe4PT+TVO1pm+g0i5Od2cef9rr/Do9sTN2bjZDI0kZxVC/aQ0hGwnYkKzBATHkZ7o3mhHwTH3ge/G+ZyIwj4mp3Pncv+8Cg9Vrs+d863Qb0Asca+QOERQU7rLoftIUo3tdCIIRBrndsbNGZcqWKIl7bQ0EO8lv/Arvp7f87/5cRXy92nk5/72P8Uf/vf/g0/rMReLxWKxWHxu82ABIV3BJ3sG0OimkE5etJKlQ3ChsgeA2APRIM0qLjnj6EBYuQV5YGqoCmfZkAgkSxehwCbgZuwkNpMICAkunjzZhEjwfdDOyunJifAzORzySMEeEFsds86jXEawqTIbZCaGoBLcxeQsys2TG6wZ4417Xs3HqN/Rt8btzZk3nj3HL3dYTiKclEk/PaJ3o9lGb73E2Ocbbm5vefL4zM0Jbm+VfmrIBWSCjKgxr+ZlZYsQXvkMaWBtY2svujPCsEqZvkznxgRI9lC0dcwg9udcLxdON2fS7RBzlyibMdgZZFc0oYnSuhEptAyukahAM4PuJSp3x/2df9gW3zv/9Vf8R+z/2I/5tBUQEcmv+L1/gT/8f/89n5bjLRaLxWKxWLzgwQJiXJxshnIsegOsC+kTMQVtFRCXSuA1OhRStYOUFiISVIJmwu7GDJCcpCeRwrk3QgKPyenUSRFkDHRQO+YdXtmU6YGJ4dKqWDCBLO2FaKNvHdXKN9BZWg0XR6xsVz28BMRWhUnbGj6d6xj0U+MDN2cut43NlI8/39mvd7xyvuVOHGcDr+Tg1m+5ffyYboISnM+3nE8ntq48OgtPXtm46YZ6EhbMHMRUdNtolJ5BmhIBl8uFZo3MwXxhzaqwoUQ04s6JR2Wwas3wEALDM0hLWtPKq/ZE1cgjW+K89UNnEZV2TdC6kK6MfdICYmbZvkoJ4J9f7t6Fx23xEF/0k34p3/zHfzM3m2H6dyZq90i++bvu+MW/87/h//Mf/6ef5itcLBaLxWKxeEkBMacgMQmEOFXqsUgvzYODhEMoaYlaK7efCZllrdqtl4bCkiRwd7qc6O3EhYFlMOcku3JqJ+bI2oXXhLOg2RkeeJTgOGRHzDCECIE0BEFwMKFJQ1Le6ohkCL0ZYx7p0ZaEOMxE3Il0tta4H1fOjzrb7PAYPJOn88Td3PnA4w/iwGW/J9vgdH5U7k031Q0wTXoLHr9yw2k70aWcjqwpp2ZkbqQHPvfSe2yJ+glJ4fz4hv2y41dHT4q60G9u4LrTuiCnXg5L5pXqvZeQ+vGrtzx7/oz3f+AJGVRB1xJphmiDDEY6SqJSRVpO2GWytc64BmbO/bMLG63C7W7WCNNnAt/vJ/1Sftlv+Gf5ZT/5S3l0enlug0dyGd/dPvptf+Kv8Rv++d/8Dl7hYrFYLBaLz3UeXKEkECNJ5rEITe79UjvXFyciyN44nzaagltZqjJBN2Fm5TlI1EJ2U6GbgDlnOwLT0ohhfOLZc25uGiZ2eEMp6V5C6+sgDERrd3/OIEwhHPTFwhkyKpQuopKsCYgMrJWIWRASYWpCzmMHXrltJ3ImYdA349XHZ667o924zh0FHp2fkAmbKafe2ZrRNDndbLRNOHVls0p/bk2xFMwMVYhd8TuvVOmhkJNA6V252wdIOSZtNx240m+M3GtxSCYyFNdAGjyyR3zgfR/gEx9/XnqHBDch5cjdAIIkJQBho+NtMiWwCHwm1sCHst3cwMW5uHPup3f6WVu8TX7Tv/Rb+E3AJ77+tz74vunBH/tLH+Pn/qO/9t25sMVisVgsFgtepoEQSBWaGkSlSsd0UqldfG1HQvUgUKQpmkloEJZk1oJWo1yHxJLQgbVGepCzgSb7J+54/4fODAePUTasllyuyakpbAYkps4kcAKdBllp1BllCysoKoJnjSuZUvkPWSnUMYOMxLrw6JUb5qxRrHF1MAgNUhpPPXnlkfJ0Cj1vkBnMhHY6cd6Mkzq3Nw1R4ebc2Dajm6FW2gIJqXuggVpDO3BKQhJPBU/CB7qduHnUACMksQ6inbFPpGndf0/GmGwqKA1COG297u0UtAlMh0hCSy+hmvRm1Z3IrG4Nhh/p1LGXta5hiCkWidnKgfhM4zocUylty/H9CxL4C9/6dBUPi8VisVgs3nUeLiAMZmjt2sdEXCBrdKmhiAT7M0dvq2ugJrTe3gptq611mBl0Ko8uU8gITJXUciLKHoyL45qIw3XfkUzOmyLSCRm1IM6EsBJyh5Mk9zMxOQoUzeqKJOCJqhLiNULkAXLYzBI1VhVJM8FOjctlYM24bZ19vkHcntlc2AnaoS0QE5rUAt66spnSt86pK6qKWtI1SxgegkwO1ySBLqgkMZxnT5+z3zsfOp3R3hC1KgSGYGdjxo5mjYb11unaK/Mia0FZzk5XrnNnowL5gkM8bo0gsQwSSAWfwZiOStQomAmRybOndzw6b8R0tuXC9BnHv/21f5Wf8EWv8cXvv8Uj+U/+7Le99bNnV+ervu6b38OrWywWi8Vi8bnKgwXEPgOfQQi0VnqGzRpXnxBOChjJuCaRYF0xVZoIKaBWXQIJJ1yOboWQGUyJt473+P03zIDL65fSLDyfbFs5HKUqNpWYteAP3/FRIXWi9W/ORFJJSdLLaUla5UdkZnU7MolZLkh5jAaZNiQSGrSbhmRwvQ6e3NxyaoOZ8MwnLQ5b2jz0Fil0UbbekJaEJl3BRJFUwquIqBCIWboEUSKrg2Jn5fF2RizhBJEDNWMP6HNyOp9K7Ewn98BtIk040chLoigZzvO7e3jcaVbWuLVPHUhmXWtUEF+3+pyihgKizilOZE60CRHBdezv9LO2+NvkX/0Vv/m9voTFYrFYLBaLv4WXjDAd+Q0ujDEwNa6+l7ZBO5IOzUijMgj2gfTGFFDVmuFHSC9dQM36VyCd+jEyczg4qUF/fIZ9sDWp7LrhWCQxKyxOjk1ya8oYgXqiL1yW0gkgo2xh85I86htt68BEEa5zJ2fi4fRNSQelFvaiAiKEQqazHZ2UGwGVpOkJIjHZ0Abb1oHkpA1RRU2RVloPtbKf0rqJJI4e2oz05PaVW5q2o4jpRAaeEwk7chsCPRyWpAtNeh0nJ9lATkI7d54+u/Dk0U1pTlQQasQMVSIBrUyO9KwuhkTpMJrgM3j0aKt7fNOQZeO6WCwWi8VisXgbPFhA+EiSgeqGhDCHg9qR5pyodYLEVTlpuQ+hgmh1A+AY4ZESM0tIiaTJoyNQC+WQ6mC0Frgr2+3GyEFGsu9OEwWRGntK8ISWMC47uYGoEhHsw2k0tAs+nOvdpG2GdYMmqCrWDHYwN6RD4hVE50FEsjWDrqQnDauCgKQ3qRGmKJGyUZ+DgLQScGuCqYIFLtUFQGpUSzLQCTftSI5ugjhoZhVG09h9cLq5OQ47MGmkBOGCapAuzHQenR/zoQ9+gE984g40CalRMSOPIqpyNTKTVNCj64MKkYJG0NS4n5NT62DJjFVBLBaLxWKxWCxezsMFhAdpirnTVPFWFqemlUXgozoM4U7fBG21hE1APFE5tAHtsFXNypLwWWFmIrXwFzewQI9MgunQxIgtuH9+R7u9IaBCJQx8DyKcsTtxH7TWSMBn6SDIQLuwDyckaShI0nsFrnHqxAgMUKvfNQJxp5/Aw4kX1rSW4Eq3oG3K9EQPbUOTo1jJLPepUFQSIUEhozojmgKtIUzc87C3SsSDuDraOyKNx7cnpg+aCfgRyidWx5TKnZASeKAk7hOoxOuIxEVJqsMiJrQXblaSzIwSTTdlXAabKH2z6pRkaUUWi8VisVgsFouX8WABEWNi1msxr4k6IMrYd+xUSdMvdurdy301ESaJZhCZWORbbYjIqDRpq0V9BqhD7mUtGlIBdJf7nc0aocGjJ7eAkmPgVP4EIZWmfGvEUHBgQFejnw05tA9Sy2kSqXGlrrzwGpKtcvA4kiQgEZUK8AqQDGBwbkZY7eJrNrpM8iT03qqgSSFiUvNDQWqJx9MrUI8sNyUB0MqwSI86vglk4p5sZ2HOCQYzHAuIcNqmhJUkPTLL/UoqyXrmzvQsYXTU/RShirJ03ALBQIQZVXhIBqqJZxCA2WFue/20P1uLxWKxWCwWi89CHu5AxKRh7C6IBKGChjBGQKuCYYaQ4iRGhtOIY/Snxo58VmqyqqLSSgcRNU6DZH3fEg+Ymsznh06B2sEneauDMYejpkePo3QWqUFS400xnTES9eo4JOUKVe5EFSSnmaV1yFq8g2ByZElYFQMRx45/GCHG1kC0rGcBVOra0wK8ig0TRZVjiEhKsKxWguZDtF1XKvRThyytQ17KbjaYJQhPhxmcto3IKgasGZpwyYltSrrS+gaRvP76U1595TFJJVYLx8dKIWmQ1bnICdOCJjWCJVk5E2pUcbFcXBeLxWKxWCwWb4OXiKgb1zHrTSPIJhUMZ+U2NDMRVYTKOJhai/J0R+QQ9QJCkPpCRJ21my+CA+4TCcEzSNdjDMmISNIqKA1N7GSM6zhE1xMRIb1GpCIDJGmnGo+ae2BNEdOjW1G78JC1S6+CARFJs6yuQwk4yHF8doOk1SiWKapS+ReuZAQ5Azl0H2aKRxzOR/X7anrE1pUWwt1BorLvRKpAMcFuOtEO1yZNWgh5FF/JhKjfjRREKx8jVTifTtyeb3jj9ac8eXSDWmk0AOLFmJQIikAITau0mRlV3GVdywwqDyPz0/1sLRaLxWKxWCw+C9GHfhj9qC+MSjzekzkCBySFLsaRXYaSmNZOf0Zyt18ZMQHwKfg18T2ONOhDoO21J58BamCatKaIgVgwCWYrwXUKaGuI1DiRpJSaOhKNBPEas+qgLSCdyGTslb4cHhgNUFTKtaj3Rt8EM0UUWq8uhG41/BQ+yZxlvzpflEIgYkhahSxQ41JzOj6d9BqBehFul5mV/zCDnIkZpFa/QEWR7Qi+0xJ0iwitGSFB03KI8gxmeLlVaV3r1o3b8011G1CaCWpBUqNjkdXgQRI1Y+tGV8PMaFJi9DmccUnmWxkZi8VisVgsFovFwzzYgdAm7FchJOii5HBSHWtGSpY7kRoznOlJxCStppMyYIyJnBSZ1AsNxErYnKPEx+UeJFgIMwairRbiBniJtps0pifbuZOjdAfpiYsfoXXVBUgSkXJBUq3FuUgQcyBpzBilkaC0GKqCooglPYURNSLVxdivFxIHFTI77oFHib67lf0rUYVAXJPYQU+C9vrwObxGiA7rWl4UQSK1Vk+t7o2CenUkMqmWzRFyJ2JIfHeXJgUQI3yW7awmY16QzOp8GHCI0/MQVksDpBK4NbRsdyXIDC53k5sbw5pi9uCjsFgsFovFYrFYAC8pIEgnTfAsgS6ZWELXWqzvMTACa53wgAT3ZB4ZCNYM/FhkH4v+nII1QagwuDmibFhnWZ6qGHbYigoQw3FTiECslRA6BSdx9ZJAq2FaDk8RHHatAu1wYELeGp+aXkLiZkfQXBx6Cqlr0lR0a6Xd8IFIEkcHIGfATPQsaFNEBR/xll3r3GtkyboS7iAOUkVQKlgYQquixp0xA8VKH4KU2NyjbF6l7G3VjHmtc8tWxZVkJWqXGHrnfr+ytTMtq7/jEsfnybLN1bKZpVx4cQKPxC+TeKQ86p2UJYJYLBaLxWKxWLychzsQWjkFOZKtNVxrhz8AiWSEl7uRCCZau+ccidHHhJF40Nph52rKPksDIZqYKNmgKcQExBCDEw0dwhiTlGRcJ2FC+GQ71YLbMmFr+DwcigSCOILcjMOaiJS6Fs3ENqsCozXsJNVdmFmDSZloEwQ7duuhtQ3RKmZmBgo0rfGlnEJadTKk1wjR3OdRkJRGhCmE1liRihxZE4546SnmPtFwem/MmFDSB1SE/UWh0A4XK6tCxYdj2hHg9nzD7e0NT58/5X1Pzm8pqM2iLGbV6l7boasARANzrdTsxludDc1VQCwWi8VisVgsXs7Dg++qNDVSBoiXU1GrHX2RhkmlNfuch2FqLbKNw/ZUSrBME6Qp1hrt3GqEp/xGsV47+a0rklmOr1Sgm5lVkeGJaB4758GL3XdRwTZDG5gdIW4V0FBrbi2LVokEEvdAIhA9igaocadW41QSimiS08kM1EC6HXoF6F2wTgnETRm7V8EgiZnQelnVuk9SwY+gOclyowqJspX1AIMYCV6jU+GKepIonuXyNGayX3c0BTvcpAjFw8mE880Nj25u2C9XfEbldmSF9dmLYL+kziegVmJ3USFVsE3wS4XerRyIxWKxWCwWi8Xb4eEciEhUGqbBm5eBCNxuRmsbCJg2RGq236NE0TOD3gxtpREQhJzCzCBzlIMTtQiPyCMlOUsnIA5pmBpjTKwpwZFUPZMmWqNSKsfYkKCi9HMFovleHQE48g1MwCHD0SN3wrUW8FvrEFXoTMpNivA6pipqndZbWdmq0TZDvH4/xDEVTqfGvs9jXIlDeO3kFNq5XJYkBVIwEbRVV6WdDXGrzokHOoUpk64d0iu5QgRaIFFjTVgVbmMObDPshZOTKD6DGZTj01Estc0wE4JkDsrbtZWjFJRoezudGG9e4drxNt/ZJ22xWCwWi8Vi8VnBgwWEKTQrG9KeiohiYozd6U2I6WgXVNtb+QMRNcmkCTmjFueZZILsQgs4nU+VoxBgIiUMzsHYkyZHkvXWIJNugozJ7omg+HCGB00FyxIhi0IKtNMRThGUslpLeyBRdrDaaoc/I4k00kelWKczPQ6Hp9Ij9NMRNDcEtBO+l+2qdnJEpV2r0E/GfqmwOjNBWq335yi9hBk1UpVVn9QYUmJ7YgiznFppZxgxOZ9OTJ9kJmxCc8PnZI7ArNHUiAl7TGQzBsHzccd1Xrk5termqNHKcAo9XKBQrfwKXmhFKvVbWxUZS0S9WCwWi8VisXg7PLxqnI6osLVKnVaRckzSRD0wEVyEMZwmlf/QWjkP+YsJIa2xpumBNiObkOrIYdmqKQRRm+zN8OGVEi1KwxFrFQI3ZrkmCcisRbkpSFPySHqWppgKGYG7k4e1LF4Bb6JGaw2PwPdETNndy5lJ89AoVNCc2iHwRkEGakZOLzG3JOOS5HZkSlhDj1TpGULi7GMiYhV6l0GTGm9qrVXBdYwStYSRgUbjirNFoFYah5TABSwFoioj24Rx9cqwyODJozNvnjpPnz7l0fn9lUVBElHWrJlgvYqmF+Jpw477LfimWEtOvb+zT9pisVgsFovF4rOCl3QgaszIrIbpTYRmQRyOTHJYoUKFt8Vh19pao5kceQdCIDV3H4lrJVyrVMZBlxJTYx1R4XLZ8TGx04agxAhcQXvSsyEb0ANDGdeBTKc9ahUAl0F0q46EB+FHLrRUEJw0qTRqq/yKTDA1xKsC6dbJTYk5mbOSnf3iqCXNlOhGzGB60LsyqXZLjWPFW05UmZXwnMcokodz5wkh3EpwvV453WwVSKe1wHcJugdjDFST1mr8SSTJZohEjYRl1sJflMtwtu3Mo0e3XPdZ4XMzoBkhAJUwHR5VVMxkv+yYNW70VLa1J3+rMFssFovFYrFYLF7Gwy5MJ0HTGOGcrHO3T84Y+KQJqPbSK2QwtZKZVY3pjjWhWSu3H6E6ArNCG+aenE6wv7GjW4dXSmidWb+jW+3yG4r2w/lpZqmzExggJmy3W2UjZFm6Nl7s+FeRIL0KhGZCHtkJU8Dk/9/e2y3JlhxndsvdI3Zm1jmNxg+HGHJGHJPp/R9IN6JmTKQkA0H0z6nK3BHh7rrwDWiuTvdN0wywWGYwA06jq6uq82L7dv++lQRBE0NEWSzIwFsg0aBBeg08kol4KzmbVkVsV2GtGoJMrQaNhKvrtpqgshEJqZdY7jWx1vAUegqvOVFTns/Ft293Vjr9OgETFshxyfIW1WwljOdl2M6owHVALiFmMtYgEqxV25VEnUuJJUvqZCkCmvQ6V2rBmzT6gmAx/fzFP2ybzWaz2Ww2m79+vn7CJPVmvLkR4rRWb+5b78iAfhhp9fbaMIYk7sn5mowRWFs8DuPT7SifgQRK2ZnHmDjgJsQ80RDCOnprfxG0pQbuFbIWbeRZZ0a9NZbH9efBiqRxnUZl1PmOCWpa1as9WSKYKhFORNW8rpdjRwWPrRuRTthEo2zRvpx+b7w+Tnglx0MJE7pVRaoohFweiSxLdniAOymBeLI8adaxe0M1SXECQ0OvrEOw/IQpvEKgL755O4iYl/BOyHDChelBSmU01KTqXsdk+WDOgS+wnqTVlsSBhiFNCFVEF0Od1hWTZOXCDiVD0J2B2Gw2m81ms9n8DL5a42oa9BTG+YFm0m5UvsGriWhFIOJYU1oXWv1FclQeQgQihdfHYi6vUyJTgkS78varByj4FKYnR2q1GdlBZjKzToTCF74mI8b/39pk8Bwn/9e//CsSikedC4lDDvDT8bmYazFfIHREOopdYWkhmqBdoUuFqUPo0esEKZMcycjJ8XYgvTGelXFYVu4HTOtc6xLZmVltYUwRNcSM1tuV9wDVxI5OvyuZFaK+q4KXQ0N6IKZV5brA80B7R7UTmqQIK+BcwXku5qpNy6/fvsXC+O6771DVy4kBuONemxJOQVPprb7vNIgI5uusTMSK/5AP3Gaz2Ww2m83mr5uvvnZu2lnn4vF2MDO5N0FJ5hNcAnwSLsjlcugIYwavcyJLeDPl0GpKygimL1IF7Q1EWBGkOXig3pgRzPeFCdw/3UGMxayKJV989y8/8Nt/+B29H3guHt/c+f39P1fhUghrVWA5NbB+DQcLPB1JcAeltikKtLuSGfQmYMERxoiyW8cK1kyOXn+/3cC1guBzOTTFKF9DWsnsTJXE6o1+CmlZ7gevU6rztWhh0JKjGyED9yBiYreGtM6nfmfkpEsj5c9uDQUSPRT1Cmn7GfhY2GG8PW7cPh+81iCtQuVNheVJRIn4tAx13K1XqPxqZMoUnmNi968rQTabzWaz2Ww2G/iJDYQ40BLJ4KbQIliaRBPOj8GK5Pms0HOOqHC0CkdrmCkaWZZpEVYmw4OPAfOsHETiHNbpreMSrBhgjhOsXKjASuU8B4nxze+/YTGZ82SsxR/+9f/lpkZvgrTGIJiA9V7CuxR666gZnlFuBUmO3ssQLYmKEQl+JnQDh/UMlMZxyNXlCs+VpDR81sM5WkbuTMW0X0K9geCQjpnQD6OrYEL9cw9FWTAD2lUDe1htWVZCLk5/IiivmMzzpGGMNej94O1+oHfDjjv91hATxECPTiLMCMacpE9cHCRhOeGD5bMC3gie1XoVWeVOvX4tm81ms9lsNpvNT/J1kVw63Qwwli7OL44FiBu32x1MEZTzDO49GOciMukd+nHQe8ddrzv+hLUgJ/JWNaJjDOQR3I4Hd+msWEjAwskTZjgik9vtBpm8vd04z8mawVJHWsMTpEpPr/OcJCVp2mhmrCxHg6bgEViDkY7QyIARgapx64pQuQJuRjjVDbuualdPliSfbgcm5VYIM1bUk3igCI3po2plLRBVTFvVqE7nlQsTIUMJqzC0vxatNSIFdS1PRI/ajBwHY07MDPesdqbDCE0eeqPfjLWCVCHmYCzBl+AxERGaCHpvSABRNbIWVwtVV/I1q11qTDLtP+DjttlsNpvNZrP5a+erGwjP4HwfZCT5goYit4ZIED4Rd0QCo7wEpg2JagXy4Sj10K4aHIdirTGB5+uD6SfakrmcH77/whr1CC5UoPmVwfff/8DpzscceATDF+1ocKu3/3/3u28RUzCITOZw1BVLg5TaiCAYSgQMn6yVZWxm8RoTtUQkGDFZPgkm3Q5Sod067WZgit6Ux/2GXk1Kz3PhvhBVFk5kkFlh6psctGhXNVMg5uQBhxn2aLSbYUG1NVn9PNbhuoMCAqMRCywhViAZte3IpAmsnMwri/J8vXh7fCbd+f5P39GiY7MRkaQEITC1NhLaOukLd8cO5XRnXZW2m81ms9lsNpvNT/HVASLXQSC8fxkE4JlYKiucXOBzwjlZr2CdZU+uoLTha/B8fyc9LvGZYzfl/umG3W40MdLrTf+ayZfXOyFOWqsbfpLebuhTMLcSv2ltLt6/+5H/45//Gc+SwZVdORivyfN1cq7JuRxPJaI8DKmCiOFSrUutHdxux5UTgOXCmAnReL6eJcwzIRHu3fh0NLQreiR2rzrXWIHPhUYJ8VDKRn2dLLVOSfBcyFEtr346ZCBauYl2dLo2tAlaPwo+S/IWlojAZBIRSARzjGpvaoJI0ASOo/GbX3+LdePH54vXXJzhqB607IgLlp1Q5eU/lCZ8OXjSj4b2Rsr6D/i4bTabzWaz2Wz+2vnqCZNzsrRqUe/amHfFp9O081qTvM5irrAC6QtMcCCmwExynpBKQ+kPQagqUs/kZsH5BDG4984Yk+PPD913Qaz8Dz6c49HwDEB4++aN/9b+K7GEKYvH/SB7cLy1+tpNqzJVg5HlZbBw1qja18AqHH6/kwnDHV8ToVXlqgJU1kNnDQTnnLQWeDZ6GAh1HjWDsdY1AChreRmyFc6PwA6w3vDzRK+mJtVyVWQo2qu5KT3rbAmQnkw/kag2p26NWJMPXxiCRP1u2nEjxyJZZAPVqIFtOdbhjGAprEh8DPLLFz5/8zu4C6QikjQxvMOPP4z/gI/bZrPZbDabzeavna+L5EQ5GuRDGcuhJdGMNR264O+DkUG/9RLFJWDKcRiijeVORLUcnefJ8sBu9ZCtpiWKO4TnmuSZ3D71CgznG6mBWiPWINNZGOc5CIF/+b//hf/tn/5XHJAQfDkmQteqYM3loFL9RQJm1K3/SkQaIQvTaoEKBMXBG+frhE9lbyZrG6GAR9CsgQahk6AhKlcLkwIV2AahdSrv4cnjfiAGoULrDT8X2pyUEtQlVIZiJHq0ylDEQo8GSxBrhK8SxPU7vJ7X9wy66uwpJcCTuzVMgnO+UEvs+v2ynB9/+AOP9g23b/6O0LJVZ6ubpWDx8RH86u3tF/2gbTabzWaz2Wz+NviJ7k6HTMQaR2vcrMOsKlRfC2+CZLKeAwnwpzN/WMz3rIEioh6UF4xz8fHlxfPHgYmiDqqNW2vc7c4Zg5gLazeSIENZc9Jbw44aDHori/Lf/+73RApW5jZ8XMNLQmZeZ0vKIkEFk8pmtMNIEW56INEZ00mvmlRpyXE3uhraBDCUsl/3VmZnCUVOxd3xddXRXvODNyc0UBGa1q/19FFmbAmkdZBGpjFSED1IqxpVORrneRIJ2kuSJ2qkJkkQA+YM2uOGHh3TjmllTUSsgu7q/ObvfoMTvD/fOawqan/88Tvebr/GuFWGRRR3WDM458JdIZLvv7z/0p+1zWaz2Ww2m83fAF/PQISgmajUJoEVSArzfRLTq340hLWCcwyyBfRkrUlk4JmM98XrYxLzCgO/kuefTiLAfTHmwn0SZ3J+DDIn0QOVoDdlZf0zTRsRwT//9/9RRmgWSbACZiZz1Rt1DydXEmOxngM8MUkyAl2JUQ6G1h2VRBIiKkuxziB8saI8CdqN0Ko6lVYP69YaJh1M6u/T8ivUV06CKC9EU45+Q8RQGtYC0UBCMIf5MWEmhxnhi+P2xhijWqVU6l/AnKxnXF81iBEgQn905Gb0m6EqSBPMGt9+/hbTzvNjEBjWG2+fv8XUaHetE7DleARrJb6S8TGQlKtta7PZbDabzWaz+TpfHSBaS9Llsj/XW2sk0S7QjAxYI8gp4AYIquWNUMl6E5/Jek1iDBrCDEebMF+LXJAkwxNJJZfwfMLHl8nTB0+fZJb12edijcX9ODjEaGJoSrUZkWQsVnq9uY+FZ5361BVPkNSVUbN6zJdU+sPqz45G78b9zRjnYH0ZqEIuJz2JhN47qXG5FqquFr2GopnoLHO1YpCKmoA4kU7gQKPdWknusmO90Y4bIge9H8ia3D/dMSBRTA1BmJmkA5poJH4uzterwuCiBIFLEioVdPf62Wc404MQgaxK3nBnLqdGrSQ8yMxqq9otTJvNZrPZbDabn8FXMxApEJJ0gy7K6YGsABPUkoGD1P/GLteDaHkPVj2YplVwOLIeWCOCMJBYjA5KR0iWB0frpCyOfhDpuAe9wwonlvOHP37HP/3DP9TDuwU+naN1AEIDaUomNL3hHpgpR2q1IFGW6rWSzIl0JRbQtZwOAfTGIVquCwETZflirIAOqsI5Bu6Lh92IrFMjQZDlQJbbQqD1jrRqZkpPVi58OTkXi8WtWZmmA3JGRSnC8XNg/Ua0STbl/uhkRrksujKfowzeM7j3GqIiE9FEW6M15XydPN8nRzOQINOxIYhChOLN0QPIxJoinvx56bHZbDabzWaz2XyNr2cgRNGjukWnO4iQVluGNCMkyfLMVS7AF7kWay7mdHyCzyBjYU0YcxJRU0t7NCKTcU7y5SV6U/ApVQ8bye0KJqPwmgMaOIvwBBdUrAaBDtoNA5o1QoykHphTIbm0y9TDtM9AJvjHSc5ZLUpaG4XjEFKSnKtEdg54sOZgzYEeYMdR4jfppIDjuFQblTZF2/VnK8GVXCBnoJm1wUlIn+SYQJKyCIW5FtIbmDJWlldCDZWGmTDnCSoc906u4Lt/f/KcjhyKmCICv/1Pv+X99eS7P/6J8IBL/Jcp+BmwJueXD2KUgE/CqrlK/Bf+qG02m81ms9ls/hb46gaii3DOZOGoBKlG+iJJlODWG68R+ArEygFxVRgRLsw5Ua0H/Yik90azOrVZayIpKEaY4jhvx4MgiQW9wQDecNKEH15f+Mf/9DvMWj2YRyJdyocQvapMASO4HUK0jngQIVgTcpb/QcXAEk9HesMoMV5KIq0alLIJMZxIIebiODrLHZGOrsnNjDmTMGiH1c+di5QKXRPXiZUq85w18ES1OZXhOklpqIJp0h53zjnRqIHHZaBe3os6uFqEKNKMDoyV5ahTZa5EXjWIiAif3j7Rjxvva/J2Lo6uWIJIMD2qOYrGejnts7AIUoSIvYLYbDabzWaz2fw0Xx0gIh0w5px105MBESx3fJWTQboiWcIzVcHU6uY+HOtyZQGSql2qMydJpbmQplgKXZW4H0gEokK79iIznecIRgatG6qNCCdIEuNAaQk0qXpWr+9PDJoJLjXMEGXEJvO6uBKcwFSxBoFUQNz+pwdpUXwtUMEvx0OkY70RwyFreFCrxlcxI3LWjxpSmYsE64a1YI1AojIfaG1N2hXEpsGhvQYMBMOIXsmJ9JM0o5uyqC+6xkQAVFmZzLMC7fVHypiLZsbLneHJQ5P+pjTryKx/62ss1iuxG+DgskMQm81ms9lsNpuf5qsDBCqoBK/T/5JjyFXXQB/jxf04sG64BxnU63MRMhO1SjeoGSEBVl1CYleQWowWlbGIEfSuNaDMRO43kER88Zzwpz/8gf/2T/+ViEm4lMPBnVdWOPlmhl7nPilXDPkaaHzVBiUy0FSSqmGV1ljhCHWahRoWNUxEXMHkFdhRX1us1YCT1V4EyVyLm5RULgGyEe7g9fMSYCqk1qARonTRGn/USsAXC83ArIEexJyVF0GQSFYm4l4bBymzdtOGx6TuwcrmHZ6sFaQ7a8LzfHG7T2690R8NGdB7MM/JcT9IT8QFd2cFCLuFabPZbDabzWbz03x9AxFVKXo7lPB60BdJhMa3nz/jAWpJ0PEZKIInZAYmIK0szk0NNHFPzlgc1tBWFaliCe6kCxnCeC4yB/dPt7rz//jCf/mH/8w4A1N4TcdU8TmR3viIRetJ/kUUp0gKkldwupUvIT0JEcIXzRSJqnHlckV4LCIoO7QJGoKlkSHMNTFryEzazXANYjmNa0thhqhiknWuxFXFKoE0wVMJTWIGdrS/DDepilkn3Bl4VbW600zIUHwOfC7SlEFi1LAm19dPKfv3nIs5gzEDj6R/+hX//q//Qrcbv/nt73hF1oZhQXblnFFbnKfTWyMTnuf5H/Bx22w2m81ms9n8tfP1DcQKemv0T43nKxEvg3LTJBzElBFJa7CiHtBzBkp5FBLB7M+VpqCSvN5fhNd2wVJ5e9wRFULiajxazOGslrxeLw5p+BRWOC5KAufrxBBarwFgrUCDykVIoM3K5RDQWrkSYtVmJLIGGVent06upJmyYpEjiTXLAUFDRPBI8gxSJ0cv/0O/K016DQiila2oPtkyYIsgKqzleAiRJdVTa7UhsUBUEamgdKYSc5EWjOW4Vl5DVHBJWgghiYvX7yqS1pWVxjkmETVwLHeec+IowxdfzpN2Dj5c+c3nOypZw89wNILM+u9NFckdot5sNpvNZrPZ/DRfHyCymn3Sk+NmKOBzMOfCWicWmAnD68G0aWUa1I0ExEpypqpVFxrCrd/5/v2D9/NHHv3g42UEi7EG7WisSJhCX4F78o+//y9ECELVlTZTBAFTyIpsz3ORIXQTWldiOBmJiRHicP29KlXtSgpzOIZhvYLOGKyclFY6WT6xZnV2lcoawf1ePozWjJmLWEk7lJAgMohIkitDkSBoDVaRxEpuR2mrNaK+HxNeY9UJlgiuyvCJfHREgmbCrXcyjIzabkRUk1Nrhq7gXMEYXidMETzX4nVOnuci+YIeb9zvDx63qrGNFYQkTZKuVXsbIRy2Q9SbzWaz2Ww2m5/mqwPEnF4nStfWwQ6htc65AmtaWYcmPLTXg/xahEDvSpNWw8f1Nl4VUiqH8NtvP/FrHjxH8P46iTUZc/D0E5FqbToO4/d/9/e4O2rtOk8C0eR4HJxnVb8qQvoVlL4s0BmAJ7TEstVmgkSv0LFI0qXhPuntVhGIVEypSlcBbeVf6KaIGOsKGluUk0FQkHJb/HkTIZGgNUREJpkJl8SuWpNAG+SsQDlyo6uzMmk3xQNux8Gf/vhCutK78nZvqNWmJSLxcFSFNRe5Sly3fPGakxnB61w8z0lq5/sff6DfPiPaOF8n3K6GrISF0Q9FFM5z0I49QGw2m81ms9lsfpqvDhB2A0z+8rbbTWjWuH+Wy0oNvTe0JYfB8wlzOSJCM8Wssg/NDDPqjEcUrqajuwmPxyd83onpSKuNwIpAUM6PyfHJKlNhhlB1q5FZoeWmlQO4WqC4sg/XDRUpAp64V7YAKjcgwjWMKB5XrkMa2hv5HPS70awxppNRWxa5VduUR+0YVAVpRoYTV3vUn2VsoVS0AsUzaQgqicQl28MQqfWJmbHc4S/hb6M35TmDFYmpcxzX5iEVBNZylsNcdSJ29MZrLc7pJaZD6PcH7+9/5Mv5I3o7eD8VpGFdK6Sdk+bKcW0hNpvNZrPZbDabn8PXPRDdUNEKBB+KBqzpdBPWSlQEU4UIWjPe3oQICKeyDwKIYFbNSYiiZrURWIKJ4h74EkT0ClsrpkZ64u50s7JESz24hwhzRZ0wmRAE2gRNIBSl/gwtUdryMimoSQnlHNCqTlUxwv9sgR6g0LVXI5Ik/daJsUhx9JDrjMhxaSQ1NNTfm9cQUjkLqFYmVeHWGu5OKEQs4gTVWkC4Vw5Br3A3WpuNt/vB3YXnPMkAX4mGIl5nUQHMcJ5z4FENTJaCkDQTpijIgaBoJunOawy0wQ1hikA4HwgenW6QvjcQm81ms9lsNpuf5qsDRCasUWlkVcNEQRUPoYkSBhBlca4joRLHNcFESIH8c+NQBqaCZAWIsxmk4llOg+zUQ7oJWgJllAZUtmK6M1eg2mlqJIpJvZTvWkPJigpIq9XwkJkEtXFIBUknQ/DlmFZGA9ProT9hVi5BtBEx60G/1fcSa6HaUKk8RKzJ8bnRVFG/sgR/bnW6XBdxhaTFFPwaOKgweRnjAs+FUluXGogc0ajBCbu8fFrbFHHwGmSSYMzF86yK2jUnsSqXUVJspfcbcU78NvlhDuAzej+QQ1E1Tg+6Ja/h9N5/yc/ZZrPZbDabzeZvhK/XuK6qN2UmZhUcvjcjopp7PJPA6218ar3dl6pFTRXUFBWrt/QpqCY+a9ugFmgIQWNq0o6GWmPlwjNIUaQJK52b1GYhswLdkPQu9N5ALslb6a3JqOrUyOR6qQ8ZdUrUytPgY6F6J+zKYmfiA0SpTULUVsOH046OhJMq5ZK4hgFrVidJ1JBCCp55Bc2d1EC71ZzgAIq2VqdMGQwXVJx0JwRWlKNBXOhdr80ELHe0GT7z8vjVGZighFfT04qF+2TNxRyBptAb9Psbz/cfsfUNmkKcwdDEcqFmNFXCYUkSc/7Sn7XNZrPZbDabzd8AXx8guHwDKvUWPoTnqorRoxmqJYdbi5JGqEAIuZy4jNTao057Sp0MFqSWkC01kEjsrld4WuliuJQIzkTJLPFcegWStSUp4OmsBaZVX0qCWVaJktQKIy8DdbPKBZxfRg0cKLJgaWBStuzUCmiHKU49TIsZHgmZVQWbZdVWA2uCipScruYHiBpU0ut7FLK+v+m0pn8ZrsjyWKhX5kFcr8pcY47BsqqeVa3MRa6sQWXBOf2yWUO7Nj2vEfzwMa6Gpho+MoPj8Znz40dMnER5vV4scfqowHhX5WXC57c7ix2E2Gw2m81ms9n8NF8dILRDekPJ6/V84A7aFBHFTJheD/cSgRzgUQ/vTSC4/v8YrStwpZepIHFKkAlHF57nQLvQTXlYx2cCC5MG6bw86s1/HmhTRsLzDHqDZiAEYllnSVHNT9KF82MhgPVGILRuSEJIIAle5jsMrQam3up7c6lhZTgopHsFry/tdM4kWlzbh6txKSvnQIK4EjMJVY7WrsamRCJJTZoYOSdJyeUsG6qTPAwn8BBmBLlgjlknWWqsEZxrlqBPjFtvPM+F0ol0TOAMmOs6n0J5/+5P2ONXSFOWBCDkrXF6As755eTxq8cv/VnbbDabzWaz2fwN8BMeCEgqV5BjYd3K8OxBWBBU41EapCralW7A5SuQTNIr1Cx2ZSEqyozWozchZYAWBDQQMQ6DkDrRoSLQSDraGzMXFq1EdZmMc6A3o3VjzsUhikbJ3ZBArP6ZknC72/U9CKiSEQz3GnY8CBFaJNP9GjIU8SRX1SolQbsqnGJdFm0RUgLwEtCpEtPr6y2wDnYvEZ54kNLIFfRbY62GR7BiIioMd1BD0nBfRCrgrNcEU1SS9GANx0dwPDpm0Lpw642