Classification3(Satelite images).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "3b063df0-873d-452d-aa08-4dbff0be51f4",
   "metadata": {},
   "source": [
    "# Satelite image 분류\n",
    "Satellite Remote Sensing Image -RSI-CB256 from https://www.kaggle.com/mahmoudreda55/satellite-image-classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "13816377-68ac-40fd-9c9d-1887dc06849e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 4506 images belonging to 4 classes.\n",
      "Found 4506 images belonging to 4 classes.\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "import tensorflow as tf\n",
    "train_datagen = ImageDataGenerator(rescale = 1./255)\n",
    "test_datagen = ImageDataGenerator(rescale = 1./255)\n",
    "import numpy as np\n",
    "from PIL import Image\n",
    "generator = ImageDataGenerator(rescale=1./255, validation_split = 0.2)\n",
    "directory_train =  '../Suhyun/satelite/data'\n",
    "\n",
    "\n",
    "train_dataset = generator.flow_from_directory(batch_size=128,\n",
    "                                                 directory = directory_train,\n",
    "                                                 shuffle=True,\n",
    "                                                 target_size=(224, 224), \n",
    "                                                 subset=\"training\",\n",
    "                                                 class_mode='categorical')\n",
    "\n",
    "\n",
    "valid_dataset = generator.flow_from_directory(batch_size=128,\n",
    "                                                 directory = directory_train,\n",
    "                                                 shuffle=True,\n",
    "                                                 target_size=(224, 224), \n",
    "                                                 subset=\"training\",\n",
    "                                                 class_mode='categorical')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6fc803c3-0038-4758-abd7-669e7a8a0df6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "physical_devices = tf.config.list_physical_devices('GPU')\n",
    "if physical_devices:\n",
    "  tf.config.experimental.set_memory_growth(physical_devices[0], True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3f36ff29-d8a3-4cee-bf7a-000d8c68d494",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "fig,ax = plt.subplots(nrows=1,ncols=10,figsize=(20,20))\n",
    "for i in range(10):\n",
    "    ax[i].imshow(train_dataset[0][0][i+10])\n",
    "    ax[i].axis('off')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "65c0443d-bfdf-4df1-be8e-55dce465daff",
   "metadata": {},
   "outputs": [],
   "source": [
    "def hard_swish(features):\n",
    "\n",
    "  \n",
    "  features = tf.convert_to_tensor(features)\n",
    "  fdtype = features.dtype\n",
    "  return features * tf.nn.relu6(features + tf.cast(3., fdtype)) * (1. / 6.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2157eb66-3576-4d42-91a2-dd0a8b4f1356",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def se_block(inputs, ch, ratio=16):\n",
    "\n",
    "    x = tf.keras.layers.GlobalAveragePooling2D()(inputs)\n",
    "    x = tf.keras.layers.Dense(ch//ratio, activation='relu')(x)\n",
    "    y = tf.keras.layers.Dense(ch, activation='sigmoid')(x)\n",
    "    y = tf.keras.layers.multiply([inputs, y])\n",
    "    out = hard_swish(y)\n",
    "    return out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d8bf2a62-5cd4-4e18-b21d-c0a486aa3e16",
   "metadata": {},
   "outputs": [],
   "source": [
    "def dep_bn(inputs, filter : int, kernel :int, stride: int):\n",
    "\n",
    "  y = tf.keras.layers.Conv2D(filters = filter, kernel_size= kernel, strides= stride, padding= 'same')(inputs)\n",
