chapter 2 Pre-training.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "956d1faa",
   "metadata": {},
   "source": [
    "# 1.加载预定义模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77c2abdd",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "14fca598",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['AlexNet',\n",
       " 'AlexNet_Weights',\n",
       " 'ConvNeXt',\n",
       " 'ConvNeXt_Base_Weights',\n",
       " 'ConvNeXt_Large_Weights',\n",
       " 'ConvNeXt_Small_Weights',\n",
       " 'ConvNeXt_Tiny_Weights',\n",
       " 'DenseNet',\n",
       " 'DenseNet121_Weights',\n",
       " 'DenseNet161_Weights',\n",
       " 'DenseNet169_Weights',\n",
       " 'DenseNet201_Weights',\n",
       " 'EfficientNet',\n",
       " 'EfficientNet_B0_Weights',\n",
       " 'EfficientNet_B1_Weights',\n",
       " 'EfficientNet_B2_Weights',\n",
       " 'EfficientNet_B3_Weights',\n",
       " 'EfficientNet_B4_Weights',\n",
       " 'EfficientNet_B5_Weights',\n",
       " 'EfficientNet_B6_Weights',\n",
       " 'EfficientNet_B7_Weights',\n",
       " 'EfficientNet_V2_L_Weights',\n",
       " 'EfficientNet_V2_M_Weights',\n",
       " 'EfficientNet_V2_S_Weights',\n",
       " 'GoogLeNet',\n",
       " 'GoogLeNetOutputs',\n",
       " 'GoogLeNet_Weights',\n",
       " 'Inception3',\n",
       " 'InceptionOutputs',\n",
       " 'Inception_V3_Weights',\n",
       " 'MNASNet',\n",
       " 'MNASNet0_5_Weights',\n",
       " 'MNASNet0_75_Weights',\n",
       " 'MNASNet1_0_Weights',\n",
       " 'MNASNet1_3_Weights',\n",
       " 'MaxVit',\n",
       " 'MaxVit_T_Weights',\n",
       " 'MobileNetV2',\n",
       " 'MobileNetV3',\n",
       " 'MobileNet_V2_Weights',\n",
       " 'MobileNet_V3_Large_Weights',\n",
       " 'MobileNet_V3_Small_Weights',\n",
       " 'RegNet',\n",
       " 'RegNet_X_16GF_Weights',\n",
       " 'RegNet_X_1_6GF_Weights',\n",
       " 'RegNet_X_32GF_Weights',\n",
       " 'RegNet_X_3_2GF_Weights',\n",
       " 'RegNet_X_400MF_Weights',\n",
       " 'RegNet_X_800MF_Weights',\n",
       " 'RegNet_X_8GF_Weights',\n",
       " 'RegNet_Y_128GF_Weights',\n",
       " 'RegNet_Y_16GF_Weights',\n",
       " 'RegNet_Y_1_6GF_Weights',\n",
       " 'RegNet_Y_32GF_Weights',\n",
       " 'RegNet_Y_3_2GF_Weights',\n",
       " 'RegNet_Y_400MF_Weights',\n",
       " 'RegNet_Y_800MF_Weights',\n",
       " 'RegNet_Y_8GF_Weights',\n",
       " 'ResNeXt101_32X8D_Weights',\n",
       " 'ResNeXt101_64X4D_Weights',\n",
       " 'ResNeXt50_32X4D_Weights',\n",
       " 'ResNet',\n",
       " 'ResNet101_Weights',\n",
       " 'ResNet152_Weights',\n",
       " 'ResNet18_Weights',\n",
       " 'ResNet34_Weights',\n",
       " 'ResNet50_Weights',\n",
       " 'ShuffleNetV2',\n",
       " 'ShuffleNet_V2_X0_5_Weights',\n",
       " 'ShuffleNet_V2_X1_0_Weights',\n",
       " 'ShuffleNet_V2_X1_5_Weights',\n",
       " 'ShuffleNet_V2_X2_0_Weights',\n",
       " 'SqueezeNet',\n",
       " 'SqueezeNet1_0_Weights',\n",
       " 'SqueezeNet1_1_Weights',\n",
       " 'SwinTransformer',\n",
       " 'Swin_B_Weights',\n",
       " 'Swin_S_Weights',\n",
       " 'Swin_T_Weights',\n",
       " 'Swin_V2_B_Weights',\n",
       " 'Swin_V2_S_Weights',\n",
       " 'Swin_V2_T_Weights',\n",
       " 'VGG',\n",
       " 'VGG11_BN_Weights',\n",
       " 'VGG11_Weights',\n",
       " 'VGG13_BN_Weights',\n",
       " 'VGG13_Weights',\n",
       " 'VGG16_BN_Weights',\n",
       " 'VGG16_Weights',\n",
       " 'VGG19_BN_Weights',\n",
       " 'VGG19_Weights',\n",
       " 'ViT_B_16_Weights',\n",
       " 'ViT_B_32_Weights',\n",
       " 'ViT_H_14_Weights',\n",
       " 'ViT_L_16_Weights',\n",
       " 'ViT_L_32_Weights',\n",
       " 'VisionTransformer',\n",
       " 'Wide_ResNet101_2_Weights',\n",
       " 'Wide_ResNet50_2_Weights',\n",
       " '_GoogLeNetOutputs',\n",
       " '_InceptionOutputs',\n",
       " '__builtins__',\n",
       " '__cached__',\n",
       " '__doc__',\n",
       " '__file__',\n",
       " '__loader__',\n",
       " '__name__',\n",
       " '__package__',\n",
       " '__path__',\n",