U4FyD6c5aC/cF0vl4/zc+98/QKpsx1+DL+eLT44ELvH/8yI/+/KU/a5vNZrPZbDabvwG+7oEYgZJ4KB4LX0BAN6uTHcnrwb8epK1VbWrAJWxTYixWeJ3WmOASVIlS1aeGwPyYRArSHMvE6UgGmcnjrnz//eDoxhJFompSBUdMESpA/fo4sdaRrCD3ClBpHE3xuf5ibzZVEMGz0huIXadaq8zTQOMKd89yJrgHJlegOaSOfY6qbRUVXK2qUL3Oi0LlClAL8zWJK0gOiSK81oKuSOm8SVV8Drx+MLoYKQ7qiNTv0V/BucofoQGhybkWi9qkIEmiPM/F83xWhiLg1u6cCSKLcTrMyYqJrxNl0h/Gc3xw+PGLftA2m81ms9lsNn8bfL2FKbze1k+HgBmVGXjOiVrS+4Fp0rUR6awIzIRcVxuSRInVVFgeaAQnV+VpAhgxg34cOAszYzwH94f9JQD9fi5MO4ijCdYFpLYFRHI0ZZ1OJFhPHIf8c71sopKsFbgIh2o5IiQxg1g1+5QXAUyCHIC2qmyVa8siZeSW1ogMMh3BSg4nCekAhAnuTlpi9Kpl1bjcDxX0XrJovWpqRY2WldVYJ5iVx2HkgivzkEFlNKagGaQHEhMZQYiyqDMlRZEMFMHXqLMxUVKT4Mnz/d/oj9+zxuBognvwej55u32LB7x2iHqz2Ww2m81m8zP4ukiuNXyUJ6G1A+ZEA9phvH88IRNvjWRx741IZTkQWlWpXbGjVW2rVxhYLwncGk5r0ETxNdFutObI2wGhxOtEmlUNqwYZ1SwUnng6KYYm6K2xvk+WTlYP7nqQmpfUTcgU3CuT4Quk1X/iCi3nyqvBNYlYVQ9LVgA7QTJpdomjsyR2khXujkg0F0jiUm4Jaa22EFGeDJUKUQ93Ho8bGpNxOp4VAD/uDWIiVr+XOvuKslOH1kASV9MTQNYGiENYs8zUnpUBMU1aU45WGRJRYASP41ven39A7A27vfH+ekc8mBh//O4dS5iyQ9SbzWaz2Ww2m5/mqwOEe0Ak9hBUwY5GjKocbWIQypyONOGcjmbZnkWqISgVFkmNGI5zyeeoViQfSdhizuDROrhW9gDHc7HWIkOxZiQwp9O6EqH0R+NxO3i9n+hDUe/46UwW2q3qZYNyTxyJqCI9oVUrlIkgoYQmpsIaggeYGCFxnWgFGeWJCFmQRkqiolXeKl5nXSJMX2CCWef5nMy5eDxu3FuvcPd11tTMkJV0U9QU8ygHRMLHc3G/Kx6Oooyr0SnCSUl8BK5VJdtMuUniQZ00RfDySRflrb+xNAgmcwTowVyD/voTt9b5fDs4v3wwxmDMk7f7r7npHiA2m81ms9lsNj/N1zcQonjzaioiMIRs0E4qe7CSmQkL5BA8nAytL9zh0Ea/NgU5HRWhNUMAR5hjsmbVxX68Jp+aIaIV2j4Un4FJx1Bec/J6jqp3bYqQeFeWV+agghFRgWmtQWVFnSZpq8B1eiKtqliP3hGBsZzlyf2uDDdyQKSDgQSgwRpAF1Tr3EpIUgJxYU5H1Wit8VoDX4l40KwByliT8OTQg5iL2ZJ2h7mqalXmRHqnHcH9aniCkvihwsccSNQmw3picuVPAjyTQ7RcGa2Rbzf0DFTr6iwR8BeiN47+wOcPPMc3PG6fajhhAoO17rVh2mw2m81ms9lsfgL92l8UVcZrcX5xsmIQoIvsgXRwqS0CC3w5/XZwezQenzq3e0Mpg/Tpk9YbhzVUA5UaIcAARQ+hq5ChlW1owXHr9ONG3JLnnPTeuPWGyg280Y4bP3x/kpG8xuTdT1pTbr3jkUxPPOvNfA0WVV9aZ03BvL4DbUYzIVcgCO2o+tPXcCKVc9b9kzVFe5mi1zmQVeFl60q71QnRoY17P7jfb2gXtEPvRjbhxagBK61OvUbV0obWFoeArr1C1McduynDB9/85hvGWKxRwrjlCzXBHs6vfmM8unCIkGtdP12rTESrrQfHjdY6nz7/juUv1BdrDo7jgXnn0+NXzPkipv+yn7TNZrPZbDabzd8EXx0gaIP724Mf//g967nwlfhUxIzsQns07t808jp/UclL5uYESYoT6gQwxiqfQSStN0Q6964cmvR2wEPRnmgIAIca2pM1vYaVdPrNeNyNW1P+/ve/IVYwfNWmIoSIYM5FeJCStGZljL6alspO7UQI/prEnMzpFRa/fhMzFsudHz6+sDi5tVbtUgK4o4Addp0xldcBF5odqAghE7sr/WhkBOdYJYILiBTGcs4JhxljeQ06ASvgy8dJqpAapARv9xtjDd5+dae9adm6WyMkWAlzLR4P5XEYrSnrXJDJvRt36xyt8al1miqmd0IaLifuzu1Q7OisMfh077VR2mw2m81ms9lsfoKvnjC9PhLrzm/+8bfkWvz43ZPb5056cj86oo7RsFUOhNfT6Q20K+F18mQYRFz2aPh4n9w96a2x3KqyFK+a1KBOg0R5MRCUO5CSRFMkBQFWOP/9f/8fqCQjIbRVWHklksHRFc0Am6g2PBKjBgxPOLIzcXw4po3A8FwVJpiCmfHbt4MZwcd4IqNxf7xB63USxSINwhUxBRNCEz0MpCMSNE1WCjRBVvJ6LVImj6Mh1phZZ1SoMJ/lgXi0GysgFWIlxMBUaffG68uqLUgKa1VjleolwjO4dePtfqBZLohDla5W3x/wOheqD9Y4afdfIRiH3ZgzkH7Qdoh6s9lsNpvNZvMz+OoA8XG+uEfnaEpq5y0ND2ExyDSezyTjiaRwpHHTksyZKrIWKQsnaFySOTVWKE4SWadMt0+N1zMRUWIuOAKRjngZmEOSvKzO2gAmnz51nq9JutCp6tSVQe8dk6SZ0lSrrSkDFa3sw3RywYrKSYDVVmJOliwOShhn4tAOzunc3oyYwvRBp86SiGTNqoB9nQs5hE+PjjuEL6QJ3QQF3t9PEKN3xVKRECJryxHe0AZiwUint6piZQXT4X50xpeB9ODowhShKfSj475wYOKogR7KDcqc3Y1S8wn3t8a/Z/KJbwn9X/ju3/5PNKxyGUfD2mdWTGLPD5vNZrPZbDabn8FXB4iexpqLmJBC+QwSpHVmCCsmj3vnkIYv5/U+iUw+H4YdB7GclY4QyJm4Qm8V7l0xCautRQPOL686FToF7UJrB7c4kAavTGI6051Hb3zzu9/w/H/+CJr4eNGtoYehUvK4TIFMDGWuxXEoIoZjmBpCMFY94MdyPs5JN+O49XItpJI4Nxrr4yRb0qUTQ5kACcMh1kRDWMOZdhArWLHoKEgjMT7dG+N8Mk2Y7xO7NfyKGwSLXAKrBii9C5ZC0jGcNYPbo1XI+1NDPmaZr19QTrnyXXgGerVDmRmHe9XSAl0VVaApx/0bUhW7GU1uwCQlQUBy/aIftM1ms9lsNpvN3waSuV89bzabzWaz2Ww2m5/H10PUm81ms9lsNpvNZvM/sQeIzWaz2Ww2m81m87PZA8Rms9lsNpvNZrP52ewBYrPZbDabzWaz2fxs9gCx2Ww2m81ms9lsfjZ7gNhsNpvNZrPZbDY/m/8PLKtgryZomUkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1008x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show some random images to verify that everything is ok\n",
    "image_ids = np.random.choice(dataset_train.image_ids, 3)\n",
    "for image_id in image_ids:\n",
    "    image = dataset_train.load_image(image_id)\n",
    "    mask, class_ids = dataset_train.load_mask(image_id)\n",
    "    visualize.display_top_masks(image, mask, class_ids, dataset_train.class_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b5a63ced",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Starting at epoch 0. LR=0.001\n",
      "\n",
      "Checkpoint Path: //logdir//train/mask_rcnn_moles_{epoch:04d}.h5\n",
      "Selecting layers to train\n",
      "fpn_c5p5               (Conv2D)\n",
      "fpn_c4p4               (Conv2D)\n",
      "fpn_c3p3               (Conv2D)\n",
      "fpn_c2p2               (Conv2D)\n",
      "fpn_p5                 (Conv2D)\n",
      "fpn_p2                 (Conv2D)\n",
      "fpn_p3                 (Conv2D)\n",
      "fpn_p4                 (Conv2D)\n",
      "In model:  rpn_model\n",
      "    rpn_conv_shared        (Conv2D)\n",
      "    rpn_class_raw          (Conv2D)\n",
      "    rpn_bbox_pred          (Conv2D)\n",
      "mrcnn_mask_conv1       (TimeDistributed)\n",
      "mrcnn_mask_bn1         (TimeDistributed)\n",
      "mrcnn_mask_conv2       (TimeDistributed)\n",
      "mrcnn_mask_bn2         (TimeDistributed)\n",
      "mrcnn_class_conv1      (TimeDistributed)\n",
      "mrcnn_class_bn1        (TimeDistributed)\n",
      "mrcnn_mask_conv3       (TimeDistributed)\n",
      "mrcnn_mask_bn3         (TimeDistributed)\n",
      "mrcnn_class_conv2      (TimeDistributed)\n",
      "mrcnn_class_bn2        (TimeDistributed)\n",
      "mrcnn_mask_conv4       (TimeDistributed)\n",
      "mrcnn_mask_bn4         (TimeDistributed)\n",
      "mrcnn_bbox_fc          (TimeDistributed)\n",
      "mrcnn_mask_deconv      (TimeDistributed)\n",
      "mrcnn_class_logits     (TimeDistributed)\n",
      "mrcnn_mask             (TimeDistributed)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-11 13:59:44.554182: I tensorflow/core/profiler/lib/profiler_session.cc:159] Profiler session started.\n",
      "/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/framework/indexed_slices.py:434: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory.\n",
      "  \"Converting sparse IndexedSlices to a dense Tensor of unknown shape. \"\n",
      "/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/engine/training_generator.py:49: UserWarning: Using a generator with `use_multiprocessing=True` and multiple workers may duplicate your data. Please consider using the `keras.utils.Sequence class.\n",
      "  UserWarning('Using a generator with `use_multiprocessing=True`'\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Model failed to serialize as JSON. Ignoring... can't pickle _thread.RLock objects\n",
      "Epoch 1/15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 55, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[250, 217, 244],\n",
      "        [254, 218, 244],\n",
      "        [252, 221, 244],\n",
      "        ...,\n",
      "        [253, 204, 222],\n",
      "        [250, 204, 219],\n",
      "        [249, 200, 214]],\n",
      "\n",
      "       [[249, 212, 239],\n",
      "        [247, 216, 244],\n",
      "        [251, 218, 244],\n",
      "        ...,\n",
      "        [252, 206, 225],\n",
      "        [250, 199, 218],\n",
      "        [243, 195, 215]],\n",
      "\n",
      "       [[241, 210, 242],\n",
      "        [254, 217, 244],\n",
      "        [253, 219, 247],\n",
      "        ...,\n",
      "        [253, 206, 224],\n",
      "        [250, 197, 219],\n",
      "        [242, 197, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[230, 196, 211],\n",
      "        [236, 199, 212],\n",
      "        [237, 200, 216],\n",
      "        ...,\n",
      "        [236, 201, 216],\n",
      "        [237, 200, 214],\n",
      "        [234, 198, 209]],\n",
      "\n",
      "       [[233, 202, 212],\n",
      "        [232, 195, 210],\n",
      "        [235, 198, 213],\n",
      "        ...,\n",
      "        [241, 203, 219],\n",
      "        [241, 203, 215],\n",
      "        [220, 189, 204]],\n",
      "\n",
      "       [[228, 198, 208],\n",
      "        [229, 198, 208],\n",
      "        [238, 198, 215],\n",
      "        ...,\n",
      "        [233, 199, 212],\n",
      "        [235, 200, 214],\n",
      "        [235, 195, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d690>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 51, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[162, 156, 191],\n",
      "        [163, 155, 191],\n",
      "        [159, 152, 192],\n",
      "        ...,\n",
      "        [126, 128, 166],\n",
      "        [116, 118, 155],\n",
      "        [102, 107, 142]],\n",
      "\n",
      "       [[152, 147, 186],\n",
      "        [160, 154, 188],\n",
      "        [165, 158, 194],\n",
      "        ...,\n",
      "        [123, 123, 164],\n",
      "        [122, 124, 160],\n",
      "        [109, 112, 149]],\n",
      "\n",
      "       [[150, 148, 187],\n",
      "        [153, 151, 187],\n",
      "        [157, 150, 187],\n",
      "        ...,\n",
      "        [136, 135, 174],\n",
      "        [131, 128, 167],\n",
      "        [114, 117, 156]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[104, 103, 138],\n",
      "        [103, 108, 142],\n",
      "        [122, 120, 155],\n",
      "        ...,\n",
      "        [ 72,  79, 112],\n",
      "        [ 68,  70, 100],\n",
      "        [ 56,  61,  83]],\n",
      "\n",
      "       [[ 97,  97, 132],\n",
      "        [108, 111, 144],\n",
      "        [109, 111, 145],\n",
      "        ...,\n",
      "        [ 66,  74, 104],\n",
      "        [ 60,  64,  89],\n",
      "        [ 53,  53,  77]],\n",
      "\n",
      "       [[ 92,  93, 124],\n",
      "        [102, 103, 138],\n",
      "        [116, 116, 147],\n",
      "        ...,\n",
      "        [ 64,  69,  97],\n",
      "        [ 51,  56,  83],\n",
      "        [ 46,  44,  58]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12910>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 55, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[250, 217, 244],\n",
      "        [254, 218, 244],\n",
      "        [252, 221, 244],\n",
      "        ...,\n",
      "        [253, 204, 222],\n",
      "        [250, 204, 219],\n",
      "        [249, 200, 214]],\n",
      "\n",
      "       [[249, 212, 239],\n",
      "        [247, 216, 244],\n",
      "        [251, 218, 244],\n",
      "        ...,\n",
      "        [252, 206, 225],\n",
      "        [250, 199, 218],\n",
      "        [243, 195, 215]],\n",
      "\n",
      "       [[241, 210, 242],\n",
      "        [254, 217, 244],\n",
      "        [253, 219, 247],\n",
      "        ...,\n",
      "        [253, 206, 224],\n",
      "        [250, 197, 219],\n",
      "        [242, 197, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[230, 196, 211],\n",
      "        [236, 199, 212],\n",
      "        [237, 200, 216],\n",
      "        ...,\n",
      "        [236, 201, 216],\n",
      "        [237, 200, 214],\n",
      "        [234, 198, 209]],\n",
      "\n",
      "       [[233, 202, 212],\n",
      "        [232, 195, 210],\n",
      "        [235, 198, 213],\n",
      "        ...,\n",
      "        [241, 203, 219],\n",
      "        [241, 203, 215],\n",
      "        [220, 189, 204]],\n",
      "\n",
      "       [[228, 198, 208],\n",
      "        [229, 198, 208],\n",
      "        [238, 198, 215],\n",
      "        ...,\n",
      "        [233, 199, 212],\n",
      "        [235, 200, 214],\n",
      "        [235, 195, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d690>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 51, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[162, 156, 191],\n",
      "        [163, 155, 191],\n",
      "        [159, 152, 192],\n",
      "        ...,\n",
      "        [126, 128, 166],\n",
      "        [116, 118, 155],\n",
      "        [102, 107, 142]],\n",
      "\n",
      "       [[152, 147, 186],\n",
      "        [160, 154, 188],\n",
      "        [165, 158, 194],\n",
      "        ...,\n",
      "        [123, 123, 164],\n",
      "        [122, 124, 160],\n",
      "        [109, 112, 149]],\n",
      "\n",
      "       [[150, 148, 187],\n",
      "        [153, 151, 187],\n",
      "        [157, 150, 187],\n",
      "        ...,\n",
      "        [136, 135, 174],\n",
      "        [131, 128, 167],\n",
      "        [114, 117, 156]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[104, 103, 138],\n",
      "        [103, 108, 142],\n",
      "        [122, 120, 155],\n",
      "        ...,\n",
      "        [ 72,  79, 112],\n",
      "        [ 68,  70, 100],\n",
      "        [ 56,  61,  83]],\n",
      "\n",
      "       [[ 97,  97, 132],\n",
      "        [108, 111, 144],\n",
      "        [109, 111, 145],\n",
      "        ...,\n",
      "        [ 66,  74, 104],\n",
      "        [ 60,  64,  89],\n",
      "        [ 53,  53,  77]],\n",
      "\n",
      "       [[ 92,  93, 124],\n",
      "        [102, 103, 138],\n",
      "        [116, 116, 147],\n",
      "        ...,\n",
      "        [ 64,  69,  97],\n",
      "        [ 51,  56,  83],\n",
      "        [ 46,  44,  58]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12910>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 55, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[250, 217, 244],\n",
      "        [254, 218, 244],\n",
      "        [252, 221, 244],\n",
      "        ...,\n",
      "        [253, 204, 222],\n",
      "        [250, 204, 219],\n",
      "        [249, 200, 214]],\n",
      "\n",
      "       [[249, 212, 239],\n",
      "        [247, 216, 244],\n",
      "        [251, 218, 244],\n",
      "        ...,\n",
      "        [252, 206, 225],\n",
      "        [250, 199, 218],\n",
      "        [243, 195, 215]],\n",
      "\n",
      "       [[241, 210, 242],\n",
      "        [254, 217, 244],\n",
      "        [253, 219, 247],\n",
      "        ...,\n",
      "        [253, 206, 224],\n",
      "        [250, 197, 219],\n",
      "        [242, 197, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[230, 196, 211],\n",
      "        [236, 199, 212],\n",
      "        [237, 200, 216],\n",
      "        ...,\n",
      "        [236, 201, 216],\n",
      "        [237, 200, 214],\n",
      "        [234, 198, 209]],\n",
      "\n",
      "       [[233, 202, 212],\n",
      "        [232, 195, 210],\n",
      "        [235, 198, 213],\n",
      "        ...,\n",
      "        [241, 203, 219],\n",
      "        [241, 203, 215],\n",
      "        [220, 189, 204]],\n",
      "\n",
      "       [[228, 198, 208],\n",
      "        [229, 198, 208],\n",
      "        [238, 198, 215],\n",
      "        ...,\n",
      "        [233, 199, 212],\n",
      "        [235, 200, 214],\n",
      "        [235, 195, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d690>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 51, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[162, 156, 191],\n",
      "        [163, 155, 191],\n",
      "        [159, 152, 192],\n",
      "        ...,\n",
      "        [126, 128, 166],\n",
      "        [116, 118, 155],\n",
      "        [102, 107, 142]],\n",
      "\n",
      "       [[152, 147, 186],\n",
      "        [160, 154, 188],\n",
      "        [165, 158, 194],\n",
      "        ...,\n",
      "        [123, 123, 164],\n",
      "        [122, 124, 160],\n",
      "        [109, 112, 149]],\n",
      "\n",
      "       [[150, 148, 187],\n",
      "        [153, 151, 187],\n",
      "        [157, 150, 187],\n",
      "        ...,\n",
      "        [136, 135, 174],\n",
      "        [131, 128, 167],\n",
      "        [114, 117, 156]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[104, 103, 138],\n",
      "        [103, 108, 142],\n",
      "        [122, 120, 155],\n",
      "        ...,\n",
      "        [ 72,  79, 112],\n",
      "        [ 68,  70, 100],\n",
      "        [ 56,  61,  83]],\n",
      "\n",
      "       [[ 97,  97, 132],\n",
      "        [108, 111, 144],\n",
      "        [109, 111, 145],\n",
      "        ...,\n",
      "        [ 66,  74, 104],\n",
      "        [ 60,  64,  89],\n",
      "        [ 53,  53,  77]],\n",
      "\n",
      "       [[ 92,  93, 124],\n",
      "        [102, 103, 138],\n",
      "        [116, 116, 147],\n",
      "        ...,\n",
      "        [ 64,  69,  97],\n",
      "        [ 51,  56,  83],\n",
      "        [ 46,  44,  58]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12910>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 51, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[162, 156, 191],\n",
      "        [163, 155, 191],\n",
      "        [159, 152, 192],\n",
      "        ...,\n",
      "        [126, 128, 166],\n",
      "        [116, 118, 155],\n",
      "        [102, 107, 142]],\n",
      "\n",
      "       [[152, 147, 186],\n",
      "        [160, 154, 188],\n",
      "        [165, 158, 194],\n",