    "  bn = tf.keras.layers.BatchNormalization(axis = 1)(y)\n",
    "  out = tf.nn.relu6(bn)\n",
    "  \n",
    "  return out\n",
    "\n",
    "\n",
    "def Inverted_residual_SEblock(x,  hs: bool, se: bool , filters, kernel, strides,  expansion_ratio ):\n",
    "\n",
    "\n",
    "    #pointwise\n",
    "    y = tf.keras.layers.Conv2D(filters = x.get_shape().as_list()[-1]*expansion_ratio, kernel_size=  1, strides = 1, padding= 'same')(x)\n",
    "    bn = tf.keras.layers.BatchNormalization(axis = 1)(y)\n",
    "    if hs:\n",
    "        y_p = hard_swish(bn)\n",
    "    else:\n",
    "        y_p = tf.nn.relu6(bn)\n",
    "    \n",
    "    \n",
    "    #depthwise\n",
    "    y = tf.keras.layers.DepthwiseConv2D(kernel_size=  kernel, strides= strides, padding= 'same', depth_multiplier = 1)(y_p)\n",
    "    bn = tf.keras.layers.BatchNormalization(axis = 1)(y)\n",
    "    if hs:\n",
    "        out = hard_swish(bn)\n",
    "    else:\n",
    "        out = tf.nn.relu6(bn)\n",
    "   \n",
    "    \n",
    "    if se: \n",
    "        channel = out.get_shape().as_list()[-1]\n",
    "        out = se_block(out, channel)\n",
    "\n",
    "    \n",
    "    #pointwise\n",
    "    y = tf.keras.layers.Conv2D(filters = filters, kernel_size=  1, strides = 1, padding= 'same')(out)\n",
    "    bn = tf.keras.layers.BatchNormalization(axis = 1)(y)\n",
    "    if hs:\n",
    "        y_p = hard_swish(bn)\n",
    "    else:\n",
    "        y_p = tf.nn.relu6(bn)\n",
    "        \n",
    "        \n",
    "    \n",
    "    if strides == 1:\n",
    "        if x.shape[3] != filters : \n",
    "           x = tf.keras.layers.Conv2D(filters = filters, kernel_size = 1, strides = strides, padding = \"same\")(x)\n",
    "        out  = tf.keras.layers.Add()([x, y_p])\n",
    "   \n",
    "    \n",
    "\n",
    "    return y\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "76efa1e0-4fab-4661-827e-aef6f6cd79cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_mobilenetV3(inputs):\n",
    "\n",
    "   x = dep_bn(inputs, 16, 3, 2)\n",
    "   x = Inverted_residual_SEblock(x, False, False, 16, 3, 1,1)\n",
    "\n",
    "   x = Inverted_residual_SEblock(x, False, False, 24, 3, 2, 4)\n",
    "   x = Inverted_residual_SEblock(x, False, False, 24, 3, 1, 3)\n",
    "   \n",
    "   x = Inverted_residual_SEblock(x, False, True, 40, 5, 2, 3)\n",
    "   x = Inverted_residual_SEblock(x, False, True, 40, 5, 1, 3)\n",
    "   x = Inverted_residual_SEblock(x, False, True, 40, 5, 1, 3)\n",
    "\n",
    "   x = Inverted_residual_SEblock(x, True, False, 80, 3, 2, 6)\n",
    "   x = Inverted_residual_SEblock(x, True, False, 80, 3, 1, 2.5)\n",
    "   x = Inverted_residual_SEblock(x, True, False, 80, 3, 1, 184/80)\n",
    "   x = Inverted_residual_SEblock(x, True, False, 80, 3, 1, 184/80)\n",
    "    \n",
    "   x = Inverted_residual_SEblock(x, True, True, 112, 3, 1, 6)\n",
    "   x = Inverted_residual_SEblock(x, True, True, 112, 3, 1, 6)\n",
    "    \n",
    "   x = Inverted_residual_SEblock(x, True, True, 160, 5, 2, 6)\n",
    "   \n",
    "   x = Inverted_residual_SEblock(x, True, True, 160, 5, 1, 6)\n",
    "   x = Inverted_residual_SEblock(x, True, True, 160, 5, 1, 6)\n",
    "   x = dep_bn(x, 960, 1, 1)\n",
    "   \n",
    "   t = tf.keras.layers.AveragePooling2D(1)(x)\n",
    "   t = tf.keras.layers.Flatten()(t)\n",
    "   outputs = tf.keras.layers.Dense(4, activation='softmax')(t)\n",
    "   return outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2411cc3f-9f3f-444e-a2bf-213b0b98804d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TensorShape([None, 4])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputs = tf.keras.Input(shape=(224,224,3))\n",