       " '__spec__',\n",
       " '_api',\n",
       " '_meta',\n",
       " '_utils',\n",
       " 'alexnet',\n",
       " 'convnext',\n",
       " 'convnext_base',\n",
       " 'convnext_large',\n",
       " 'convnext_small',\n",
       " 'convnext_tiny',\n",
       " 'densenet',\n",
       " 'densenet121',\n",
       " 'densenet161',\n",
       " 'densenet169',\n",
       " 'densenet201',\n",
       " 'detection',\n",
       " 'efficientnet',\n",
       " 'efficientnet_b0',\n",
       " 'efficientnet_b1',\n",
       " 'efficientnet_b2',\n",
       " 'efficientnet_b3',\n",
       " 'efficientnet_b4',\n",
       " 'efficientnet_b5',\n",
       " 'efficientnet_b6',\n",
       " 'efficientnet_b7',\n",
       " 'efficientnet_v2_l',\n",
       " 'efficientnet_v2_m',\n",
       " 'efficientnet_v2_s',\n",
       " 'get_model',\n",
       " 'get_model_builder',\n",
       " 'get_model_weights',\n",
       " 'get_weight',\n",
       " 'googlenet',\n",
       " 'inception',\n",
       " 'inception_v3',\n",
       " 'list_models',\n",
       " 'maxvit',\n",
       " 'maxvit_t',\n",
       " 'mnasnet',\n",
       " 'mnasnet0_5',\n",
       " 'mnasnet0_75',\n",
       " 'mnasnet1_0',\n",
       " 'mnasnet1_3',\n",
       " 'mobilenet',\n",
       " 'mobilenet_v2',\n",
       " 'mobilenet_v3_large',\n",
       " 'mobilenet_v3_small',\n",
       " 'mobilenetv2',\n",
       " 'mobilenetv3',\n",
       " 'optical_flow',\n",
       " 'quantization',\n",
       " 'regnet',\n",
       " 'regnet_x_16gf',\n",
       " 'regnet_x_1_6gf',\n",
       " 'regnet_x_32gf',\n",
       " 'regnet_x_3_2gf',\n",
       " 'regnet_x_400mf',\n",
       " 'regnet_x_800mf',\n",
       " 'regnet_x_8gf',\n",
       " 'regnet_y_128gf',\n",
       " 'regnet_y_16gf',\n",
       " 'regnet_y_1_6gf',\n",
       " 'regnet_y_32gf',\n",
       " 'regnet_y_3_2gf',\n",
       " 'regnet_y_400mf',\n",
       " 'regnet_y_800mf',\n",
       " 'regnet_y_8gf',\n",
       " 'resnet',\n",
       " 'resnet101',\n",
       " 'resnet152',\n",
       " 'resnet18',\n",
       " 'resnet34',\n",
       " 'resnet50',\n",
       " 'resnext101_32x8d',\n",
       " 'resnext101_64x4d',\n",
       " 'resnext50_32x4d',\n",
       " 'segmentation',\n",
       " 'shufflenet_v2_x0_5',\n",
       " 'shufflenet_v2_x1_0',\n",
       " 'shufflenet_v2_x1_5',\n",
       " 'shufflenet_v2_x2_0',\n",
       " 'shufflenetv2',\n",
       " 'squeezenet',\n",
       " 'squeezenet1_0',\n",
       " 'squeezenet1_1',\n",
       " 'swin_b',\n",
       " 'swin_s',\n",
       " 'swin_t',\n",
       " 'swin_transformer',\n",
       " 'swin_v2_b',\n",
       " 'swin_v2_s',\n",
       " 'swin_v2_t',\n",
       " 'vgg',\n",
       " 'vgg11',\n",
       " 'vgg11_bn',\n",
       " 'vgg13',\n",
       " 'vgg13_bn',\n",
       " 'vgg16',\n",
       " 'vgg16_bn',\n",
       " 'vgg19',\n",
       " 'vgg19_bn',\n",
       " 'video',\n",
       " 'vision_transformer',\n",
       " 'vit_b_16',\n",
       " 'vit_b_32',\n",
       " 'vit_h_14',\n",
       " 'vit_l_16',\n",
       " 'vit_l_32',\n",
       " 'wide_resnet101_2',\n",
       " 'wide_resnet50_2']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dir(models)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a75ffa5",
   "metadata": {},
   "source": [
    "# 2.加载某个预定义模型"
   ]
  },
  {
   "cell_type": "raw",
   "id": "8fbfe211",
   "metadata": {},
   "source": [
    "a.加载AlexNet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a23282a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "alexnet = models.AlexNet()"
   ]
  },
  {
   "cell_type": "raw",
   "id": "50320a65",
   "metadata": {},
   "source": [
    "b.加载ResNet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "41408fa9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "F:\\aconda-env-win\\DL_Pytorch\\lib\\site-packages\\torchvision\\models\\_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "F:\\aconda-env-win\\DL_Pytorch\\lib\\site-packages\\torchvision\\models\\_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet101_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet101_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n",