      "        ...,\n",
      "        [123, 123, 164],\n",
      "        [122, 124, 160],\n",
      "        [109, 112, 149]],\n",
      "\n",
      "       [[150, 148, 187],\n",
      "        [153, 151, 187],\n",
      "        [157, 150, 187],\n",
      "        ...,\n",
      "        [136, 135, 174],\n",
      "        [131, 128, 167],\n",
      "        [114, 117, 156]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[104, 103, 138],\n",
      "        [103, 108, 142],\n",
      "        [122, 120, 155],\n",
      "        ...,\n",
      "        [ 72,  79, 112],\n",
      "        [ 68,  70, 100],\n",
      "        [ 56,  61,  83]],\n",
      "\n",
      "       [[ 97,  97, 132],\n",
      "        [108, 111, 144],\n",
      "        [109, 111, 145],\n",
      "        ...,\n",
      "        [ 66,  74, 104],\n",
      "        [ 60,  64,  89],\n",
      "        [ 53,  53,  77]],\n",
      "\n",
      "       [[ 92,  93, 124],\n",
      "        [102, 103, 138],\n",
      "        [116, 116, 147],\n",
      "        ...,\n",
      "        [ 64,  69,  97],\n",
      "        [ 51,  56,  83],\n",
      "        [ 46,  44,  58]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12910>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[148, 137, 165],\n",
      "        [147, 135, 165],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [160, 147, 171],\n",
      "        [163, 147, 172]],\n",
      "\n",
      "       [[149, 135, 163],\n",
      "        [149, 137, 167],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [157, 143, 167],\n",
      "        [162, 147, 171],\n",
      "        [159, 145, 169]],\n",
      "\n",
      "       [[150, 136, 164],\n",
      "        [147, 135, 165],\n",
      "        [147, 135, 165],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [158, 142, 166],\n",
      "        [160, 146, 170]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[155, 142, 168],\n",
      "        [151, 141, 168],\n",
      "        [153, 141, 168],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [154, 140, 164],\n",
      "        [155, 141, 165]],\n",
      "\n",
      "       [[156, 143, 168],\n",
      "        [154, 141, 167],\n",
      "        [157, 144, 167],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [155, 141, 165],\n",
      "        [154, 140, 164]],\n",
      "\n",
      "       [[156, 142, 166],\n",
      "        [154, 141, 167],\n",
      "        [153, 141, 164],\n",
      "        ...,\n",
      "        [155, 139, 164],\n",
      "        [154, 141, 165],\n",
      "        [154, 140, 164]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d290>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[148, 137, 165],\n",
      "        [147, 135, 165],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [160, 147, 171],\n",
      "        [163, 147, 172]],\n",
      "\n",
      "       [[149, 135, 163],\n",
      "        [149, 137, 167],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [157, 143, 167],\n",
      "        [162, 147, 171],\n",
      "        [159, 145, 169]],\n",
      "\n",
      "       [[150, 136, 164],\n",
      "        [147, 135, 165],\n",
      "        [147, 135, 165],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [158, 142, 166],\n",
      "        [160, 146, 170]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[155, 142, 168],\n",
      "        [151, 141, 168],\n",
      "        [153, 141, 168],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [154, 140, 164],\n",
      "        [155, 141, 165]],\n",
      "\n",
      "       [[156, 143, 168],\n",
      "        [154, 141, 167],\n",
      "        [157, 144, 167],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [155, 141, 165],\n",
      "        [154, 140, 164]],\n",
      "\n",
      "       [[156, 142, 166],\n",
      "        [154, 141, 167],\n",
      "        [153, 141, 164],\n",
      "        ...,\n",
      "        [155, 139, 164],\n",
      "        [154, 141, 165],\n",
      "        [154, 140, 164]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d290>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 44, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 167, 223],\n",
      "        [158, 162, 227],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [163, 169, 222],\n",
      "        [172, 173, 220],\n",
      "        [155, 161, 215]],\n",
      "\n",
      "       [[173, 166, 234],\n",
      "        [161, 160, 225],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [168, 168, 223],\n",
      "        [174, 172, 224],\n",
      "        [156, 162, 212]],\n",
      "\n",
      "       [[151, 153, 220],\n",
      "        [150, 151, 221],\n",
      "        [138, 141, 215],\n",
      "        ...,\n",
      "        [168, 169, 220],\n",
      "        [165, 169, 221],\n",
      "        [160, 165, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[144, 156, 205],\n",
      "        [148, 152, 204],\n",
      "        [148, 153, 205],\n",
      "        ...,\n",
      "        [149, 162, 208],\n",
      "        [149, 161, 204],\n",
      "        [161, 168, 212]],\n",
      "\n",
      "       [[144, 156, 209],\n",
      "        [143, 155, 208],\n",
      "        [144, 151, 206],\n",
      "        ...,\n",
      "        [154, 163, 207],\n",
      "        [152, 162, 206],\n",
      "        [158, 164, 210]],\n",
      "\n",
      "       [[146, 157, 205],\n",
      "        [161, 167, 214],\n",
      "        [139, 153, 208],\n",
      "        ...,\n",
      "        [158, 168, 210],\n",
      "        [158, 167, 209],\n",
      "        [159, 168, 211]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0dd50>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 55, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[250, 217, 244],\n",
      "        [254, 218, 244],\n",
      "        [252, 221, 244],\n",
      "        ...,\n",
      "        [253, 204, 222],\n",
      "        [250, 204, 219],\n",
      "        [249, 200, 214]],\n",
      "\n",
      "       [[249, 212, 239],\n",
      "        [247, 216, 244],\n",
      "        [251, 218, 244],\n",
      "        ...,\n",
      "        [252, 206, 225],\n",
      "        [250, 199, 218],\n",
      "        [243, 195, 215]],\n",
      "\n",
      "       [[241, 210, 242],\n",
      "        [254, 217, 244],\n",
      "        [253, 219, 247],\n",
      "        ...,\n",
      "        [253, 206, 224],\n",
      "        [250, 197, 219],\n",
      "        [242, 197, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[230, 196, 211],\n",
      "        [236, 199, 212],\n",
      "        [237, 200, 216],\n",
      "        ...,\n",
      "        [236, 201, 216],\n",
      "        [237, 200, 214],\n",
      "        [234, 198, 209]],\n",
      "\n",
      "       [[233, 202, 212],\n",
      "        [232, 195, 210],\n",
      "        [235, 198, 213],\n",
      "        ...,\n",
      "        [241, 203, 219],\n",
      "        [241, 203, 215],\n",
      "        [220, 189, 204]],\n",
      "\n",
      "       [[228, 198, 208],\n",
      "        [229, 198, 208],\n",
      "        [238, 198, 215],\n",
      "        ...,\n",
      "        [233, 199, 212],\n",
      "        [235, 200, 214],\n",
      "        [235, 195, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d690>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[148, 137, 165],\n",
      "        [147, 135, 165],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [160, 147, 171],\n",
      "        [163, 147, 172]],\n",
      "\n",
      "       [[149, 135, 163],\n",
      "        [149, 137, 167],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [157, 143, 167],\n",
      "        [162, 147, 171],\n",
      "        [159, 145, 169]],\n",
      "\n",
      "       [[150, 136, 164],\n",
      "        [147, 135, 165],\n",
      "        [147, 135, 165],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [158, 142, 166],\n",
      "        [160, 146, 170]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[155, 142, 168],\n",
      "        [151, 141, 168],\n",
      "        [153, 141, 168],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [154, 140, 164],\n",
      "        [155, 141, 165]],\n",
      "\n",
      "       [[156, 143, 168],\n",
      "        [154, 141, 167],\n",
      "        [157, 144, 167],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [155, 141, 165],\n",
      "        [154, 140, 164]],\n",
      "\n",
      "       [[156, 142, 166],\n",
      "        [154, 141, 167],\n",
      "        [153, 141, 164],\n",
      "        ...,\n",
      "        [155, 139, 164],\n",
      "        [154, 141, 165],\n",
      "        [154, 140, 164]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d290>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 58, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 84,  96, 139],\n",
      "        [ 85,  96, 138],\n",
      "        [ 89,  99, 144],\n",
      "        ...,\n",
      "        [ 72,  80, 123],\n",
      "        [ 67,  76, 119],\n",
      "        [ 68,  77, 119]],\n",
      "\n",
      "       [[ 77,  89, 131],\n",
      "        [ 79,  94, 138],\n",
      "        [ 84,  94, 137],\n",
      "        ...,\n",
      "        [ 76,  84, 127],\n",
      "        [ 68,  77, 120],\n",
      "        [ 71,  80, 123]],\n",
      "\n",
      "       [[ 77,  89, 132],\n",
      "        [ 78,  90, 133],\n",
      "        [ 85,  92, 137],\n",
      "        ...,\n",
      "        [ 78,  87, 130],\n",
      "        [ 72,  81, 124],\n",
      "        [ 70,  79, 123]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[ 74,  86, 128],\n",
      "        [ 71,  83, 125],\n",
      "        [ 75,  84, 127],\n",
      "        ...,\n",
      "        [ 75,  78, 112],\n",
      "        [ 77,  75, 111],\n",
      "        [ 75,  74, 108]],\n",
      "\n",
      "       [[ 76,  85, 128],\n",
      "        [ 75,  84, 127],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 75,  73, 110],\n",
      "        [ 74,  72, 109],\n",
      "        [ 70,  69, 101]],\n",
      "\n",
      "       [[ 72,  82, 121],\n",
      "        [ 72,  81, 125],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 69,  68, 107],\n",
      "        [ 67,  69, 104],\n",
      "        [ 66,  64,  99]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 44, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 167, 223],\n",
      "        [158, 162, 227],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [163, 169, 222],\n",
      "        [172, 173, 220],\n",
      "        [155, 161, 215]],\n",
      "\n",
      "       [[173, 166, 234],\n",
      "        [161, 160, 225],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [168, 168, 223],\n",
      "        [174, 172, 224],\n",
      "        [156, 162, 212]],\n",
      "\n",
      "       [[151, 153, 220],\n",
      "        [150, 151, 221],\n",
      "        [138, 141, 215],\n",
      "        ...,\n",
      "        [168, 169, 220],\n",
      "        [165, 169, 221],\n",
      "        [160, 165, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[144, 156, 205],\n",
      "        [148, 152, 204],\n",
      "        [148, 153, 205],\n",
      "        ...,\n",
      "        [149, 162, 208],\n",
      "        [149, 161, 204],\n",
      "        [161, 168, 212]],\n",
      "\n",
      "       [[144, 156, 209],\n",
      "        [143, 155, 208],\n",
      "        [144, 151, 206],\n",
      "        ...,\n",
      "        [154, 163, 207],\n",
      "        [152, 162, 206],\n",
      "        [158, 164, 210]],\n",
      "\n",
      "       [[146, 157, 205],\n",
      "        [161, 167, 214],\n",
      "        [139, 153, 208],\n",
      "        ...,\n",
      "        [158, 168, 210],\n",
      "        [158, 167, 209],\n",
      "        [159, 168, 211]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0dd50>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 58, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 84,  96, 139],\n",
      "        [ 85,  96, 138],\n",
      "        [ 89,  99, 144],\n",
      "        ...,\n",
      "        [ 72,  80, 123],\n",
      "        [ 67,  76, 119],\n",
      "        [ 68,  77, 119]],\n",
      "\n",
      "       [[ 77,  89, 131],\n",
      "        [ 79,  94, 138],\n",
      "        [ 84,  94, 137],\n",
      "        ...,\n",
      "        [ 76,  84, 127],\n",
      "        [ 68,  77, 120],\n",
      "        [ 71,  80, 123]],\n",
      "\n",
      "       [[ 77,  89, 132],\n",
      "        [ 78,  90, 133],\n",
      "        [ 85,  92, 137],\n",
      "        ...,\n",
      "        [ 78,  87, 130],\n",
      "        [ 72,  81, 124],\n",
      "        [ 70,  79, 123]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[ 74,  86, 128],\n",
      "        [ 71,  83, 125],\n",
      "        [ 75,  84, 127],\n",
      "        ...,\n",
      "        [ 75,  78, 112],\n",
      "        [ 77,  75, 111],\n",
      "        [ 75,  74, 108]],\n",
      "\n",
      "       [[ 76,  85, 128],\n",
      "        [ 75,  84, 127],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 75,  73, 110],\n",
      "        [ 74,  72, 109],\n",
      "        [ 70,  69, 101]],\n",
      "\n",
      "       [[ 72,  82, 121],\n",
      "        [ 72,  81, 125],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 69,  68, 107],\n",
      "        [ 67,  69, 104],\n",
      "        [ 66,  64,  99]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 44, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 167, 223],\n",
      "        [158, 162, 227],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [163, 169, 222],\n",
      "        [172, 173, 220],\n",
      "        [155, 161, 215]],\n",
      "\n",
      "       [[173, 166, 234],\n",
      "        [161, 160, 225],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [168, 168, 223],\n",
      "        [174, 172, 224],\n",
      "        [156, 162, 212]],\n",
      "\n",
      "       [[151, 153, 220],\n",
      "        [150, 151, 221],\n",
      "        [138, 141, 215],\n",
      "        ...,\n",
      "        [168, 169, 220],\n",
      "        [165, 169, 221],\n",
      "        [160, 165, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[144, 156, 205],\n",
      "        [148, 152, 204],\n",
      "        [148, 153, 205],\n",
      "        ...,\n",
      "        [149, 162, 208],\n",
      "        [149, 161, 204],\n",
      "        [161, 168, 212]],\n",
      "\n",
      "       [[144, 156, 209],\n",
      "        [143, 155, 208],\n",
      "        [144, 151, 206],\n",
      "        ...,\n",
      "        [154, 163, 207],\n",
      "        [152, 162, 206],\n",
      "        [158, 164, 210]],\n",
      "\n",
      "       [[146, 157, 205],\n",
      "        [161, 167, 214],\n",
      "        [139, 153, 208],\n",
      "        ...,\n",
      "        [158, 168, 210],\n",
      "        [158, 167, 209],\n",
      "        [159, 168, 211]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0dd50>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[148, 137, 165],\n",
      "        [147, 135, 165],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [160, 147, 171],\n",
      "        [163, 147, 172]],\n",
      "\n",
      "       [[149, 135, 163],\n",
      "        [149, 137, 167],\n",
      "        [148, 136, 166],\n",
      "        ...,\n",
      "        [157, 143, 167],\n",
      "        [162, 147, 171],\n",
      "        [159, 145, 169]],\n",
      "\n",
      "       [[150, 136, 164],\n",
      "        [147, 135, 165],\n",
      "        [147, 135, 165],\n",
      "        ...,\n",
      "        [155, 141, 165],\n",
      "        [158, 142, 166],\n",
      "        [160, 146, 170]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[155, 142, 168],\n",
      "        [151, 141, 168],\n",
      "        [153, 141, 168],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [154, 140, 164],\n",
      "        [155, 141, 165]],\n",
      "\n",
      "       [[156, 143, 168],\n",
      "        [154, 141, 167],\n",
      "        [157, 144, 167],\n",
      "        ...,\n",
      "        [156, 142, 166],\n",
      "        [155, 141, 165],\n",
      "        [154, 140, 164]],\n",
      "\n",
      "       [[156, 142, 166],\n",
      "        [154, 141, 167],\n",
      "        [153, 141, 164],\n",
      "        ...,\n",
      "        [155, 139, 164],\n",
      "        [154, 141, 165],\n",
      "        [154, 140, 164]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d290>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 44, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 167, 223],\n",
      "        [158, 162, 227],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [163, 169, 222],\n",
      "        [172, 173, 220],\n",
      "        [155, 161, 215]],\n",
      "\n",