    "outputs = create_mobilenetV3(inputs)\n",
    "outputs.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3a12258d-9639-44a6-84f6-12daca3e3f21",
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = tf.keras.Input(shape=(224,224,3))\n",
    "outputs = create_mobilenetV3(inputs)\n",
    "\n",
    "model = tf.keras.Model(inputs, outputs)\n",
    "\n",
    "opt = tf.keras.optimizers.Adam(learning_rate=0.001)\n",
    "model.compile(\n",
    "        optimizer=opt,\n",
    "        loss='categorical_crossentropy',\n",
    "        metrics=['accuracy']\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "4ba1050e-08df-470f-a528-a808487dd7dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "36/36 [==============================] - 23s 479ms/step - loss: 2.1636 - accuracy: 0.4015 - val_loss: 1.3906 - val_accuracy: 0.2663\n",
      "Epoch 2/50\n",
      "36/36 [==============================] - 16s 450ms/step - loss: 1.3786 - accuracy: 0.3691 - val_loss: 1.3811 - val_accuracy: 0.2663\n",
      "Epoch 3/50\n",
      "36/36 [==============================] - 16s 444ms/step - loss: 0.9732 - accuracy: 0.4587 - val_loss: 1.3804 - val_accuracy: 0.2663\n",
      "Epoch 4/50\n",
      "36/36 [==============================] - 16s 441ms/step - loss: 0.9475 - accuracy: 0.4949 - val_loss: 1.3933 - val_accuracy: 0.2663\n",
      "Epoch 5/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.8193 - accuracy: 0.4989 - val_loss: 1.4140 - val_accuracy: 0.2663\n",
      "Epoch 6/50\n",
      "36/36 [==============================] - 16s 444ms/step - loss: 0.8392 - accuracy: 0.5195 - val_loss: 1.4851 - val_accuracy: 0.2663\n",
      "Epoch 7/50\n",
      "36/36 [==============================] - 16s 447ms/step - loss: 0.7769 - accuracy: 0.5226 - val_loss: 1.6258 - val_accuracy: 0.2663\n",
      "Epoch 8/50\n",
      "36/36 [==============================] - 16s 448ms/step - loss: 0.8061 - accuracy: 0.5632 - val_loss: 1.6131 - val_accuracy: 0.2663\n",
      "Epoch 9/50\n",
      "36/36 [==============================] - 16s 446ms/step - loss: 0.9929 - accuracy: 0.5107 - val_loss: 1.6126 - val_accuracy: 0.2663\n",
      "Epoch 10/50\n",
      "36/36 [==============================] - 16s 445ms/step - loss: 1.1100 - accuracy: 0.4998 - val_loss: 1.5977 - val_accuracy: 0.2663\n",
      "Epoch 11/50\n",
      "36/36 [==============================] - 16s 448ms/step - loss: 0.7691 - accuracy: 0.5384 - val_loss: 2.4217 - val_accuracy: 0.2663\n",
      "Epoch 12/50\n",
      "36/36 [==============================] - 16s 447ms/step - loss: 0.8863 - accuracy: 0.4920 - val_loss: 2.7235 - val_accuracy: 0.2663\n",
      "Epoch 13/50\n",
      "36/36 [==============================] - 16s 441ms/step - loss: 0.7562 - accuracy: 0.5502 - val_loss: 3.2264 - val_accuracy: 0.2705\n",
      "Epoch 14/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.8694 - accuracy: 0.5102 - val_loss: 2.6620 - val_accuracy: 0.2663\n",
      "Epoch 15/50\n",
      "36/36 [==============================] - 16s 446ms/step - loss: 0.8320 - accuracy: 0.5300 - val_loss: 1.0979 - val_accuracy: 0.5322\n",
      "Epoch 16/50\n",
      "36/36 [==============================] - 16s 446ms/step - loss: 0.7398 - accuracy: 0.5335 - val_loss: 3.4219 - val_accuracy: 0.3367\n",
      "Epoch 17/50\n",
      "36/36 [==============================] - 16s 448ms/step - loss: 0.7197 - accuracy: 0.5644 - val_loss: 7.9168 - val_accuracy: 0.4982\n",
      "Epoch 18/50\n",
      "36/36 [==============================] - 16s 444ms/step - loss: 0.7694 - accuracy: 0.5439 - val_loss: 0.7181 - val_accuracy: 0.5195\n",
      "Epoch 19/50\n",
      "36/36 [==============================] - 16s 441ms/step - loss: 0.8521 - accuracy: 0.5213 - val_loss: 104.3991 - val_accuracy: 0.5346\n",