      "Downloading: \"https://download.pytorch.org/models/resnet101-63fe2227.pth\" to C:\\Users\\WIN10/.cache\\torch\\hub\\checkpoints\\resnet101-63fe2227.pth\n",
      "100.0%\n"
     ]
    }
   ],
   "source": [
    "resnet = models.resnet101(pretrained=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "13e4bb30",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ResNet(\n",
       "  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
       "  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "  (relu): ReLU(inplace=True)\n",
       "  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "  (layer1): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (layer2): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (3): Bottleneck(\n",
       "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (layer3): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (3): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (4): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (5): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (6): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (7): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (8): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (9): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (10): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (11): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (12): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (13): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (14): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (15): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (16): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (17): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (18): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (19): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (20): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (21): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (22): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (layer4): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
       "  (fc): Linear(in_features=2048, out_features=1000, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resnet"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7083f2a3",
   "metadata": {},
   "source": [
    "# 3.定义函数预处理管道"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5acf92cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import transforms"
   ]
  },
  {
   "cell_type": "raw",
   "id": "eb20ed72",
   "metadata": {},
   "source": [
    "transforms是组合容器，类似管道: Resize + CenterCrop--剪裁，ToTensor--转张量， Normalize--RGB归一化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d5acf004",
   "metadata": {},
   "outputs": [],
   "source": [
    "preprocess = transforms.Compose([transforms.Resize(256),transforms.CenterCrop(224),transforms.ToTensor(),transforms.Normalize(mean=[0.485,0.456,0.406],std=[0.229,0.224,0.225])])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e9e4a59",
   "metadata": {},
   "source": [
    "# 4.加载图片"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "4a4361d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3b78a4d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "img = Image.open(\"F:\\\\jupyter home\\\\dome.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a7c4629c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<PIL.PngImagePlugin.PngImageFile image mode=RGB size=256x256>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img"
   ]
  },
  {
   "cell_type": "raw",
   "id": "f4640083",
   "metadata": {},
   "source": [
    "重塑图片"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3dd59576",
   "metadata": {},
   "outputs": [],
   "source": [
    "img_t = preprocess(img)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2882b783",
   "metadata": {},
   "source": [
    "# 5.运行模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "bc49eb36",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "b5943579",