      "       [[173, 166, 234],\n",
      "        [161, 160, 225],\n",
      "        [156, 158, 222],\n",
      "        ...,\n",
      "        [168, 168, 223],\n",
      "        [174, 172, 224],\n",
      "        [156, 162, 212]],\n",
      "\n",
      "       [[151, 153, 220],\n",
      "        [150, 151, 221],\n",
      "        [138, 141, 215],\n",
      "        ...,\n",
      "        [168, 169, 220],\n",
      "        [165, 169, 221],\n",
      "        [160, 165, 216]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[144, 156, 205],\n",
      "        [148, 152, 204],\n",
      "        [148, 153, 205],\n",
      "        ...,\n",
      "        [149, 162, 208],\n",
      "        [149, 161, 204],\n",
      "        [161, 168, 212]],\n",
      "\n",
      "       [[144, 156, 209],\n",
      "        [143, 155, 208],\n",
      "        [144, 151, 206],\n",
      "        ...,\n",
      "        [154, 163, 207],\n",
      "        [152, 162, 206],\n",
      "        [158, 164, 210]],\n",
      "\n",
      "       [[146, 157, 205],\n",
      "        [161, 167, 214],\n",
      "        [139, 153, 208],\n",
      "        ...,\n",
      "        [158, 168, 210],\n",
      "        [158, 167, 209],\n",
      "        [159, 168, 211]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0dd50>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[46, 40, 62],\n",
      "        [55, 52, 73],\n",
      "        [68, 64, 91],\n",
      "        ...,\n",
      "        [56, 50, 68],\n",
      "        [41, 35, 52],\n",
      "        [34, 27, 43]],\n",
      "\n",
      "       [[48, 45, 68],\n",
      "        [61, 60, 81],\n",
      "        [70, 66, 97],\n",
      "        ...,\n",
      "        [63, 56, 77],\n",
      "        [47, 42, 58],\n",
      "        [39, 32, 45]],\n",
      "\n",
      "       [[52, 47, 77],\n",
      "        [64, 59, 90],\n",
      "        [75, 75, 98],\n",
      "        ...,\n",
      "        [70, 64, 83],\n",
      "        [55, 50, 66],\n",
      "        [41, 35, 53]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[34, 33, 48],\n",
      "        [39, 37, 56],\n",
      "        [49, 43, 69],\n",
      "        ...,\n",
      "        [18, 14, 21],\n",
      "        [ 8,  7, 15],\n",
      "        [ 6,  5, 13]],\n",
      "\n",
      "       [[31, 27, 45],\n",
      "        [38, 32, 54],\n",
      "        [44, 37, 61],\n",
      "        ...,\n",
      "        [15,  7, 16],\n",
      "        [ 5,  4, 13],\n",
      "        [ 4,  3, 12]],\n",
      "\n",
      "       [[27, 26, 44],\n",
      "        [33, 30, 50],\n",
      "        [40, 34, 58],\n",
      "        ...,\n",
      "        [ 5,  4, 13],\n",
      "        [ 6,  5, 14],\n",
      "        [ 1,  2, 11]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0da10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 47, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 147, 237],\n",
      "        [164, 150, 233],\n",
      "        [167, 150, 238],\n",
      "        ...,\n",
      "        [159, 146, 238],\n",
      "        [158, 145, 235],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       [[167, 149, 238],\n",
      "        [166, 149, 238],\n",
      "        [163, 149, 236],\n",
      "        ...,\n",
      "        [156, 142, 234],\n",
      "        [154, 143, 233],\n",
      "        [152, 141, 232]],\n",
      "\n",
      "       [[163, 148, 236],\n",
      "        [165, 149, 236],\n",
      "        [165, 150, 238],\n",
      "        ...,\n",
      "        [156, 143, 233],\n",
      "        [155, 138, 230],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[151, 146, 223],\n",
      "        [154, 148, 221],\n",
      "        [150, 145, 222],\n",
      "        ...,\n",
      "        [149, 145, 217],\n",
      "        [149, 143, 214],\n",
      "        [147, 144, 215]],\n",
      "\n",
      "       [[143, 138, 218],\n",
      "        [150, 146, 222],\n",
      "        [154, 152, 225],\n",
      "        ...,\n",
      "        [155, 148, 222],\n",
      "        [150, 145, 217],\n",
      "        [146, 143, 214]],\n",
      "\n",
      "       [[148, 144, 223],\n",
      "        [149, 145, 220],\n",
      "        [154, 149, 225],\n",
      "        ...,\n",
      "        [148, 146, 214],\n",
      "        [152, 144, 217],\n",
      "        [153, 144, 214]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7cc1b61b10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[46, 40, 62],\n",
      "        [55, 52, 73],\n",
      "        [68, 64, 91],\n",
      "        ...,\n",
      "        [56, 50, 68],\n",
      "        [41, 35, 52],\n",
      "        [34, 27, 43]],\n",
      "\n",
      "       [[48, 45, 68],\n",
      "        [61, 60, 81],\n",
      "        [70, 66, 97],\n",
      "        ...,\n",
      "        [63, 56, 77],\n",
      "        [47, 42, 58],\n",
      "        [39, 32, 45]],\n",
      "\n",
      "       [[52, 47, 77],\n",
      "        [64, 59, 90],\n",
      "        [75, 75, 98],\n",
      "        ...,\n",
      "        [70, 64, 83],\n",
      "        [55, 50, 66],\n",
      "        [41, 35, 53]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[34, 33, 48],\n",
      "        [39, 37, 56],\n",
      "        [49, 43, 69],\n",
      "        ...,\n",
      "        [18, 14, 21],\n",
      "        [ 8,  7, 15],\n",
      "        [ 6,  5, 13]],\n",
      "\n",
      "       [[31, 27, 45],\n",
      "        [38, 32, 54],\n",
      "        [44, 37, 61],\n",
      "        ...,\n",
      "        [15,  7, 16],\n",
      "        [ 5,  4, 13],\n",
      "        [ 4,  3, 12]],\n",
      "\n",
      "       [[27, 26, 44],\n",
      "        [33, 30, 50],\n",
      "        [40, 34, 58],\n",
      "        ...,\n",
      "        [ 5,  4, 13],\n",
      "        [ 6,  5, 14],\n",
      "        [ 1,  2, 11]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0da10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 47, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 147, 237],\n",
      "        [164, 150, 233],\n",
      "        [167, 150, 238],\n",
      "        ...,\n",
      "        [159, 146, 238],\n",
      "        [158, 145, 235],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       [[167, 149, 238],\n",
      "        [166, 149, 238],\n",
      "        [163, 149, 236],\n",
      "        ...,\n",
      "        [156, 142, 234],\n",
      "        [154, 143, 233],\n",
      "        [152, 141, 232]],\n",
      "\n",
      "       [[163, 148, 236],\n",
      "        [165, 149, 236],\n",
      "        [165, 150, 238],\n",
      "        ...,\n",
      "        [156, 143, 233],\n",
      "        [155, 138, 230],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[151, 146, 223],\n",
      "        [154, 148, 221],\n",
      "        [150, 145, 222],\n",
      "        ...,\n",
      "        [149, 145, 217],\n",
      "        [149, 143, 214],\n",
      "        [147, 144, 215]],\n",
      "\n",
      "       [[143, 138, 218],\n",
      "        [150, 146, 222],\n",
      "        [154, 152, 225],\n",
      "        ...,\n",
      "        [155, 148, 222],\n",
      "        [150, 145, 217],\n",
      "        [146, 143, 214]],\n",
      "\n",
      "       [[148, 144, 223],\n",
      "        [149, 145, 220],\n",
      "        [154, 149, 225],\n",
      "        ...,\n",
      "        [148, 146, 214],\n",
      "        [152, 144, 217],\n",
      "        [153, 144, 214]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7cc1b61b10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[46, 40, 62],\n",
      "        [55, 52, 73],\n",
      "        [68, 64, 91],\n",
      "        ...,\n",
      "        [56, 50, 68],\n",
      "        [41, 35, 52],\n",
      "        [34, 27, 43]],\n",
      "\n",
      "       [[48, 45, 68],\n",
      "        [61, 60, 81],\n",
      "        [70, 66, 97],\n",
      "        ...,\n",
      "        [63, 56, 77],\n",
      "        [47, 42, 58],\n",
      "        [39, 32, 45]],\n",
      "\n",
      "       [[52, 47, 77],\n",
      "        [64, 59, 90],\n",
      "        [75, 75, 98],\n",
      "        ...,\n",
      "        [70, 64, 83],\n",
      "        [55, 50, 66],\n",
      "        [41, 35, 53]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[34, 33, 48],\n",
      "        [39, 37, 56],\n",
      "        [49, 43, 69],\n",
      "        ...,\n",
      "        [18, 14, 21],\n",
      "        [ 8,  7, 15],\n",
      "        [ 6,  5, 13]],\n",
      "\n",
      "       [[31, 27, 45],\n",
      "        [38, 32, 54],\n",
      "        [44, 37, 61],\n",
      "        ...,\n",
      "        [15,  7, 16],\n",
      "        [ 5,  4, 13],\n",
      "        [ 4,  3, 12]],\n",
      "\n",
      "       [[27, 26, 44],\n",
      "        [33, 30, 50],\n",
      "        [40, 34, 58],\n",
      "        ...,\n",
      "        [ 5,  4, 13],\n",
      "        [ 6,  5, 14],\n",
      "        [ 1,  2, 11]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0da10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 30, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[189, 155, 181],\n",
      "        [190, 158, 184],\n",
      "        [188, 154, 181],\n",
      "        ...,\n",
      "        [172, 141, 170],\n",
      "        [173, 143, 175],\n",
      "        [174, 146, 176]],\n",
      "\n",
      "       [[187, 158, 182],\n",
      "        [190, 157, 183],\n",
      "        [188, 155, 182],\n",
      "        ...,\n",
      "        [131,  93, 126],\n",
      "        [159,  98, 124],\n",
      "        [166, 125, 148]],\n",
      "\n",
      "       [[188, 156, 183],\n",
      "        [191, 157, 183],\n",
      "        [192, 158, 185],\n",
      "        ...,\n",
      "        [132, 142, 172],\n",
      "        [115, 124, 159],\n",
      "        [122, 119, 150]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[170, 142, 165],\n",
      "        [177, 146, 171],\n",
      "        [182, 148, 173],\n",
      "        ...,\n",
      "        [171, 147, 172],\n",
      "        [168, 145, 170],\n",
      "        [162, 143, 166]],\n",
      "\n",
      "       [[168, 140, 164],\n",
      "        [172, 146, 169],\n",
      "        [180, 147, 170],\n",
      "        ...,\n",
      "        [164, 141, 171],\n",
      "        [164, 140, 172],\n",
      "        [160, 138, 167]],\n",
      "\n",
      "       [[163, 137, 161],\n",
      "        [169, 143, 167],\n",
      "        [173, 145, 169],\n",
      "        ...,\n",
      "        [161, 137, 168],\n",
      "        [161, 138, 164],\n",
      "        [159, 136, 163]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d2d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[161, 152, 231],\n",
      "        [160, 147, 229],\n",
      "        [160, 148, 229],\n",
      "        ...,\n",
      "        [164, 153, 229],\n",
      "        [164, 154, 230],\n",
      "        [161, 152, 232]],\n",
      "\n",
      "       [[162, 153, 231],\n",
      "        [158, 148, 231],\n",
      "        [158, 149, 229],\n",
      "        ...,\n",
      "        [159, 152, 231],\n",
      "        [163, 152, 226],\n",
      "        [168, 157, 236]],\n",
      "\n",
      "       [[155, 149, 228],\n",
      "        [157, 151, 229],\n",
      "        [154, 146, 227],\n",
      "        ...,\n",
      "        [161, 155, 230],\n",
      "        [162, 153, 227],\n",
      "        [165, 159, 231]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[139, 144, 215],\n",
      "        [141, 143, 213],\n",
      "        [145, 144, 215],\n",
      "        ...,\n",
      "        [153, 153, 214],\n",
      "        [153, 151, 216],\n",
      "        [155, 154, 218]],\n",
      "\n",
      "       [[132, 135, 210],\n",
      "        [136, 142, 212],\n",
      "        [141, 147, 216],\n",
      "        ...,\n",
      "        [154, 152, 216],\n",
      "        [153, 152, 214],\n",
      "        [154, 153, 214]],\n",
      "\n",
      "       [[130, 131, 210],\n",
      "        [129, 130, 207],\n",
      "        [140, 140, 215],\n",
      "        ...,\n",
      "        [151, 153, 214],\n",
      "        [150, 150, 212],\n",
      "        [150, 143, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 58, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 84,  96, 139],\n",
      "        [ 85,  96, 138],\n",
      "        [ 89,  99, 144],\n",
      "        ...,\n",
      "        [ 72,  80, 123],\n",
      "        [ 67,  76, 119],\n",
      "        [ 68,  77, 119]],\n",
      "\n",
      "       [[ 77,  89, 131],\n",
      "        [ 79,  94, 138],\n",
      "        [ 84,  94, 137],\n",
      "        ...,\n",
      "        [ 76,  84, 127],\n",
      "        [ 68,  77, 120],\n",
      "        [ 71,  80, 123]],\n",
      "\n",
      "       [[ 77,  89, 132],\n",
      "        [ 78,  90, 133],\n",
      "        [ 85,  92, 137],\n",
      "        ...,\n",
      "        [ 78,  87, 130],\n",
      "        [ 72,  81, 124],\n",
      "        [ 70,  79, 123]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[ 74,  86, 128],\n",
      "        [ 71,  83, 125],\n",
      "        [ 75,  84, 127],\n",
      "        ...,\n",
      "        [ 75,  78, 112],\n",
      "        [ 77,  75, 111],\n",
      "        [ 75,  74, 108]],\n",
      "\n",
      "       [[ 76,  85, 128],\n",
      "        [ 75,  84, 127],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 75,  73, 110],\n",
      "        [ 74,  72, 109],\n",
      "        [ 70,  69, 101]],\n",
      "\n",
      "       [[ 72,  82, 121],\n",
      "        [ 72,  81, 125],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 69,  68, 107],\n",
      "        [ 67,  69, 104],\n",
      "        [ 66,  64,  99]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 47, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 147, 237],\n",
      "        [164, 150, 233],\n",
      "        [167, 150, 238],\n",
      "        ...,\n",
      "        [159, 146, 238],\n",
      "        [158, 145, 235],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       [[167, 149, 238],\n",
      "        [166, 149, 238],\n",
      "        [163, 149, 236],\n",
      "        ...,\n",
      "        [156, 142, 234],\n",
      "        [154, 143, 233],\n",
      "        [152, 141, 232]],\n",
      "\n",
      "       [[163, 148, 236],\n",
      "        [165, 149, 236],\n",
      "        [165, 150, 238],\n",
      "        ...,\n",
      "        [156, 143, 233],\n",
      "        [155, 138, 230],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[151, 146, 223],\n",
      "        [154, 148, 221],\n",
      "        [150, 145, 222],\n",
      "        ...,\n",
      "        [149, 145, 217],\n",
      "        [149, 143, 214],\n",
      "        [147, 144, 215]],\n",
      "\n",
      "       [[143, 138, 218],\n",
      "        [150, 146, 222],\n",
      "        [154, 152, 225],\n",
      "        ...,\n",
      "        [155, 148, 222],\n",
      "        [150, 145, 217],\n",
      "        [146, 143, 214]],\n",
      "\n",
      "       [[148, 144, 223],\n",
      "        [149, 145, 220],\n",
      "        [154, 149, 225],\n",
      "        ...,\n",
      "        [148, 146, 214],\n",
      "        [152, 144, 217],\n",
      "        [153, 144, 214]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7cc1b61b10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 58, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 84,  96, 139],\n",
      "        [ 85,  96, 138],\n",
      "        [ 89,  99, 144],\n",
      "        ...,\n",
      "        [ 72,  80, 123],\n",
      "        [ 67,  76, 119],\n",
      "        [ 68,  77, 119]],\n",
      "\n",
      "       [[ 77,  89, 131],\n",
      "        [ 79,  94, 138],\n",
      "        [ 84,  94, 137],\n",
      "        ...,\n",
      "        [ 76,  84, 127],\n",
      "        [ 68,  77, 120],\n",
      "        [ 71,  80, 123]],\n",
      "\n",
      "       [[ 77,  89, 132],\n",
      "        [ 78,  90, 133],\n",
      "        [ 85,  92, 137],\n",
      "        ...,\n",
      "        [ 78,  87, 130],\n",
      "        [ 72,  81, 124],\n",
      "        [ 70,  79, 123]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[ 74,  86, 128],\n",
      "        [ 71,  83, 125],\n",
      "        [ 75,  84, 127],\n",
      "        ...,\n",
      "        [ 75,  78, 112],\n",
      "        [ 77,  75, 111],\n",
      "        [ 75,  74, 108]],\n",
      "\n",
      "       [[ 76,  85, 128],\n",
      "        [ 75,  84, 127],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 75,  73, 110],\n",
      "        [ 74,  72, 109],\n",
      "        [ 70,  69, 101]],\n",
      "\n",
      "       [[ 72,  82, 121],\n",
      "        [ 72,  81, 125],\n",
      "        [ 71,  80, 123],\n",
      "        ...,\n",
      "        [ 69,  68, 107],\n",