      "Epoch 20/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.9331 - accuracy: 0.5266 - val_loss: 68.2078 - val_accuracy: 0.5293\n",
      "Epoch 21/50\n",
      "36/36 [==============================] - 16s 441ms/step - loss: 0.7516 - accuracy: 0.5513 - val_loss: 33.5262 - val_accuracy: 0.5408\n",
      "Epoch 22/50\n",
      "36/36 [==============================] - 16s 439ms/step - loss: 0.7232 - accuracy: 0.5568 - val_loss: 7.7001 - val_accuracy: 0.5415\n",
      "Epoch 23/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.6829 - accuracy: 0.6092 - val_loss: 0.7671 - val_accuracy: 0.6070\n",
      "Epoch 24/50\n",
      "36/36 [==============================] - 16s 448ms/step - loss: 0.6628 - accuracy: 0.6389 - val_loss: 0.8121 - val_accuracy: 0.5719\n",
      "Epoch 25/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.6476 - accuracy: 0.6691 - val_loss: 0.6637 - val_accuracy: 0.6170\n",
      "Epoch 26/50\n",
      "36/36 [==============================] - 16s 448ms/step - loss: 0.6350 - accuracy: 0.6558 - val_loss: 0.9102 - val_accuracy: 0.5158\n",
      "Epoch 27/50\n",
      "36/36 [==============================] - 16s 446ms/step - loss: 0.6265 - accuracy: 0.6682 - val_loss: 0.8325 - val_accuracy: 0.5182\n",
      "Epoch 28/50\n",
      "36/36 [==============================] - 16s 447ms/step - loss: 0.6356 - accuracy: 0.6977 - val_loss: 3.8703 - val_accuracy: 0.2736\n",
      "Epoch 29/50\n",
      "36/36 [==============================] - 16s 450ms/step - loss: 0.6985 - accuracy: 0.6760 - val_loss: 0.6416 - val_accuracy: 0.6398\n",
      "Epoch 30/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.6921 - accuracy: 0.6367 - val_loss: 12.2395 - val_accuracy: 0.5706\n",
      "Epoch 31/50\n",
      "36/36 [==============================] - 16s 449ms/step - loss: 0.6448 - accuracy: 0.6338 - val_loss: 11.0450 - val_accuracy: 0.4854\n",
      "Epoch 32/50\n",
      "36/36 [==============================] - 16s 445ms/step - loss: 0.5599 - accuracy: 0.6842 - val_loss: 9.4868 - val_accuracy: 0.5189\n",
      "Epoch 33/50\n",
      "36/36 [==============================] - 16s 440ms/step - loss: 0.5131 - accuracy: 0.7301 - val_loss: 3.5471 - val_accuracy: 0.4470\n",
      "Epoch 34/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.5040 - accuracy: 0.7390 - val_loss: 3.9167 - val_accuracy: 0.4379\n",
      "Epoch 35/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.4983 - accuracy: 0.7415 - val_loss: 0.6613 - val_accuracy: 0.6744\n",
      "Epoch 36/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.5072 - accuracy: 0.7432 - val_loss: 1.3851 - val_accuracy: 0.6507\n",
      "Epoch 37/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.4830 - accuracy: 0.7565 - val_loss: 2.8002 - val_accuracy: 0.5348\n",
      "Epoch 38/50\n",
      "36/36 [==============================] - 16s 446ms/step - loss: 0.4723 - accuracy: 0.7605 - val_loss: 1.2650 - val_accuracy: 0.5504\n",
      "Epoch 39/50\n",
      "36/36 [==============================] - 16s 448ms/step - loss: 0.9275 - accuracy: 0.5672 - val_loss: 185.2067 - val_accuracy: 0.2665\n",
      "Epoch 40/50\n",
      "36/36 [==============================] - 16s 450ms/step - loss: 0.7801 - accuracy: 0.6067 - val_loss: 176.6229 - val_accuracy: 0.2663\n",
      "Epoch 41/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.7146 - accuracy: 0.6232 - val_loss: 19.1926 - val_accuracy: 0.4503\n",
      "Epoch 42/50\n",
      "36/36 [==============================] - 16s 440ms/step - loss: 0.6537 - accuracy: 0.6669 - val_loss: 54.9915 - val_accuracy: 0.5044\n",
      "Epoch 43/50\n",