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_t = torch.unsqueeze(img_t, 0)"
   ]
  },
  {
   "cell_type": "raw",
   "id": "7a2e0039",
   "metadata": {},
   "source": [
    "eval 模式 --评估模式"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "988cc3d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ResNet(\n",
       "  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
       "  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "  (relu): ReLU(inplace=True)\n",
       "  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "  (layer1): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (layer2): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (3): Bottleneck(\n",
       "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (layer3): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (3): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (4): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (5): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (6): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (7): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (8): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (9): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (10): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (11): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (12): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (13): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (14): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (15): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (16): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (17): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (18): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (19): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (20): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (21): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (22): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (layer4): Sequential(\n",
       "    (0): Bottleneck(\n",
       "      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "      (downsample): Sequential(\n",
       "        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      )\n",
       "    )\n",
       "    (1): Bottleneck(\n",
       "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "    (2): Bottleneck(\n",
       "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "      (relu): ReLU(inplace=True)\n",
       "    )\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
       "  (fc): Linear(in_features=2048, out_features=1000, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resnet.eval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "5d44a7d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "out = resnet(batch_t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "6e9d9237",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[-2.4001e+00,  2.8959e-01,  2.3363e+00,  2.1809e+00,  2.7404e+00,\n",
       "          1.2835e-01, -7.4239e-01, -1.4432e+00, -1.7717e+00, -3.5675e+00,\n",
       "         -3.1370e+00, -2.3162e+00, -1.1310e+00, -2.4895e+00, -1.0910e+00,\n",
       "         -2.3436e+00, -3.7751e+00, -1.4518e+00, -9.7028e-01, -2.7080e+00,\n",
       "         -3.1844e+00, -2.6203e+00, -1.7676e+00, -3.3299e-01, -2.1045e-01,\n",
       "         -4.0211e+00, -2.9857e+00, -3.0741e+00, -3.4934e+00, -2.5480e+00,\n",
       "         -1.6836e+00, -1.7619e+00, -2.3255e+00,  1.0627e+00,  1.8741e+00,\n",
       "         -2.9454e+00, -7.3058e-01, -1.1092e+00, -1.1886e+00, -1.3963e-01,\n",
       "         -1.3808e+00, -1.1147e+00, -2.8379e+00, -1.9736e+00, -1.7106e+00,\n",
       "         -9.9270e-01,  8.5779e-02, -1.9646e+00, -1.1526e+00, -1.8292e+00,\n",
       "         -1.3703e+00,  5.6460e-01, -2.3639e+00, -3.5106e+00, -2.8571e+00,\n",
       "         -8.2920e-01, -2.9348e+00, -2.9211e+00, -1.2175e+00, -1.4904e+00,\n",
       "         -1.0336e+00, -2.5605e+00, -2.9267e+00, -1.4687e+00, -7.3422e-01,\n",
       "         -1.1219e+00, -2.6086e+00, -2.3936e+00, -3.3001e+00, -1.6810e+00,\n",