      "        [ 67,  69, 104],\n",
      "        [ 66,  64,  99]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[161, 152, 231],\n",
      "        [160, 147, 229],\n",
      "        [160, 148, 229],\n",
      "        ...,\n",
      "        [164, 153, 229],\n",
      "        [164, 154, 230],\n",
      "        [161, 152, 232]],\n",
      "\n",
      "       [[162, 153, 231],\n",
      "        [158, 148, 231],\n",
      "        [158, 149, 229],\n",
      "        ...,\n",
      "        [159, 152, 231],\n",
      "        [163, 152, 226],\n",
      "        [168, 157, 236]],\n",
      "\n",
      "       [[155, 149, 228],\n",
      "        [157, 151, 229],\n",
      "        [154, 146, 227],\n",
      "        ...,\n",
      "        [161, 155, 230],\n",
      "        [162, 153, 227],\n",
      "        [165, 159, 231]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[139, 144, 215],\n",
      "        [141, 143, 213],\n",
      "        [145, 144, 215],\n",
      "        ...,\n",
      "        [153, 153, 214],\n",
      "        [153, 151, 216],\n",
      "        [155, 154, 218]],\n",
      "\n",
      "       [[132, 135, 210],\n",
      "        [136, 142, 212],\n",
      "        [141, 147, 216],\n",
      "        ...,\n",
      "        [154, 152, 216],\n",
      "        [153, 152, 214],\n",
      "        [154, 153, 214]],\n",
      "\n",
      "       [[130, 131, 210],\n",
      "        [129, 130, 207],\n",
      "        [140, 140, 215],\n",
      "        ...,\n",
      "        [151, 153, 214],\n",
      "        [150, 150, 212],\n",
      "        [150, 143, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 47, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[164, 147, 237],\n",
      "        [164, 150, 233],\n",
      "        [167, 150, 238],\n",
      "        ...,\n",
      "        [159, 146, 238],\n",
      "        [158, 145, 235],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       [[167, 149, 238],\n",
      "        [166, 149, 238],\n",
      "        [163, 149, 236],\n",
      "        ...,\n",
      "        [156, 142, 234],\n",
      "        [154, 143, 233],\n",
      "        [152, 141, 232]],\n",
      "\n",
      "       [[163, 148, 236],\n",
      "        [165, 149, 236],\n",
      "        [165, 150, 238],\n",
      "        ...,\n",
      "        [156, 143, 233],\n",
      "        [155, 138, 230],\n",
      "        [159, 143, 234]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[151, 146, 223],\n",
      "        [154, 148, 221],\n",
      "        [150, 145, 222],\n",
      "        ...,\n",
      "        [149, 145, 217],\n",
      "        [149, 143, 214],\n",
      "        [147, 144, 215]],\n",
      "\n",
      "       [[143, 138, 218],\n",
      "        [150, 146, 222],\n",
      "        [154, 152, 225],\n",
      "        ...,\n",
      "        [155, 148, 222],\n",
      "        [150, 145, 217],\n",
      "        [146, 143, 214]],\n",
      "\n",
      "       [[148, 144, 223],\n",
      "        [149, 145, 220],\n",
      "        [154, 149, 225],\n",
      "        ...,\n",
      "        [148, 146, 214],\n",
      "        [152, 144, 217],\n",
      "        [153, 144, 214]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7cc1b61b10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 30, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[189, 155, 181],\n",
      "        [190, 158, 184],\n",
      "        [188, 154, 181],\n",
      "        ...,\n",
      "        [172, 141, 170],\n",
      "        [173, 143, 175],\n",
      "        [174, 146, 176]],\n",
      "\n",
      "       [[187, 158, 182],\n",
      "        [190, 157, 183],\n",
      "        [188, 155, 182],\n",
      "        ...,\n",
      "        [131,  93, 126],\n",
      "        [159,  98, 124],\n",
      "        [166, 125, 148]],\n",
      "\n",
      "       [[188, 156, 183],\n",
      "        [191, 157, 183],\n",
      "        [192, 158, 185],\n",
      "        ...,\n",
      "        [132, 142, 172],\n",
      "        [115, 124, 159],\n",
      "        [122, 119, 150]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[170, 142, 165],\n",
      "        [177, 146, 171],\n",
      "        [182, 148, 173],\n",
      "        ...,\n",
      "        [171, 147, 172],\n",
      "        [168, 145, 170],\n",
      "        [162, 143, 166]],\n",
      "\n",
      "       [[168, 140, 164],\n",
      "        [172, 146, 169],\n",
      "        [180, 147, 170],\n",
      "        ...,\n",
      "        [164, 141, 171],\n",
      "        [164, 140, 172],\n",
      "        [160, 138, 167]],\n",
      "\n",
      "       [[163, 137, 161],\n",
      "        [169, 143, 167],\n",
      "        [173, 145, 169],\n",
      "        ...,\n",
      "        [161, 137, 168],\n",
      "        [161, 138, 164],\n",
      "        [159, 136, 163]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d2d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[46, 40, 62],\n",
      "        [55, 52, 73],\n",
      "        [68, 64, 91],\n",
      "        ...,\n",
      "        [56, 50, 68],\n",
      "        [41, 35, 52],\n",
      "        [34, 27, 43]],\n",
      "\n",
      "       [[48, 45, 68],\n",
      "        [61, 60, 81],\n",
      "        [70, 66, 97],\n",
      "        ...,\n",
      "        [63, 56, 77],\n",
      "        [47, 42, 58],\n",
      "        [39, 32, 45]],\n",
      "\n",
      "       [[52, 47, 77],\n",
      "        [64, 59, 90],\n",
      "        [75, 75, 98],\n",
      "        ...,\n",
      "        [70, 64, 83],\n",
      "        [55, 50, 66],\n",
      "        [41, 35, 53]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[34, 33, 48],\n",
      "        [39, 37, 56],\n",
      "        [49, 43, 69],\n",
      "        ...,\n",
      "        [18, 14, 21],\n",
      "        [ 8,  7, 15],\n",
      "        [ 6,  5, 13]],\n",
      "\n",
      "       [[31, 27, 45],\n",
      "        [38, 32, 54],\n",
      "        [44, 37, 61],\n",
      "        ...,\n",
      "        [15,  7, 16],\n",
      "        [ 5,  4, 13],\n",
      "        [ 4,  3, 12]],\n",
      "\n",
      "       [[27, 26, 44],\n",
      "        [33, 30, 50],\n",
      "        [40, 34, 58],\n",
      "        ...,\n",
      "        [ 5,  4, 13],\n",
      "        [ 6,  5, 14],\n",
      "        [ 1,  2, 11]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0da10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[161, 152, 231],\n",
      "        [160, 147, 229],\n",
      "        [160, 148, 229],\n",
      "        ...,\n",
      "        [164, 153, 229],\n",
      "        [164, 154, 230],\n",
      "        [161, 152, 232]],\n",
      "\n",
      "       [[162, 153, 231],\n",
      "        [158, 148, 231],\n",
      "        [158, 149, 229],\n",
      "        ...,\n",
      "        [159, 152, 231],\n",
      "        [163, 152, 226],\n",
      "        [168, 157, 236]],\n",
      "\n",
      "       [[155, 149, 228],\n",
      "        [157, 151, 229],\n",
      "        [154, 146, 227],\n",
      "        ...,\n",
      "        [161, 155, 230],\n",
      "        [162, 153, 227],\n",
      "        [165, 159, 231]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[139, 144, 215],\n",
      "        [141, 143, 213],\n",
      "        [145, 144, 215],\n",
      "        ...,\n",
      "        [153, 153, 214],\n",
      "        [153, 151, 216],\n",
      "        [155, 154, 218]],\n",
      "\n",
      "       [[132, 135, 210],\n",
      "        [136, 142, 212],\n",
      "        [141, 147, 216],\n",
      "        ...,\n",
      "        [154, 152, 216],\n",
      "        [153, 152, 214],\n",
      "        [154, 153, 214]],\n",
      "\n",
      "       [[130, 131, 210],\n",
      "        [129, 130, 207],\n",
      "        [140, 140, 215],\n",
      "        ...,\n",
      "        [151, 153, 214],\n",
      "        [150, 150, 212],\n",
      "        [150, 143, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 31, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 86,  97, 190],\n",
      "        [108, 116, 199],\n",
      "        [128, 132, 216],\n",
      "        ...,\n",
      "        [139, 124, 198],\n",
      "        [142, 122, 198],\n",
      "        [135, 123, 194]],\n",
      "\n",
      "       [[ 98, 107, 195],\n",
      "        [126, 130, 214],\n",
      "        [129, 136, 218],\n",
      "        ...,\n",
      "        [135, 123, 202],\n",
      "        [ 97,  87, 170],\n",
      "        [ 56,  47, 113]],\n",
      "\n",
      "       [[121, 126, 214],\n",
      "        [118, 126, 213],\n",
      "        [122, 126, 212],\n",
      "        ...,\n",
      "        [ 71,  61, 118],\n",
      "        [ 89,  80, 142],\n",
      "        [123, 111, 178]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[125, 125, 190],\n",
      "        [ 70,  65, 128],\n",
      "        [ 75,  70, 130],\n",
      "        ...,\n",
      "        [127, 118, 180],\n",
      "        [123, 116, 179],\n",
      "        [124, 112, 178]],\n",
      "\n",
      "       [[ 61,  57, 118],\n",
      "        [ 80,  71, 129],\n",
      "        [128, 119, 188],\n",
      "        ...,\n",
      "        [122, 119, 180],\n",
      "        [125, 116, 178],\n",
      "        [120, 111, 174]],\n",
      "\n",
      "       [[ 89,  77, 136],\n",
      "        [123, 120, 185],\n",
      "        [110, 108, 173],\n",
      "        ...,\n",
      "        [121, 113, 177],\n",
      "        [120, 111, 177],\n",
      "        [120, 111, 173]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d3d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 6, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[161, 152, 231],\n",
      "        [160, 147, 229],\n",
      "        [160, 148, 229],\n",
      "        ...,\n",
      "        [164, 153, 229],\n",
      "        [164, 154, 230],\n",
      "        [161, 152, 232]],\n",
      "\n",
      "       [[162, 153, 231],\n",
      "        [158, 148, 231],\n",
      "        [158, 149, 229],\n",
      "        ...,\n",
      "        [159, 152, 231],\n",
      "        [163, 152, 226],\n",
      "        [168, 157, 236]],\n",
      "\n",
      "       [[155, 149, 228],\n",
      "        [157, 151, 229],\n",
      "        [154, 146, 227],\n",
      "        ...,\n",
      "        [161, 155, 230],\n",
      "        [162, 153, 227],\n",
      "        [165, 159, 231]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[139, 144, 215],\n",
      "        [141, 143, 213],\n",
      "        [145, 144, 215],\n",
      "        ...,\n",
      "        [153, 153, 214],\n",
      "        [153, 151, 216],\n",
      "        [155, 154, 218]],\n",
      "\n",
      "       [[132, 135, 210],\n",
      "        [136, 142, 212],\n",
      "        [141, 147, 216],\n",
      "        ...,\n",
      "        [154, 152, 216],\n",
      "        [153, 152, 214],\n",
      "        [154, 153, 214]],\n",
      "\n",
      "       [[130, 131, 210],\n",
      "        [129, 130, 207],\n",
      "        [140, 140, 215],\n",
      "        ...,\n",
      "        [151, 153, 214],\n",
      "        [150, 150, 212],\n",
      "        [150, 143, 206]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12f90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 67, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[153, 135, 219],\n",
      "        [143, 136, 220],\n",
      "        [152, 139, 220],\n",
      "        ...,\n",
      "        [130, 119, 210],\n",
      "        [127, 109, 206],\n",
      "        [137, 121, 214]],\n",
      "\n",
      "       [[139, 129, 213],\n",
      "        [136, 130, 218],\n",
      "        [143, 137, 218],\n",
      "        ...,\n",
      "        [141, 123, 217],\n",
      "        [131, 118, 211],\n",
      "        [129, 116, 207]],\n",
      "\n",
      "       [[135, 125, 212],\n",
      "        [138, 133, 215],\n",
      "        [149, 138, 224],\n",
      "        ...,\n",
      "        [140, 126, 214],\n",
      "        [128, 119, 209],\n",
      "        [125, 114, 206]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[132, 121, 202],\n",
      "        [117, 113, 199],\n",
      "        [129, 123, 205],\n",
      "        ...,\n",
      "        [119, 107, 198],\n",
      "        [108, 103, 189],\n",
      "        [114, 108, 189]],\n",
      "\n",
      "       [[120, 121, 202],\n",
      "        [126, 122, 204],\n",
      "        [117, 120, 202],\n",
      "        ...,\n",
      "        [114, 108, 192],\n",
      "        [103, 104, 188],\n",
      "        [112, 108, 194]],\n",
      "\n",
      "       [[112, 117, 198],\n",
      "        [120, 117, 203],\n",
      "        [115, 115, 203],\n",
      "        ...,\n",
      "        [124, 114, 196],\n",
      "        [105, 106, 193],\n",
      "        [111, 107, 194]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0de90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 30, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[189, 155, 181],\n",
      "        [190, 158, 184],\n",
      "        [188, 154, 181],\n",
      "        ...,\n",
      "        [172, 141, 170],\n",
      "        [173, 143, 175],\n",
      "        [174, 146, 176]],\n",
      "\n",
      "       [[187, 158, 182],\n",
      "        [190, 157, 183],\n",
      "        [188, 155, 182],\n",
      "        ...,\n",
      "        [131,  93, 126],\n",
      "        [159,  98, 124],\n",
      "        [166, 125, 148]],\n",
      "\n",
      "       [[188, 156, 183],\n",
      "        [191, 157, 183],\n",
      "        [192, 158, 185],\n",
      "        ...,\n",
      "        [132, 142, 172],\n",
      "        [115, 124, 159],\n",
      "        [122, 119, 150]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[170, 142, 165],\n",
      "        [177, 146, 171],\n",
      "        [182, 148, 173],\n",
      "        ...,\n",
      "        [171, 147, 172],\n",
      "        [168, 145, 170],\n",
      "        [162, 143, 166]],\n",
      "\n",
      "       [[168, 140, 164],\n",
      "        [172, 146, 169],\n",
      "        [180, 147, 170],\n",
      "        ...,\n",
      "        [164, 141, 171],\n",
      "        [164, 140, 172],\n",
      "        [160, 138, 167]],\n",
      "\n",
      "       [[163, 137, 161],\n",
      "        [169, 143, 167],\n",
      "        [173, 145, 169],\n",
      "        ...,\n",
      "        [161, 137, 168],\n",
      "        [161, 138, 164],\n",
      "        [159, 136, 163]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d2d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 30, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[189, 155, 181],\n",
      "        [190, 158, 184],\n",
      "        [188, 154, 181],\n",
      "        ...,\n",
      "        [172, 141, 170],\n",
      "        [173, 143, 175],\n",
      "        [174, 146, 176]],\n",
      "\n",
      "       [[187, 158, 182],\n",
      "        [190, 157, 183],\n",
      "        [188, 155, 182],\n",
      "        ...,\n",
      "        [131,  93, 126],\n",
      "        [159,  98, 124],\n",
      "        [166, 125, 148]],\n",
      "\n",
      "       [[188, 156, 183],\n",
      "        [191, 157, 183],\n",
      "        [192, 158, 185],\n",
      "        ...,\n",
      "        [132, 142, 172],\n",
      "        [115, 124, 159],\n",
      "        [122, 119, 150]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[170, 142, 165],\n",
      "        [177, 146, 171],\n",
      "        [182, 148, 173],\n",
      "        ...,\n",
      "        [171, 147, 172],\n",
      "        [168, 145, 170],\n",
      "        [162, 143, 166]],\n",
      "\n",
      "       [[168, 140, 164],\n",
      "        [172, 146, 169],\n",
      "        [180, 147, 170],\n",
      "        ...,\n",
      "        [164, 141, 171],\n",
      "        [164, 140, 172],\n",
      "        [160, 138, 167]],\n",
      "\n",
      "       [[163, 137, 161],\n",
      "        [169, 143, 167],\n",
      "        [173, 145, 169],\n",
      "        ...,\n",
      "        [161, 137, 168],\n",
      "        [161, 138, 164],\n",
      "        [159, 136, 163]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d2d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 31, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 86,  97, 190],\n",
      "        [108, 116, 199],\n",
      "        [128, 132, 216],\n",
      "        ...,\n",
      "        [139, 124, 198],\n",
      "        [142, 122, 198],\n",
      "        [135, 123, 194]],\n",
      "\n",
      "       [[ 98, 107, 195],\n",
      "        [126, 130, 214],\n",
      "        [129, 136, 218],\n",
      "        ...,\n",
      "        [135, 123, 202],\n",
      "        [ 97,  87, 170],\n",
      "        [ 56,  47, 113]],\n",
      "\n",
      "       [[121, 126, 214],\n",