      "36/36 [==============================] - 16s 444ms/step - loss: 0.6062 - accuracy: 0.6842 - val_loss: 51.4400 - val_accuracy: 0.5049\n",
      "Epoch 44/50\n",
      "36/36 [==============================] - 16s 450ms/step - loss: 0.5181 - accuracy: 0.7268 - val_loss: 0.6985 - val_accuracy: 0.6755\n",
      "Epoch 45/50\n",
      "36/36 [==============================] - 16s 447ms/step - loss: 0.5230 - accuracy: 0.7399 - val_loss: 1.1563 - val_accuracy: 0.5144\n",
      "Epoch 46/50\n",
      "36/36 [==============================] - 16s 446ms/step - loss: 0.6115 - accuracy: 0.6826 - val_loss: 1.6340 - val_accuracy: 0.5513\n",
      "Epoch 47/50\n",
      "36/36 [==============================] - 16s 455ms/step - loss: 0.4753 - accuracy: 0.7577 - val_loss: 1.1788 - val_accuracy: 0.5688\n",
      "Epoch 48/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.4612 - accuracy: 0.7810 - val_loss: 2.3614 - val_accuracy: 0.5843\n",
      "Epoch 49/50\n",
      "36/36 [==============================] - 16s 443ms/step - loss: 0.4857 - accuracy: 0.7545 - val_loss: 1.4476 - val_accuracy: 0.5104\n",
      "Epoch 50/50\n",
      "36/36 [==============================] - 16s 442ms/step - loss: 0.4797 - accuracy: 0.7477 - val_loss: 7.0176 - val_accuracy: 0.5917\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7fd4c8232ee0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(train_dataset, validation_data=valid_dataset, epochs = 50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c4978726-02ec-4a30-ab09-cd1efb67aa64",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = model.predict(train_dataset[0][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c95d34fb-7ff4-446c-a324-3b25b408ab1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def index_output(prediction) :\n",
    "    \n",
    "    index_list = []\n",
    "    for pred in prediction:\n",
    "        i = np.argmax(pred)\n",
    "        index_list.append(i)\n",
    "    return index_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8f61622c-be31-4c83-9df2-690be1a3e5b6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(128,)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = np.array(index_output(predictions))\n",
    "y_pred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "b515e953-7f7f-49ef-9c67-3a84612e255c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(128,)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_true =[]\n",
    "for y in train_dataset[0][1]:\n",
    "    Y_true.append(np.where(y==1))\n",
    "Y_true = np.squeeze(np.asarray(Y_true))\n",
    "Y_true.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "145009a5-f023-437a-a9bd-4c041ce05cc1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.58      0.94      0.72        34\n",
      "           1       0.00      0.00      0.00        23\n",
      "           2       0.56      0.97      0.71        34\n",
      "           3       0.79      0.30      0.43        37\n",
      "\n",
      "    accuracy                           0.59       128\n",
      "   macro avg       0.48      0.55      0.47       128\n",
      "weighted avg       0.53      0.59      0.50       128\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/sklearn/metrics/_classification.py:1248: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/usr/local/lib/python3.8/dist-packages/sklearn/metrics/_classification.py:1248: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/usr/local/lib/python3.8/dist-packages/sklearn/metrics/_classification.py:1248: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "print(classification_report(Y_true, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a493df8a-a278-4564-9cc7-44ab923b013b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}