       "         -2.8400e+00,  1.1388e+00,  1.0468e-01, -2.9528e-01, -8.4638e-01,\n",
       "         -1.1254e+00, -1.0375e+00, -7.3092e-01, -1.4591e+00, -1.7003e+00,\n",
       "         -2.8816e+00, -3.6270e+00, -1.2672e+00, -1.9848e+00, -1.6586e+00,\n",
       "         -2.4345e+00, -1.5103e+00, -1.3195e-01, -3.9978e-01, -1.7189e+00,\n",
       "         -2.6516e+00, -2.8522e+00, -2.8696e+00, -3.4341e+00, -2.0616e+00,\n",
       "         -4.2855e+00, -1.0882e+00, -2.6754e+00, -3.0050e+00, -1.8109e+00,\n",
       "         -1.9780e+00, -2.8285e+00, -3.9088e+00, -2.2650e+00, -3.2735e+00,\n",
       "         -1.0237e+00, -2.5012e+00, -6.4343e-01, -9.3740e-01, -1.2967e+00,\n",
       "         -3.6304e+00,  1.8106e+00, -7.9527e-01, -1.6053e+00, -1.8967e-01,\n",
       "         -3.3030e+00, -3.5326e+00, -3.3427e-02, -1.6455e+00, -2.1601e+00,\n",
       "         -3.3451e+00, -1.5170e+00, -1.3606e+00, -1.7584e-01, -1.4898e+00,\n",
       "         -3.6208e+00, -9.4170e-01, -2.1196e-01, -1.3550e+00, -4.7541e-01,\n",
       "         -1.4832e+00, -1.9109e+00, -9.3331e-01, -1.4784e+00, -8.2239e-01,\n",
       "         -2.8532e+00, -2.6251e-01, -9.7098e-01, -3.0534e+00, -3.2395e+00,\n",
       "         -3.2438e+00, -2.4776e+00, -2.6923e+00, -2.9161e+00,  1.4259e-01,\n",
       "         -2.0264e+00, -9.1773e-01, -1.8750e+00,  1.3020e+00,  3.4237e-01,\n",
       "          2.1179e+00, -9.7908e-01, -3.1118e+00, -1.6214e+00, -3.8318e-01,\n",
       "         -1.2244e-01, -1.2398e+00, -4.0036e-01, -1.6559e+00, -1.5751e+00,\n",
       "         -8.1643e-01, -1.2592e+00, -1.0716e+00, -2.4234e-01, -3.9480e-02,\n",
       "         -1.2151e+00, -1.1966e+00, -1.6425e+00, -2.1872e+00, -2.0092e+00,\n",
       "         -7.2424e-01, -1.4107e+00, -1.0923e+00, -2.0435e+00, -5.5658e-01,\n",
       "         -1.9597e+00, -2.1538e+00, -2.1284e+00,  3.6300e-01, -7.5152e-01,\n",
       "         -1.3100e+00, -9.5301e-01, -1.2194e+00, -8.3126e-01, -7.9362e-01,\n",
       "         -1.4582e+00, -2.1667e+00, -1.4657e+00, -1.9514e+00, -1.3674e+00,\n",
       "         -1.6872e+00, -7.2464e-01, -1.1395e+00, -1.3454e+00, -1.4303e+00,\n",
       "         -7.1831e-01, -6.1025e-01, -1.0096e+00, -1.7686e+00, -9.2277e-01,\n",
       "         -1.8652e+00, -1.7935e+00, -6.9055e-01, -4.8785e-01, -1.1172e+00,\n",
       "         -1.4795e+00, -2.2625e+00, -4.3496e-01, -1.4930e+00, -6.0445e-01,\n",
       "         -2.7239e+00,  4.4320e-01, -1.2786e+00, -9.4442e-01, -1.1288e+00,\n",
       "         -7.5815e-01, -2.5180e+00, -1.1971e+00, -9.9660e-01, -1.8679e+00,\n",
       "         -2.7981e+00, -2.1361e+00, -5.1716e-01, -9.3469e-01, -6.3479e-01,\n",
       "         -3.9408e-01, -6.8884e-01, -2.7236e+00, -2.0933e+00, -1.0354e-01,\n",
       "         -1.5041e+00, -5.3179e-01, -6.4150e-01, -4.7835e-01, -3.1057e-01,\n",
       "         -1.1051e+00, -2.0697e+00, -2.0155e+00, -1.0824e+00,  2.9829e-01,\n",
       "         -5.4867e-01, -1.4517e+00,  7.0803e-01,  5.8244e-02, -1.9980e+00,\n",
       "         -1.5443e+00, -1.3289e+00, -5.9368e-01, -1.6063e+00, -2.6460e+00,\n",
       "         -8.2484e-01, -1.3635e-01, -2.1242e+00, -2.2754e-02,  8.8346e-01,\n",
       "         -2.0270e+00, -7.2902e-01, -2.5312e-01, -9.6499e-01, -1.3695e+00,\n",
       "         -1.7972e+00, -1.9738e+00, -1.8942e+00,  6.9111e-01, -2.0721e-03,\n",
       "         -1.7133e+00, -2.5117e+00, -2.4884e+00, -2.8404e+00, -2.2709e+00,\n",
       "         -2.4444e+00, -2.1885e+00, -1.6089e+00, -2.1778e+00, -2.8215e+00,\n",
       "         -2.6668e+00, -3.0938e+00, -1.9828e+00, -2.4502e+00, -1.9471e+00,\n",
       "         -1.0129e+00,  4.5672e-01,  6.2891e-01, -9.0360e-01,  2.4585e-01,\n",
       "         -8.9819e-01, -1.2952e+00, -2.5941e+00, -9.6441e-01, -5.4356e-01,\n",
       "         -1.5412e+00, -3.6781e+00,  1.5148e-01, -3.7450e+00, -1.7112e+00,\n",
       "         -2.2630e+00,  2.1514e-01, -1.6681e+00, -1.8405e+00, -5.5157e-01,\n",
       "         -2.0185e+00, -3.4435e+00, -2.9638e+00, -1.9697e+00, -2.9135e+00,\n",