      "        [118, 126, 213],\n",
      "        [122, 126, 212],\n",
      "        ...,\n",
      "        [ 71,  61, 118],\n",
      "        [ 89,  80, 142],\n",
      "        [123, 111, 178]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[125, 125, 190],\n",
      "        [ 70,  65, 128],\n",
      "        [ 75,  70, 130],\n",
      "        ...,\n",
      "        [127, 118, 180],\n",
      "        [123, 116, 179],\n",
      "        [124, 112, 178]],\n",
      "\n",
      "       [[ 61,  57, 118],\n",
      "        [ 80,  71, 129],\n",
      "        [128, 119, 188],\n",
      "        ...,\n",
      "        [122, 119, 180],\n",
      "        [125, 116, 178],\n",
      "        [120, 111, 174]],\n",
      "\n",
      "       [[ 89,  77, 136],\n",
      "        [123, 120, 185],\n",
      "        [110, 108, 173],\n",
      "        ...,\n",
      "        [121, 113, 177],\n",
      "        [120, 111, 177],\n",
      "        [120, 111, 173]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d3d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 31, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 86,  97, 190],\n",
      "        [108, 116, 199],\n",
      "        [128, 132, 216],\n",
      "        ...,\n",
      "        [139, 124, 198],\n",
      "        [142, 122, 198],\n",
      "        [135, 123, 194]],\n",
      "\n",
      "       [[ 98, 107, 195],\n",
      "        [126, 130, 214],\n",
      "        [129, 136, 218],\n",
      "        ...,\n",
      "        [135, 123, 202],\n",
      "        [ 97,  87, 170],\n",
      "        [ 56,  47, 113]],\n",
      "\n",
      "       [[121, 126, 214],\n",
      "        [118, 126, 213],\n",
      "        [122, 126, 212],\n",
      "        ...,\n",
      "        [ 71,  61, 118],\n",
      "        [ 89,  80, 142],\n",
      "        [123, 111, 178]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[125, 125, 190],\n",
      "        [ 70,  65, 128],\n",
      "        [ 75,  70, 130],\n",
      "        ...,\n",
      "        [127, 118, 180],\n",
      "        [123, 116, 179],\n",
      "        [124, 112, 178]],\n",
      "\n",
      "       [[ 61,  57, 118],\n",
      "        [ 80,  71, 129],\n",
      "        [128, 119, 188],\n",
      "        ...,\n",
      "        [122, 119, 180],\n",
      "        [125, 116, 178],\n",
      "        [120, 111, 174]],\n",
      "\n",
      "       [[ 89,  77, 136],\n",
      "        [123, 120, 185],\n",
      "        [110, 108, 173],\n",
      "        ...,\n",
      "        [121, 113, 177],\n",
      "        [120, 111, 177],\n",
      "        [120, 111, 173]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d3d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 67, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[153, 135, 219],\n",
      "        [143, 136, 220],\n",
      "        [152, 139, 220],\n",
      "        ...,\n",
      "        [130, 119, 210],\n",
      "        [127, 109, 206],\n",
      "        [137, 121, 214]],\n",
      "\n",
      "       [[139, 129, 213],\n",
      "        [136, 130, 218],\n",
      "        [143, 137, 218],\n",
      "        ...,\n",
      "        [141, 123, 217],\n",
      "        [131, 118, 211],\n",
      "        [129, 116, 207]],\n",
      "\n",
      "       [[135, 125, 212],\n",
      "        [138, 133, 215],\n",
      "        [149, 138, 224],\n",
      "        ...,\n",
      "        [140, 126, 214],\n",
      "        [128, 119, 209],\n",
      "        [125, 114, 206]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[132, 121, 202],\n",
      "        [117, 113, 199],\n",
      "        [129, 123, 205],\n",
      "        ...,\n",
      "        [119, 107, 198],\n",
      "        [108, 103, 189],\n",
      "        [114, 108, 189]],\n",
      "\n",
      "       [[120, 121, 202],\n",
      "        [126, 122, 204],\n",
      "        [117, 120, 202],\n",
      "        ...,\n",
      "        [114, 108, 192],\n",
      "        [103, 104, 188],\n",
      "        [112, 108, 194]],\n",
      "\n",
      "       [[112, 117, 198],\n",
      "        [120, 117, 203],\n",
      "        [115, 115, 203],\n",
      "        ...,\n",
      "        [124, 114, 196],\n",
      "        [105, 106, 193],\n",
      "        [111, 107, 194]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0de90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 67, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[153, 135, 219],\n",
      "        [143, 136, 220],\n",
      "        [152, 139, 220],\n",
      "        ...,\n",
      "        [130, 119, 210],\n",
      "        [127, 109, 206],\n",
      "        [137, 121, 214]],\n",
      "\n",
      "       [[139, 129, 213],\n",
      "        [136, 130, 218],\n",
      "        [143, 137, 218],\n",
      "        ...,\n",
      "        [141, 123, 217],\n",
      "        [131, 118, 211],\n",
      "        [129, 116, 207]],\n",
      "\n",
      "       [[135, 125, 212],\n",
      "        [138, 133, 215],\n",
      "        [149, 138, 224],\n",
      "        ...,\n",
      "        [140, 126, 214],\n",
      "        [128, 119, 209],\n",
      "        [125, 114, 206]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[132, 121, 202],\n",
      "        [117, 113, 199],\n",
      "        [129, 123, 205],\n",
      "        ...,\n",
      "        [119, 107, 198],\n",
      "        [108, 103, 189],\n",
      "        [114, 108, 189]],\n",
      "\n",
      "       [[120, 121, 202],\n",
      "        [126, 122, 204],\n",
      "        [117, 120, 202],\n",
      "        ...,\n",
      "        [114, 108, 192],\n",
      "        [103, 104, 188],\n",
      "        [112, 108, 194]],\n",
      "\n",
      "       [[112, 117, 198],\n",
      "        [120, 117, 203],\n",
      "        [115, 115, 203],\n",
      "        ...,\n",
      "        [124, 114, 196],\n",
      "        [105, 106, 193],\n",
      "        [111, 107, 194]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0de90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 67, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[153, 135, 219],\n",
      "        [143, 136, 220],\n",
      "        [152, 139, 220],\n",
      "        ...,\n",
      "        [130, 119, 210],\n",
      "        [127, 109, 206],\n",
      "        [137, 121, 214]],\n",
      "\n",
      "       [[139, 129, 213],\n",
      "        [136, 130, 218],\n",
      "        [143, 137, 218],\n",
      "        ...,\n",
      "        [141, 123, 217],\n",
      "        [131, 118, 211],\n",
      "        [129, 116, 207]],\n",
      "\n",
      "       [[135, 125, 212],\n",
      "        [138, 133, 215],\n",
      "        [149, 138, 224],\n",
      "        ...,\n",
      "        [140, 126, 214],\n",
      "        [128, 119, 209],\n",
      "        [125, 114, 206]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[132, 121, 202],\n",
      "        [117, 113, 199],\n",
      "        [129, 123, 205],\n",
      "        ...,\n",
      "        [119, 107, 198],\n",
      "        [108, 103, 189],\n",
      "        [114, 108, 189]],\n",
      "\n",
      "       [[120, 121, 202],\n",
      "        [126, 122, 204],\n",
      "        [117, 120, 202],\n",
      "        ...,\n",
      "        [114, 108, 192],\n",
      "        [103, 104, 188],\n",
      "        [112, 108, 194]],\n",
      "\n",
      "       [[112, 117, 198],\n",
      "        [120, 117, 203],\n",
      "        [115, 115, 203],\n",
      "        ...,\n",
      "        [124, 114, 196],\n",
      "        [105, 106, 193],\n",
      "        [111, 107, 194]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0de90>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[172, 150, 174],\n",
      "        [182, 158, 185],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [176, 151, 181],\n",
      "        [172, 149, 173],\n",
      "        [154, 131, 150]],\n",
      "\n",
      "       [[175, 152, 179],\n",
      "        [181, 155, 183],\n",
      "        [187, 162, 188],\n",
      "        ...,\n",
      "        [175, 152, 180],\n",
      "        [177, 151, 177],\n",
      "        [163, 139, 161]],\n",
      "\n",
      "       [[180, 155, 184],\n",
      "        [181, 156, 184],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [181, 156, 184],\n",
      "        [176, 151, 179],\n",
      "        [169, 146, 169]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[172, 150, 175],\n",
      "        [170, 150, 175],\n",
      "        [172, 149, 177],\n",
      "        ...,\n",
      "        [165, 144, 168],\n",
      "        [144, 122, 150],\n",
      "        [122,  98, 123]],\n",
      "\n",
      "       [[165, 147, 169],\n",
      "        [172, 150, 176],\n",
      "        [168, 146, 174],\n",
      "        ...,\n",
      "        [156, 135, 162],\n",
      "        [135, 116, 144],\n",
      "        [ 94,  80,  88]],\n",
      "\n",
      "       [[161, 140, 162],\n",
      "        [167, 148, 170],\n",
      "        [170, 149, 176],\n",
      "        ...,\n",
      "        [149, 130, 157],\n",
      "        [123, 105, 126],\n",
      "        [ 72,  48,  49]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12a10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "ERROR:root:Error processing image {'id': 31, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[ 86,  97, 190],\n",
      "        [108, 116, 199],\n",
      "        [128, 132, 216],\n",
      "        ...,\n",
      "        [139, 124, 198],\n",
      "        [142, 122, 198],\n",
      "        [135, 123, 194]],\n",
      "\n",
      "       [[ 98, 107, 195],\n",
      "        [126, 130, 214],\n",
      "        [129, 136, 218],\n",
      "        ...,\n",
      "        [135, 123, 202],\n",
      "        [ 97,  87, 170],\n",
      "        [ 56,  47, 113]],\n",
      "\n",
      "       [[121, 126, 214],\n",
      "        [118, 126, 213],\n",
      "        [122, 126, 212],\n",
      "        ...,\n",
      "        [ 71,  61, 118],\n",
      "        [ 89,  80, 142],\n",
      "        [123, 111, 178]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[125, 125, 190],\n",
      "        [ 70,  65, 128],\n",
      "        [ 75,  70, 130],\n",
      "        ...,\n",
      "        [127, 118, 180],\n",
      "        [123, 116, 179],\n",
      "        [124, 112, 178]],\n",
      "\n",
      "       [[ 61,  57, 118],\n",
      "        [ 80,  71, 129],\n",
      "        [128, 119, 188],\n",
      "        ...,\n",
      "        [122, 119, 180],\n",
      "        [125, 116, 178],\n",
      "        [120, 111, 174]],\n",
      "\n",
      "       [[ 89,  77, 136],\n",
      "        [123, 120, 185],\n",
      "        [110, 108, 173],\n",
      "        ...,\n",
      "        [121, 113, 177],\n",
      "        [120, 111, 177],\n",
      "        [120, 111, 173]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d0d3d0>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[172, 150, 174],\n",
      "        [182, 158, 185],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [176, 151, 181],\n",
      "        [172, 149, 173],\n",
      "        [154, 131, 150]],\n",
      "\n",
      "       [[175, 152, 179],\n",
      "        [181, 155, 183],\n",
      "        [187, 162, 188],\n",
      "        ...,\n",
      "        [175, 152, 180],\n",
      "        [177, 151, 177],\n",
      "        [163, 139, 161]],\n",
      "\n",
      "       [[180, 155, 184],\n",
      "        [181, 156, 184],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [181, 156, 184],\n",
      "        [176, 151, 179],\n",
      "        [169, 146, 169]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[172, 150, 175],\n",
      "        [170, 150, 175],\n",
      "        [172, 149, 177],\n",
      "        ...,\n",
      "        [165, 144, 168],\n",
      "        [144, 122, 150],\n",
      "        [122,  98, 123]],\n",
      "\n",
      "       [[165, 147, 169],\n",
      "        [172, 150, 176],\n",
      "        [168, 146, 174],\n",
      "        ...,\n",
      "        [156, 135, 162],\n",
      "        [135, 116, 144],\n",
      "        [ 94,  80,  88]],\n",
      "\n",
      "       [[161, 140, 162],\n",
      "        [167, 148, 170],\n",
      "        [170, 149, 176],\n",
      "        ...,\n",
      "        [149, 130, 157],\n",
      "        [123, 105, 126],\n",
      "        [ 72,  48,  49]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12a10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[172, 150, 174],\n",
      "        [182, 158, 185],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [176, 151, 181],\n",
      "        [172, 149, 173],\n",
      "        [154, 131, 150]],\n",
      "\n",
      "       [[175, 152, 179],\n",
      "        [181, 155, 183],\n",
      "        [187, 162, 188],\n",
      "        ...,\n",
      "        [175, 152, 180],\n",
      "        [177, 151, 177],\n",
      "        [163, 139, 161]],\n",
      "\n",
      "       [[180, 155, 184],\n",
      "        [181, 156, 184],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [181, 156, 184],\n",
      "        [176, 151, 179],\n",
      "        [169, 146, 169]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[172, 150, 175],\n",
      "        [170, 150, 175],\n",
      "        [172, 149, 177],\n",
      "        ...,\n",
      "        [165, 144, 168],\n",
      "        [144, 122, 150],\n",
      "        [122,  98, 123]],\n",
      "\n",
      "       [[165, 147, 169],\n",
      "        [172, 150, 176],\n",
      "        [168, 146, 174],\n",
      "        ...,\n",
      "        [156, 135, 162],\n",
      "        [135, 116, 144],\n",
      "        [ 94,  80,  88]],\n",
      "\n",
      "       [[161, 140, 162],\n",
      "        [167, 148, 170],\n",
      "        [170, 149, 176],\n",
      "        ...,\n",
      "        [149, 130, 157],\n",
      "        [123, 105, 126],\n",
      "        [ 72,  48,  49]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12a10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n",
      "ERROR:root:Error processing image {'id': 3, 'source': 'moles', 'path': None, 'width': 128, 'height': 128, 'img': array([[[172, 150, 174],\n",
      "        [182, 158, 185],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [176, 151, 181],\n",
      "        [172, 149, 173],\n",
      "        [154, 131, 150]],\n",
      "\n",
      "       [[175, 152, 179],\n",
      "        [181, 155, 183],\n",
      "        [187, 162, 188],\n",
      "        ...,\n",
      "        [175, 152, 180],\n",
      "        [177, 151, 177],\n",
      "        [163, 139, 161]],\n",
      "\n",
      "       [[180, 155, 184],\n",
      "        [181, 156, 184],\n",
      "        [185, 157, 186],\n",
      "        ...,\n",
      "        [181, 156, 184],\n",
      "        [176, 151, 179],\n",
      "        [169, 146, 169]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[172, 150, 175],\n",
      "        [170, 150, 175],\n",
      "        [172, 149, 177],\n",
      "        ...,\n",
      "        [165, 144, 168],\n",
      "        [144, 122, 150],\n",
      "        [122,  98, 123]],\n",
      "\n",
      "       [[165, 147, 169],\n",
      "        [172, 150, 176],\n",
      "        [168, 146, 174],\n",
      "        ...,\n",
      "        [156, 135, 162],\n",
      "        [135, 116, 144],\n",
      "        [ 94,  80,  88]],\n",
      "\n",
      "       [[161, 140, 162],\n",
      "        [167, 148, 170],\n",
      "        [170, 149, 176],\n",
      "        ...,\n",
      "        [149, 130, 157],\n",
      "        [123, 105, 126],\n",
      "        [ 72,  48,  49]]], dtype=uint8), 'shape': ('malignant', 3, (128, 128)), 'mask': array([[[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       ...,\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]],\n",
      "\n",
      "       [[0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        ...,\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0],\n",
      "        [0, 0, 0]]], dtype=uint8), 'extra': <__main__.Metadata object at 0x7f7c92d12a10>}\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n",
      "    use_mini_mask=config.USE_MINI_MASK)\n",
      "  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n",
      "    class_ids = class_ids[_idx]\n",
      "IndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mRemoteTraceback\u001b[0m                           Traceback (most recent call last)",