       "         -3.2533e+00, -2.8667e+00, -3.1027e+00, -1.5643e+00, -2.3170e+00,\n",
       "         -7.6777e-01, -7.4115e-01, -7.2218e-01,  1.2673e-01, -1.4125e-01,\n",
       "          1.2745e+00, -4.5096e-01, -6.6848e-01, -9.2197e-01,  2.3829e-01,\n",
       "         -1.2783e+00, -2.9051e+00, -2.4824e+00, -1.6237e+00, -1.1942e+00,\n",
       "         -2.6083e+00, -1.9739e+00,  1.2781e+00, -4.5326e-02, -2.2571e+00,\n",
       "         -1.2390e+00, -1.6364e+00, -1.0534e+00, -6.8012e-01, -2.2570e+00,\n",
       "         -3.5960e+00, -2.4003e+00, -1.0054e+00, -8.8000e-01, -3.1576e+00,\n",
       "         -1.3200e+00, -1.4641e+00, -3.3727e+00, -4.4853e+00, -2.7726e+00,\n",
       "         -7.2399e-01, -1.6413e+00, -3.3002e+00, -2.9716e+00, -3.4046e+00,\n",
       "         -3.2036e+00, -3.6052e+00, -4.1071e+00, -3.1202e+00, -3.0314e+00,\n",
       "         -2.0816e+00, -8.1830e-01, -2.8549e+00, -9.6292e-01, -1.5541e+00,\n",
       "         -2.5270e+00, -2.3987e-01, -9.2485e-01, -1.8808e+00, -2.2972e+00,\n",
       "          4.2524e-01, -8.8892e-01,  8.7390e-02, -7.1920e-01,  1.7469e-02,\n",
       "         -2.1092e+00, -3.5436e+00, -2.8301e+00, -8.3478e-01, -1.0096e+00,\n",
       "         -2.4372e+00,  1.1361e-02, -2.0433e+00, -2.1080e+00, -3.4628e+00,\n",
       "         -7.5814e-01, -9.7347e-01, -1.6500e+00, -7.7745e-01, -1.5538e+00,\n",
       "         -2.3880e+00, -2.0124e+00, -1.2627e+00, -1.0559e+00, -1.4425e+00,\n",
       "         -1.9431e+00,  2.8617e-01, -1.7946e+00,  4.5973e-01, -6.1013e-01,\n",
       "         -5.8147e-02,  3.9376e-01, -6.0569e-01,  1.6695e+00,  8.8196e-02,\n",
       "         -6.5317e-01,  1.7052e+00,  9.4042e-01,  7.3055e+00,  5.0567e+00,\n",
       "          5.1579e+00,  3.2690e+00,  2.4110e+00,  1.5785e+00,  5.5660e+00,\n",
       "         -1.5168e+00, -3.6003e-01,  2.6723e+00,  1.1574e+00,  1.6636e-01,\n",
       "          2.3012e+00,  1.7252e+00,  1.5339e+00,  2.0362e-01,  2.4705e+00,\n",
       "          1.1046e+00,  5.0679e+00,  2.5292e+00,  3.2104e+00,  3.1704e+00,\n",
       "          1.4085e+00,  6.4354e-01,  1.1091e-01,  2.1260e+00,  2.1204e-01,\n",
       "          4.2798e+00,  4.7251e-01, -8.6084e-01,  1.2957e+00,  8.5389e-02,\n",
       "          9.2061e-01,  7.4155e-01,  1.8819e+00,  2.0274e+00, -1.3540e+00,\n",
       "          1.2060e+00, -8.7002e-01,  3.3323e+00,  4.3380e-01, -4.1805e-01,\n",
       "          1.3157e+00,  1.7106e+00,  2.0024e+00,  3.2653e+00,  1.1750e+00,\n",
       "          1.7384e+00, -1.1428e-01,  3.1200e-01,  9.4922e-01,  3.4153e+00,\n",
       "          2.9690e-01,  5.3132e-01,  1.4617e-01,  1.2543e+00,  2.5793e+00,\n",
       "          2.5757e+00,  1.3562e+00,  9.0559e-01,  1.1760e+00, -1.8060e+00,\n",
       "          3.9093e-01,  4.2770e+00,  1.3162e+00,  4.3706e+00,  1.3603e+00,\n",
       "          1.2265e+00,  2.5688e+00,  6.4047e-03, -6.9964e-01, -1.2864e+00,\n",
       "          1.7954e+00,  1.8803e+00,  1.6820e+00,  3.7969e+00, -8.7182e-01,\n",
       "          3.9997e+00,  2.6692e+00,  3.2891e-01,  1.5408e+00,  1.5569e+00,\n",
       "          9.5540e-01, -6.7698e-01,  1.8251e+00,  1.2368e+00,  1.0949e+00,\n",
       "         -1.9040e+00,  4.9125e-01, -1.9469e-01, -3.0532e+00,  9.2891e-01,\n",
       "         -3.3203e-01,  1.4417e+00,  2.6349e+00,  3.5323e+00,  9.2684e-01,\n",
       "         -8.3753e-01,  3.9162e-02, -3.2911e-01, -8.9803e-01,  5.0460e-01,\n",
       "          3.7263e-01,  5.1831e+00, -2.7414e-01,  1.8191e+00,  6.2925e-02,\n",
       "          3.5512e+00, -2.0836e+00,  6.8795e-01,  2.9510e-02, -6.5817e-01,\n",
       "          2.0832e-01,  1.4604e+00,  6.7942e+00, -9.0405e-02,  2.9712e+00,\n",
       "          4.8765e-03, -2.1323e+00, -9.6153e-01,  2.6496e+00, -2.1346e-01,\n",
       "          4.1049e+00,  3.1905e+00,  2.7412e+00, -2.1050e-01,  1.0606e-01,\n",
       "          3.3213e+00,  6.5648e-01,  2.5061e+00, -4.9503e-01,  1.5661e-01,\n",
       "          3.0358e-01,  3.4460e+00, -2.7030e-01,  5.8069e+00,  3.7571e+00,\n",
       "          4.4956e+00,  1.0819e+00,  3.5462e-01,  2.9937e+00, -1.0382e+00,\n",
       "          2.0562e+00,  1.1355e+00,  2.2946e+00, -1.0094e+00,  3.0525e+00,\n",
       "         -1.6201e+00, -1.6913e+00,  6.1798e-01,  7.1010e-01,  1.5168e+00,\n",
       "          1.3555e+00,  3.6181e-01,  3.6755e+00, -2.2957e-01,  2.9738e+00,\n",
       "          1.6265e+00,  3.5140e+00,  3.6585e+00, -1.1588e+00,  5.4733e-01,\n",
       "          1.8680e+00, -6.4025e-01, -8.0439e-01, -1.6112e+00,  3.1577e+00,\n",
       "          1.7077e+00,  5.0034e-01,  6.1624e-01, -1.4589e+00,  1.0056e+00,\n",
       "          1.6426e-01,  3.8143e+00,  7.1351e-02, -5.7368e-02,  9.3387e-01,\n",
       "          3.1585e+00, -1.3717e+00,  1.6341e+00,  5.3982e+00, -3.7985e-01,\n",
       "         -4.8727e-01,  2.6155e+00,  1.9213e+00, -2.7111e+00, -8.3644e-01,\n",
       "         -1.2053e-01,  1.1975e+00,  2.2660e+00, -3.4252e-02,  7.2354e-01,\n",
       "          1.1427e+00,  7.3142e-01,  5.3440e-01, -4.0732e-01,  9.4155e-01,\n",
       "          3.4178e+00,  1.6313e-01, -3.7309e-01, -6.6142e-01,  8.8273e-01,\n",
       "          8.6917e-01,  5.7700e-01,  1.0292e+00,  1.6164e+00,  4.0654e-03,\n",
       "          9.5009e-01,  3.6195e+00,  1.8913e+00,  3.7240e+00,  4.1273e-01,\n",
       "         -4.8685e-01, -9.3485e-01,  2.1304e+00,  1.7044e+00,  2.8668e+00,\n",
       "          3.3334e+00, -1.3155e-01,  1.5010e+00,  1.1216e+00,  2.8253e+00,\n",
       "          5.4108e-01, -6.6444e-02,  4.6729e-01,  3.7208e+00, -7.5115e-01,\n",
       "         -1.9145e+00, -1.7777e+00,  9.1578e-01,  1.8117e+00,  2.2021e+00,\n",
       "          3.0379e+00, -1.6271e-01,  1.3964e+00, -7.5369e-01, -1.1809e-01,\n",
       "          2.5396e+00,  1.0145e-01, -2.4757e-03,  8.4746e-01,  4.9425e+00,\n",
       "          1.4031e+00,  7.1907e+00, -1.4644e-01,  2.7724e-01, -1.3332e+00,\n",
       "          1.8049e+00, -4.8314e-01,  1.0290e+00,  1.0053e+00,  8.4951e-01,\n",
       "          4.7899e-01,  1.8487e+00,  5.7188e+00,  6.8895e-01, -1.4952e+00,\n",
       "          1.3211e+00, -1.5312e+00,  2.8292e+00,  1.0620e+00,  5.0322e+00,\n",
       "          4.2145e-01,  6.7707e-03,  5.8262e-01,  4.8909e+00,  3.1565e+00,\n",
       "          4.2913e-01,  3.5653e-01, -5.1730e-01,  1.6396e+00,  2.0502e+00,\n",
       "          1.4861e+00, -3.1515e-01,  1.8525e+00, -3.6191e-01, -4.0123e-01,\n",
       "         -2.9294e-01,  2.0704e+00,  3.5920e+00,  3.9113e-01, -6.7039e-01,\n",
       "         -9.5847e-01, -6.0259e-01,  1.2443e+00,  2.0496e+00, -1.7994e+00,\n",
       "         -9.1910e-01,  1.9044e+00,  2.2801e+00, -8.8976e-01,  3.8600e+00,\n",
       "          7.8397e-01,  2.7347e+00, -1.1411e+00,  2.6955e+00, -7.4103e-01,\n",
       "          3.9845e-01,  2.7645e+00,  6.9929e-01,  6.4630e-01,  1.2629e+00,\n",
       "          2.5055e+00,  3.2344e+00,  1.7323e+00,  3.4532e+00, -1.3277e-01,\n",
       "          4.3985e-01, -6.2693e-01,  2.3624e+00,  5.0224e+00, -9.0294e-01,\n",
       "         -1.7326e+00,  1.9651e+00, -1.3836e+00,  4.0516e+00,  1.4679e-01,\n",
       "          2.4885e+00,  2.3318e+00,  4.3149e+00,  1.2020e+00,  7.7511e-01,\n",
       "          1.0201e+00, -8.4074e-01,  2.8256e+00,  2.8484e+00, -1.5586e-01,\n",
       "          7.1221e-01, -6.6077e-03,  5.7033e-01,  2.7175e+00,  1.2346e+00,\n",
       "          7.4662e-01,  2.4784e+00,  1.9718e+00,  4.6296e-01,  9.7795e-01,\n",
       "          1.4655e+00,  3.7953e+00,  1.9916e+00,  2.2232e+00,  4.9911e+00,\n",
       "          4.2429e+00,  1.0375e+00,  1.6968e+00, -7.5233e-01, -2.3020e-01,\n",
       "          1.4182e+00,  9.5862e-01,  2.4008e+00,  9.0297e-01,  1.4115e+00,\n",
       "          7.1380e+00,  1.6920e+00,  2.0888e+00, -2.6489e-01,  3.7625e-01,\n",
       "          1.2251e+00,  2.1426e+00,  1.1627e+00, -2.9614e-01,  8.5588e-01,\n",
       "          3.4527e+00, -2.9526e-01,  2.2397e+00,  2.6147e+00,  4.0119e+00,\n",
       "          3.1566e+00,  6.1701e-01,  4.3964e-01,  3.6242e-01,  1.0961e+00,\n",