      "\u001b[0;31mRemoteTraceback\u001b[0m: \n\"\"\"\nTraceback (most recent call last):\n  File \"/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/multiprocessing/pool.py\", line 121, in worker\n    result = (True, func(*args, **kwds))\n  File \"/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/utils/data_utils.py\", line 650, in next_sample\n    return six.next(_SHARED_SEQUENCES[uid])\n  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1709, in data_generator\n    use_mini_mask=config.USE_MINI_MASK)\n  File \"/home/michael/Mask-RCNN-TF2/mrcnn/model.py\", line 1265, in load_image_gt\n    class_ids = class_ids[_idx]\nIndexError: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3\n\"\"\"",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_7171/3816800673.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     13\u001b[0m             \u001b[0mlearning_rate\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLEARNING_RATE\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m             \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m15\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 15\u001b[0;31m             layers='heads')\n\u001b[0m\u001b[1;32m     16\u001b[0m \u001b[0;31m# After all the layers\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0;31m#model.train(dataset_train, dataset_val,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, train_dataset, val_dataset, learning_rate, epochs, layers, augmentation, custom_callbacks, no_augmentation_sources)\u001b[0m\n\u001b[1;32m   2383\u001b[0m             \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m100\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2384\u001b[0m             \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2385\u001b[0;31m             \u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2386\u001b[0m         )\n\u001b[1;32m   2387\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/legacy/interfaces.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     89\u001b[0m                 warnings.warn('Update your `' + object_name + '` call to the ' +\n\u001b[1;32m     90\u001b[0m                               'Keras 2 API: ' + signature, stacklevel=2)\n\u001b[0;32m---> 91\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     92\u001b[0m         \u001b[0mwrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     93\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, validation_freq, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m   1730\u001b[0m             \u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1731\u001b[0m             \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mshuffle\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1732\u001b[0;31m             initial_epoch=initial_epoch)\n\u001b[0m\u001b[1;32m   1733\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1734\u001b[0m     \u001b[0;34m@\u001b[0m\u001b[0minterfaces\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlegacy_generator_methods_support\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/engine/training_generator.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(model, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, validation_freq, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m    183\u001b[0m             \u001b[0mbatch_index\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    184\u001b[0m             \u001b[0;32mwhile\u001b[0m \u001b[0msteps_done\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 185\u001b[0;31m                 \u001b[0mgenerator_output\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput_generator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    186\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    187\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgenerator_output\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'__len__'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/utils/data_utils.py\u001b[0m in \u001b[0;36mget\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    740\u001b[0m                     \u001b[0;34m\"`use_multiprocessing=False, workers > 1`.\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    741\u001b[0m                     \"For more information see issue #1638.\")\n\u001b[0;32m--> 742\u001b[0;31m             \u001b[0msix\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreraise\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexc_info\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/six.py\u001b[0m in \u001b[0;36mreraise\u001b[0;34m(tp, value, tb)\u001b[0m\n\u001b[1;32m    717\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mtb\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    718\u001b[0m                 \u001b[0;32mraise\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 719\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    720\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    721\u001b[0m             \u001b[0mvalue\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/utils/data_utils.py\u001b[0m in \u001b[0;36mget\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    709\u001b[0m                 \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    710\u001b[0m                     \u001b[0mfuture\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mqueue\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mblock\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 711\u001b[0;31m                     \u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfuture\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m30\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    712\u001b[0m                     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mqueue\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtask_done\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    713\u001b[0m                 \u001b[0;32mexcept\u001b[0m \u001b[0mmp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTimeoutError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/multiprocessing/pool.py\u001b[0m in \u001b[0;36mget\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m    655\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_value\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    656\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 657\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_value\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    658\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    659\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_set\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mIndexError\u001b[0m: boolean index did not match indexed array along dimension 0; dimension is 1 but corresponding boolean dimension is 3"
     ]
    }
   ],
   "source": [
    "# Create the MaskRCNN model\n",
    "model = modellib.MaskRCNN(mode=\"training\", config=config, model_dir=MODEL_DIR)\n",
    "\n",
    "# Use as start point the coco model\n",
    "model.load_weights(COCO_MODEL_PATH, by_name=True,\n",
    "                   exclude=[\"mrcnn_class_logits\", \"mrcnn_bbox_fc\",\n",
    "                            \"mrcnn_bbox\", \"mrcnn_mask\"])\n",
    "\n",
    "# Train the model on the train dataset\n",
    "# First only the header layers\n",
    "model.log_dir = './logs/train'\n",
    "model.train(dataset_train, dataset_val,\n",
    "            learning_rate=config.LEARNING_RATE,\n",
    "            epochs=15,\n",
    "            layers='heads')\n",
    "# After all the layers \n",
    "#model.train(dataset_train, dataset_val,\n",
    "#            learning_rate=config.LEARNING_RATE/10,\n",
    "#            epochs=25,\n",
    " #           layers=\"all\")\n",
    "\n",
    "#print(\"Trained finished!\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "798a4983",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = modellib.MaskRCNN(mode=\"training\", model_dir=MODEL_DIR, config=config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "52bc3bf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.load_weights(COCO_MODEL_PATH, by_name=True,\n",
    "                   exclude=[\"mrcnn_class_logits\", \"mrcnn_bbox_fc\",\n",
    "                            \"mrcnn_bbox\", \"mrcnn_mask\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "1d64f9f4",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Starting at epoch 0. LR=0.0001\n",
      "\n",
      "Checkpoint Path: //logdir//train/mask_rcnn_moles_{epoch:04d}.h5\n",
      "Selecting layers to train\n",
      "conv1                  (Conv2D)\n",
      "bn_conv1               (BatchNorm)\n",
      "res2a_branch2a         (Conv2D)\n",
      "bn2a_branch2a          (BatchNorm)\n",
      "res2a_branch2b         (Conv2D)\n",
      "bn2a_branch2b          (BatchNorm)\n",
      "res2a_branch2c         (Conv2D)\n",
      "res2a_branch1          (Conv2D)\n",
      "bn2a_branch2c          (BatchNorm)\n",
      "bn2a_branch1           (BatchNorm)\n",
      "res2b_branch2a         (Conv2D)\n",
      "bn2b_branch2a          (BatchNorm)\n",
      "res2b_branch2b         (Conv2D)\n",
      "bn2b_branch2b          (BatchNorm)\n",
      "res2b_branch2c         (Conv2D)\n",
      "bn2b_branch2c          (BatchNorm)\n",
      "res2c_branch2a         (Conv2D)\n",
      "bn2c_branch2a          (BatchNorm)\n",
      "res2c_branch2b         (Conv2D)\n",
      "bn2c_branch2b          (BatchNorm)\n",
      "res2c_branch2c         (Conv2D)\n",
      "bn2c_branch2c          (BatchNorm)\n",
      "res3a_branch2a         (Conv2D)\n",
      "bn3a_branch2a          (BatchNorm)\n",
      "res3a_branch2b         (Conv2D)\n",
      "bn3a_branch2b          (BatchNorm)\n",
      "res3a_branch2c         (Conv2D)\n",
      "res3a_branch1          (Conv2D)\n",
      "bn3a_branch2c          (BatchNorm)\n",
      "bn3a_branch1           (BatchNorm)\n",
      "res3b_branch2a         (Conv2D)\n",
      "bn3b_branch2a          (BatchNorm)\n",
      "res3b_branch2b         (Conv2D)\n",
      "bn3b_branch2b          (BatchNorm)\n",
      "res3b_branch2c         (Conv2D)\n",
      "bn3b_branch2c          (BatchNorm)\n",
      "res3c_branch2a         (Conv2D)\n",
      "bn3c_branch2a          (BatchNorm)\n",
      "res3c_branch2b         (Conv2D)\n",
      "bn3c_branch2b          (BatchNorm)\n",
      "res3c_branch2c         (Conv2D)\n",
      "bn3c_branch2c          (BatchNorm)\n",
      "res3d_branch2a         (Conv2D)\n",
      "bn3d_branch2a          (BatchNorm)\n",
      "res3d_branch2b         (Conv2D)\n",
      "bn3d_branch2b          (BatchNorm)\n",
      "res3d_branch2c         (Conv2D)\n",
      "bn3d_branch2c          (BatchNorm)\n",
      "res4a_branch2a         (Conv2D)\n",
      "bn4a_branch2a          (BatchNorm)\n",
      "res4a_branch2b         (Conv2D)\n",
      "bn4a_branch2b          (BatchNorm)\n",
      "res4a_branch2c         (Conv2D)\n",
      "res4a_branch1          (Conv2D)\n",
      "bn4a_branch2c          (BatchNorm)\n",
      "bn4a_branch1           (BatchNorm)\n",
      "res4b_branch2a         (Conv2D)\n",
      "bn4b_branch2a          (BatchNorm)\n",
      "res4b_branch2b         (Conv2D)\n",
      "bn4b_branch2b          (BatchNorm)\n",
      "res4b_branch2c         (Conv2D)\n",
      "bn4b_branch2c          (BatchNorm)\n",
      "res4c_branch2a         (Conv2D)\n",
      "bn4c_branch2a          (BatchNorm)\n",
      "res4c_branch2b         (Conv2D)\n",
      "bn4c_branch2b          (BatchNorm)\n",
      "res4c_branch2c         (Conv2D)\n",
      "bn4c_branch2c          (BatchNorm)\n",
      "res4d_branch2a         (Conv2D)\n",
      "bn4d_branch2a          (BatchNorm)\n",
      "res4d_branch2b         (Conv2D)\n",
      "bn4d_branch2b          (BatchNorm)\n",
      "res4d_branch2c         (Conv2D)\n",
      "bn4d_branch2c          (BatchNorm)\n",
      "res4e_branch2a         (Conv2D)\n",
      "bn4e_branch2a          (BatchNorm)\n",
      "res4e_branch2b         (Conv2D)\n",
      "bn4e_branch2b          (BatchNorm)\n",
      "res4e_branch2c         (Conv2D)\n",
      "bn4e_branch2c          (BatchNorm)\n",
      "res4f_branch2a         (Conv2D)\n",
      "bn4f_branch2a          (BatchNorm)\n",
      "res4f_branch2b         (Conv2D)\n",
      "bn4f_branch2b          (BatchNorm)\n",
      "res4f_branch2c         (Conv2D)\n",
      "bn4f_branch2c          (BatchNorm)\n",
      "res4g_branch2a         (Conv2D)\n",
      "bn4g_branch2a          (BatchNorm)\n",
      "res4g_branch2b         (Conv2D)\n",
      "bn4g_branch2b          (BatchNorm)\n",
      "res4g_branch2c         (Conv2D)\n",
      "bn4g_branch2c          (BatchNorm)\n",
      "res4h_branch2a         (Conv2D)\n",
      "bn4h_branch2a          (BatchNorm)\n",
      "res4h_branch2b         (Conv2D)\n",
      "bn4h_branch2b          (BatchNorm)\n",
      "res4h_branch2c         (Conv2D)\n",
      "bn4h_branch2c          (BatchNorm)\n",
      "res4i_branch2a         (Conv2D)\n",
      "bn4i_branch2a          (BatchNorm)\n",
      "res4i_branch2b         (Conv2D)\n",
      "bn4i_branch2b          (BatchNorm)\n",
      "res4i_branch2c         (Conv2D)\n",
      "bn4i_branch2c          (BatchNorm)\n",
      "res4j_branch2a         (Conv2D)\n",
      "bn4j_branch2a          (BatchNorm)\n",
      "res4j_branch2b         (Conv2D)\n",
      "bn4j_branch2b          (BatchNorm)\n",
      "res4j_branch2c         (Conv2D)\n",
      "bn4j_branch2c          (BatchNorm)\n",
      "res4k_branch2a         (Conv2D)\n",
      "bn4k_branch2a          (BatchNorm)\n",
      "res4k_branch2b         (Conv2D)\n",
      "bn4k_branch2b          (BatchNorm)\n",
      "res4k_branch2c         (Conv2D)\n",
      "bn4k_branch2c          (BatchNorm)\n",
      "res4l_branch2a         (Conv2D)\n",
      "bn4l_branch2a          (BatchNorm)\n",
      "res4l_branch2b         (Conv2D)\n",
      "bn4l_branch2b          (BatchNorm)\n",
      "res4l_branch2c         (Conv2D)\n",
      "bn4l_branch2c          (BatchNorm)\n",
      "res4m_branch2a         (Conv2D)\n",
      "bn4m_branch2a          (BatchNorm)\n",
      "res4m_branch2b         (Conv2D)\n",
      "bn4m_branch2b          (BatchNorm)\n",
      "res4m_branch2c         (Conv2D)\n",
      "bn4m_branch2c          (BatchNorm)\n",
      "res4n_branch2a         (Conv2D)\n",
      "bn4n_branch2a          (BatchNorm)\n",
      "res4n_branch2b         (Conv2D)\n",
      "bn4n_branch2b          (BatchNorm)\n",
      "res4n_branch2c         (Conv2D)\n",
      "bn4n_branch2c          (BatchNorm)\n",
      "res4o_branch2a         (Conv2D)\n",
      "bn4o_branch2a          (BatchNorm)\n",
      "res4o_branch2b         (Conv2D)\n",
      "bn4o_branch2b          (BatchNorm)\n",
      "res4o_branch2c         (Conv2D)\n",
      "bn4o_branch2c          (BatchNorm)\n",
      "res4p_branch2a         (Conv2D)\n",
      "bn4p_branch2a          (BatchNorm)\n",
      "res4p_branch2b         (Conv2D)\n",
      "bn4p_branch2b          (BatchNorm)\n",
      "res4p_branch2c         (Conv2D)\n",
      "bn4p_branch2c          (BatchNorm)\n",
      "res4q_branch2a         (Conv2D)\n",
      "bn4q_branch2a          (BatchNorm)\n",
      "res4q_branch2b         (Conv2D)\n",
      "bn4q_branch2b          (BatchNorm)\n",
      "res4q_branch2c         (Conv2D)\n",
      "bn4q_branch2c          (BatchNorm)\n",
      "res4r_branch2a         (Conv2D)\n",
      "bn4r_branch2a          (BatchNorm)\n",
      "res4r_branch2b         (Conv2D)\n",
      "bn4r_branch2b          (BatchNorm)\n",
      "res4r_branch2c         (Conv2D)\n",
      "bn4r_branch2c          (BatchNorm)\n",
      "res4s_branch2a         (Conv2D)\n",
      "bn4s_branch2a          (BatchNorm)\n",
      "res4s_branch2b         (Conv2D)\n",
      "bn4s_branch2b          (BatchNorm)\n",
      "res4s_branch2c         (Conv2D)\n",
      "bn4s_branch2c          (BatchNorm)\n",
      "res4t_branch2a         (Conv2D)\n",
      "bn4t_branch2a          (BatchNorm)\n",
      "res4t_branch2b         (Conv2D)\n",
      "bn4t_branch2b          (BatchNorm)\n",
      "res4t_branch2c         (Conv2D)\n",
      "bn4t_branch2c          (BatchNorm)\n",
      "res4u_branch2a         (Conv2D)\n",
      "bn4u_branch2a          (BatchNorm)\n",
      "res4u_branch2b         (Conv2D)\n",
      "bn4u_branch2b          (BatchNorm)\n",
      "res4u_branch2c         (Conv2D)\n",
      "bn4u_branch2c          (BatchNorm)\n",
      "res4v_branch2a         (Conv2D)\n",
      "bn4v_branch2a          (BatchNorm)\n",
      "res4v_branch2b         (Conv2D)\n",
      "bn4v_branch2b          (BatchNorm)\n",
      "res4v_branch2c         (Conv2D)\n",
      "bn4v_branch2c          (BatchNorm)\n",
      "res4w_branch2a         (Conv2D)\n",
      "bn4w_branch2a          (BatchNorm)\n",
      "res4w_branch2b         (Conv2D)\n",
      "bn4w_branch2b          (BatchNorm)\n",
      "res4w_branch2c         (Conv2D)\n",
      "bn4w_branch2c          (BatchNorm)\n",
      "res5a_branch2a         (Conv2D)\n",
      "bn5a_branch2a          (BatchNorm)\n",
      "res5a_branch2b         (Conv2D)\n",
      "bn5a_branch2b          (BatchNorm)\n",
      "res5a_branch2c         (Conv2D)\n",
      "res5a_branch1          (Conv2D)\n",
      "bn5a_branch2c          (BatchNorm)\n",
      "bn5a_branch1           (BatchNorm)\n",
      "res5b_branch2a         (Conv2D)\n",
      "bn5b_branch2a          (BatchNorm)\n",