       "          2.9592e-01,  9.8718e-01,  1.1779e+00,  3.3096e+00,  2.9111e-01,\n",
       "          3.7370e-01,  3.6476e+00,  3.4548e+00,  2.3546e+00,  8.4180e-01,\n",
       "         -7.7477e-01,  9.0827e-01,  2.2939e+00,  1.6167e+00, -3.9659e-01,\n",
       "         -8.7072e-01,  7.3500e-01,  1.1950e+00,  4.1554e-01,  1.1087e+00,\n",
       "          1.9651e+00,  2.3903e-01, -4.4412e-01,  3.8391e-01,  2.4762e+00,\n",
       "         -1.0591e+00,  3.1988e-01,  1.0927e-01,  1.4653e+00, -1.4787e+00,\n",
       "          2.7668e+00,  1.3625e+00,  4.5962e+00,  7.8745e-01,  1.4960e-01,\n",
       "         -2.0344e-01, -3.6843e-01,  4.4974e+00, -1.5371e-01,  1.1504e+00,\n",
       "          6.5093e-01, -5.8496e-01, -9.5276e-01,  3.5543e+00,  3.1853e+00,\n",
       "          1.2016e+00,  4.6053e+00,  7.8007e-01,  1.2895e+00, -1.0965e+00,\n",
       "         -1.0284e+00,  1.3692e-01,  2.0165e-01,  9.9519e-02,  3.9775e+00,\n",
       "          3.1008e+00,  5.2659e-01,  1.7276e+00,  1.8866e+00,  6.5559e-01,\n",
       "          5.9781e+00,  1.7284e+00,  5.2466e-01, -1.8473e+00,  3.3517e+00,\n",
       "          1.6882e+00,  1.5279e+00,  2.0300e+00,  1.7989e-01,  9.9772e-01,\n",
       "          4.1856e+00,  2.1888e+00,  2.2118e+00,  1.0206e+00, -3.0741e-01,\n",
       "          8.8683e-01,  2.8578e+00,  2.6988e+00,  1.7902e-01,  1.3170e+00,\n",
       "         -3.2466e-03, -7.4750e-01,  2.3990e+00,  6.0669e-01, -7.2414e-02,\n",
       "          1.8538e+00,  9.9379e-01,  4.1547e+00,  4.2302e+00,  2.0125e+00,\n",
       "          2.3722e+00,  1.5571e-01,  1.2744e+00,  3.7714e+00, -2.7043e-01,\n",
       "          8.6235e-01,  2.6401e+00,  3.1064e+00,  2.7739e+00,  9.5304e-01,\n",
       "         -4.8317e-01,  1.2366e+00,  4.0596e+00, -3.9821e-01,  3.0103e+00,\n",
       "          1.2963e+00, -3.2181e-01, -8.7488e-02,  2.9418e+00,  4.7900e+00,\n",
       "         -9.4617e-01,  1.7587e+00,  1.0592e+00,  2.7318e+00,  2.7327e-01,\n",
       "          6.3489e-01, -2.7459e+00,  3.5363e+00,  1.2783e+00,  3.0556e-01,\n",
       "          4.3175e+00,  1.9159e+00,  1.6042e+00,  1.4982e+00, -1.0432e+00,\n",
       "          3.2290e+00, -8.8712e-01,  3.8480e-01, -7.9816e-01,  3.3432e+00,\n",
       "          2.4576e-01, -5.3330e-01, -3.8600e-01,  6.8085e+00,  7.8510e-02,\n",
       "         -1.2946e+00, -9.9652e-01, -9.2344e-01,  2.6258e+00, -2.0872e+00,\n",
       "          9.4692e-01,  1.0253e+00,  5.9279e+00,  2.6597e+00,  6.6413e+00,\n",
       "          5.0809e+00,  6.6158e+00,  8.8401e-01, -1.1948e+00, -2.4609e+00,\n",
       "         -2.1828e+00, -3.7056e+00, -2.6900e+00, -2.9325e+00,  1.5482e+00,\n",
       "         -3.2257e+00, -2.6260e+00, -1.4674e+00, -1.9310e+00, -3.0176e-01,\n",
       "         -2.4328e+00,  8.5877e-02,  4.0022e-01, -8.9348e-01, -1.4316e+00,\n",
       "         -2.4121e+00, -1.6651e+00, -8.0847e-01,  2.5205e-01,  7.2795e-01,\n",
       "         -1.1263e+00,  5.8610e-01, -8.5070e-01, -5.7005e-01, -1.8322e+00,\n",
       "          6.7074e-01, -1.6032e+00,  6.6961e-01, -7.7268e-01, -7.5780e-01,\n",
       "         -8.0009e-01, -2.0472e+00,  4.8752e-01, -7.9287e-01, -1.7352e+00,\n",
       "         -4.2430e+00, -3.1350e+00, -2.1516e+00, -2.0381e+00, -2.6304e+00,\n",
       "         -2.0892e+00, -4.1557e-01, -8.3216e-01,  1.0836e+00, -2.6839e+00,\n",
       "          1.8823e+00,  1.2626e+00,  2.0422e-01,  2.8415e-01, -1.3497e+00,\n",
       "          8.4579e-01, -5.5787e-01, -1.7910e+00,  4.2057e-01, -3.1710e-01,\n",
       "          1.1655e+00,  2.6026e-01, -4.2221e-01,  1.8761e+00,  6.7828e-01,\n",
       "          1.5701e-01, -2.3656e+00, -1.0051e+00,  4.6910e-02, -1.7720e+00,\n",
       "         -2.5121e+00, -1.5802e+00, -1.6638e+00, -4.1282e+00, -2.0186e+00,\n",
       "         -1.4908e+00, -2.4191e+00, -2.7869e+00, -8.8768e-01,  1.6975e+00]],\n",
       "       grad_fn=<AddmmBackward0>)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "DL_Pytorch",
   "language": "python",
   "name": "dl_pytorch"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}