      "res5b_branch2b         (Conv2D)\n",
      "bn5b_branch2b          (BatchNorm)\n",
      "res5b_branch2c         (Conv2D)\n",
      "bn5b_branch2c          (BatchNorm)\n",
      "res5c_branch2a         (Conv2D)\n",
      "bn5c_branch2a          (BatchNorm)\n",
      "res5c_branch2b         (Conv2D)\n",
      "bn5c_branch2b          (BatchNorm)\n",
      "res5c_branch2c         (Conv2D)\n",
      "bn5c_branch2c          (BatchNorm)\n",
      "fpn_c5p5               (Conv2D)\n",
      "fpn_c4p4               (Conv2D)\n",
      "fpn_c3p3               (Conv2D)\n",
      "fpn_c2p2               (Conv2D)\n",
      "fpn_p5                 (Conv2D)\n",
      "fpn_p2                 (Conv2D)\n",
      "fpn_p3                 (Conv2D)\n",
      "fpn_p4                 (Conv2D)\n",
      "In model:  rpn_model\n",
      "    rpn_conv_shared        (Conv2D)\n",
      "    rpn_class_raw          (Conv2D)\n",
      "    rpn_bbox_pred          (Conv2D)\n",
      "mrcnn_mask_conv1       (TimeDistributed)\n",
      "mrcnn_mask_bn1         (TimeDistributed)\n",
      "mrcnn_mask_conv2       (TimeDistributed)\n",
      "mrcnn_mask_bn2         (TimeDistributed)\n",
      "mrcnn_class_conv1      (TimeDistributed)\n",
      "mrcnn_class_bn1        (TimeDistributed)\n",
      "mrcnn_mask_conv3       (TimeDistributed)\n",
      "mrcnn_mask_bn3         (TimeDistributed)\n",
      "mrcnn_class_conv2      (TimeDistributed)\n",
      "mrcnn_class_bn2        (TimeDistributed)\n",
      "mrcnn_mask_conv4       (TimeDistributed)\n",
      "mrcnn_mask_bn4         (TimeDistributed)\n",
      "mrcnn_bbox_fc          (TimeDistributed)\n",
      "mrcnn_mask_deconv      (TimeDistributed)\n",
      "mrcnn_class_logits     (TimeDistributed)\n",
      "mrcnn_mask             (TimeDistributed)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-02-10 19:57:21.062157: I tensorflow/core/profiler/lib/profiler_session.cc:159] Profiler session started.\n",
      "Exception ignored in: <function _EagerDefinedFunctionDeleter.__del__ at 0x7f5e5b3175f0>\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/eager/function.py\", line 434, in __del__\n",
      "    context.remove_function(self.name)\n",
      "  File \"/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/eager/context.py\", line 2237, in remove_function\n",
      "    context().remove_function(name)\n",
      "  File \"/home/michael/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/eager/context.py\", line 1105, in remove_function\n",
      "    pywrap_tfe.TFE_ContextRemoveFunction(self._handle, name)\n",
      "KeyboardInterrupt: \n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_3522/3357500426.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      2\u001b[0m model.train(dataset_train, dataset_val,\n\u001b[1;32m      3\u001b[0m             \u001b[0mlearning_rate\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLEARNING_RATE\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m             epochs=25,layers=\"all\")\n\u001b[0m",
      "\u001b[0;32m~/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, train_dataset, val_dataset, learning_rate, epochs, layers, augmentation, custom_callbacks, no_augmentation_sources)\u001b[0m\n\u001b[1;32m   2363\u001b[0m         \u001b[0mlog\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Checkpoint Path: {}\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcheckpoint_path\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2364\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_trainable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlayers\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2365\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlearning_rate\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLEARNING_MOMENTUM\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2366\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2367\u001b[0m         \u001b[0;31m# Work-around for Windows: Keras fails on Windows when using\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mcompile\u001b[0;34m(self, learning_rate, momentum)\u001b[0m\n\u001b[1;32m   2205\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeras_model\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics_names\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2206\u001b[0m             loss = (\n\u001b[0;32m-> 2207\u001b[0;31m                 \u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreduce_mean\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkeepdims\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2208\u001b[0m                 * self.config.LOSS_WEIGHTS.get(name, 1.))\n\u001b[1;32m   2209\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeras_model\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics_tensors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    178\u001b[0m     \u001b[0;34m\"\"\"Call target, and fall back on dispatchers if there is a TypeError.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    179\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 180\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    181\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mTypeError\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    182\u001b[0m       \u001b[0;31m# Note: convert_to_eager_tensor currently raises a ValueError, not a\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/ops/math_ops.py\u001b[0m in \u001b[0;36mreduce_mean\u001b[0;34m(input_tensor, axis, keepdims, name)\u001b[0m\n\u001b[1;32m   2064\u001b[0m       gen_math_ops.mean(\n\u001b[1;32m   2065\u001b[0m           \u001b[0minput_tensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_ReductionDims\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkeepdims\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2066\u001b[0;31m           name=name))\n\u001b[0m\u001b[1;32m   2067\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2068\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/ops/gen_math_ops.py\u001b[0m in \u001b[0;36mmean\u001b[0;34m(input, axis, keep_dims, name)\u001b[0m\n\u001b[1;32m   5814\u001b[0m   _, _, _op, _outputs = _op_def_library._apply_op_helper(\n\u001b[1;32m   5815\u001b[0m         \u001b[0;34m\"Mean\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreduction_indices\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maxis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkeep_dims\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkeep_dims\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 5816\u001b[0;31m                 name=name)\n\u001b[0m\u001b[1;32m   5817\u001b[0m   \u001b[0m_result\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_outputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   5818\u001b[0m   \u001b[0;32mif\u001b[0m \u001b[0m_execute\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmust_record_gradient\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/framework/op_def_library.py\u001b[0m in \u001b[0;36m_apply_op_helper\u001b[0;34m(op_type_name, name, **keywords)\u001b[0m\n\u001b[1;32m    351\u001b[0m   \u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    352\u001b[0m   \u001b[0minput_types\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 353\u001b[0;31m   \u001b[0;32mwith\u001b[0m \u001b[0mg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_default\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mops\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname_scope\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mscope\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    354\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    355\u001b[0m     \u001b[0;31m# Perform input type inference\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/framework/ops.py\u001b[0m in \u001b[0;36m__enter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   6294\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   6295\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_name_scope\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname_scope\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 6296\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_name_scope\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__enter__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   6297\u001b[0m     \u001b[0;32mexcept\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   6298\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_g_manager\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/contextlib.py\u001b[0m in \u001b[0;36m__enter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    110\u001b[0m         \u001b[0;32mdel\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    111\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 112\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mnext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgen\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    113\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mStopIteration\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    114\u001b[0m             \u001b[0;32mraise\u001b[0m \u001b[0mRuntimeError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"generator didn't yield\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/tensorflow/python/framework/ops.py\u001b[0m in \u001b[0;36mname_scope\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m   4029\u001b[0m         \u001b[0;31m# Scopes created in the root must match the more restrictive\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4030\u001b[0m         \u001b[0;31m# op name regex, which constrains the initial character.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 4031\u001b[0;31m         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0m_VALID_OP_NAME_REGEX\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   4032\u001b[0m           \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"'%s' is not a valid scope name\"\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4033\u001b[0m     \u001b[0mold_stack\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_name_stack\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "\n",
    "model.train(dataset_train, dataset_val,\n",
    "            learning_rate=config.LEARNING_RATE,\n",
    "            epochs=25,layers=\"all\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2f0ce80a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import cv2\n",
    "import json\n",
    "import glob\n",
    "\n",
    "from mrcnn.config import Config\n",
    "import mrcnn.utils as utils\n",
    "import mrcnn.model as modellib\n",
    "import mrcnn.visualize as visualize\n",
    "\n",
    "np.set_printoptions(threshold=np.inf)\n",
    "\n",
    "# path of the trained model\n",
    "dir_path = './'\n",
    "MODEL_DIR = dir_path + \"/models/\"\n",
    "MODEL_PATH = MODEL_DIR +\"/mask_rcnn_coco.h5\"\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "7fe66417",
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Layer #389 (named \"mrcnn_bbox_fc\"), weight <tf.Variable 'mrcnn_bbox_fc_5/kernel:0' shape=(1024, 12) dtype=float32, numpy=\narray([[-0.0438811 , -0.04495583,  0.03014846, ..., -0.00831714,\n        -0.07000256,  0.05572653],\n       [-0.03052835, -0.05996221, -0.06562172, ..., -0.03630493,\n         0.06238019,  0.01063577],\n       [ 0.00787228, -0.00513127, -0.01102631, ..., -0.02811338,\n         0.04203645,  0.04163819],\n       ...,\n       [-0.01048563, -0.01884444,  0.06367481, ...,  0.06031197,\n        -0.04036516,  0.06056623],\n       [ 0.06560457, -0.04252248,  0.0632337 , ..., -0.06357488,\n         0.01784682, -0.06170159],\n       [-0.03592445, -0.03315556,  0.04010619, ...,  0.07032625,\n        -0.02199942,  0.04615058]], dtype=float32)> has shape (1024, 12), but the saved weight has shape (1024, 324).",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_5327/2559863039.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0;31m# take the trained model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodellib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mMaskRCNN\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"inference\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mMODEL_DIR\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 19\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_weights\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mCOCO_MODEL_PATH\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mby_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m \u001b[0;31m# background + (malignant , benign)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Mask-RCNN-TF2/mrcnn/model.py\u001b[0m in \u001b[0;36mload_weights\u001b[0;34m(self, filepath, by_name, exclude)\u001b[0m\n\u001b[1;32m   2128\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2129\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mby_name\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2130\u001b[0;31m             \u001b[0msaving\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_weights_from_hdf5_group_by_name\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlayers\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2131\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2132\u001b[0m             \u001b[0msaving\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_weights_from_hdf5_group\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlayers\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/MaskRCNNTF2/lib/python3.7/site-packages/keras/engine/saving.py\u001b[0m in \u001b[0;36mload_weights_from_hdf5_group_by_name\u001b[0;34m(f, layers, skip_mismatch, reshape)\u001b[0m\n\u001b[1;32m   1326\u001b[0m                                          \u001b[0;34m' has shape {}'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msymbolic_shape\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1327\u001b[0m                                          \u001b[0;34m', but the saved weight has shape '\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1328\u001b[0;31m                                          str(weight_values[i].shape) + '.')\n\u001b[0m\u001b[1;32m   1329\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1330\u001b[0m                     weight_value_tuples.append((symbolic_weights[i],\n",
      "\u001b[0;31mValueError\u001b[0m: Layer #389 (named \"mrcnn_bbox_fc\"), weight <tf.Variable 'mrcnn_bbox_fc_5/kernel:0' shape=(1024, 12) dtype=float32, numpy=\narray([[-0.0438811 , -0.04495583,  0.03014846, ..., -0.00831714,\n        -0.07000256,  0.05572653],\n       [-0.03052835, -0.05996221, -0.06562172, ..., -0.03630493,\n         0.06238019,  0.01063577],\n       [ 0.00787228, -0.00513127, -0.01102631, ..., -0.02811338,\n         0.04203645,  0.04163819],\n       ...,\n       [-0.01048563, -0.01884444,  0.06367481, ...,  0.06031197,\n        -0.04036516,  0.06056623],\n       [ 0.06560457, -0.04252248,  0.0632337 , ..., -0.06357488,\n         0.01784682, -0.06170159],\n       [-0.03592445, -0.03315556,  0.04010619, ...,  0.07032625,\n        -0.02199942,  0.04615058]], dtype=float32)> has shape (1024, 12), but the saved weight has shape (1024, 324)."
     ]
    }
   ],
   "source": [
    "class CocoConfig(Config):\n",
    "    ''' \n",
    "    MolesConfig:\n",
    "        Contain the configuration for the dataset + those in Config\n",
    "    '''\n",
    "    NAME = \"moles\"\n",
    "    NUM_CLASSES = 1 + 2 # background + (malignant , benign)\n",
    "    IMAGE_MIN_DIM = 128\n",
    "    IMAGE_MAX_DIM = 128\n",
    "    GPU_COUNT = 1\n",
    "    IMAGES_PER_GPU = 1\n",
    "    DETECTION_MAX_INSTANCES = 3\n",
    "\n",
    "# create and instance of config\n",
    "config = CocoConfig()\n",
    "\n",
    "# take the trained model\n",
    "model = modellib.MaskRCNN(mode=\"inference\", model_dir=MODEL_DIR, config=config)\n",
    "model.load_weights(COCO_MODEL_PATH, by_name=True)\n",
    "\n",
    "# background + (malignant , benign)\n",
    "class_names = [\"BG\", \"malignant\", \"benign\"]\n",
    "'''\n",
    "img = cv2.imread(dir_path +\"/ISIC_0034262.jpg\")\n",
    "img = cv2.resize(img, (128, 128))\n",
    "if img.all():\n",
    "   print(\"No Image\")\n",
    "#ground truth of the class\n",
    "print(data[\"meta\"][\"clinical\"][\"benign_malignant\"])\n",
    "''' \n",
    "\n",
    "# predict the mask, bounding box and class of the image\n",
    "for file in os.listdir(dir_path +\"/test\"):\n",
    "    print(file)\n",
    "    img = cv2.imread(dir_path +\"/test/\"+file)\n",
    "    img = cv2.resize(img, (128, 128))\n",
    "    r = model.detect([img])[0]\n",
    "    print(r['class_ids'] , class_names , r['scores'])\n",
    "    visualize.display_instances(img, r['rois'], r['masks'], r['class_ids'],\n",
    "                                class_names, r['scores'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d6b22b3f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}