diff --git a/08.ipynb b/08.ipynb new file mode 100644 index 0000000..abd337d --- /dev/null +++ b/08.ipynb @@ -0,0 +1,153 @@ +{ + "cells": [ + { + "cell_type": "code", + "id": "initial_id", + "metadata": { + "collapsed": true, + "ExecuteTime": { + "end_time": "2025-06-13T08:32:44.715688Z", + "start_time": "2025-06-13T08:32:44.700427Z" + } + }, + "source": [ + "import torchvision.models as models\n", + "from sympy.printing.pytorch import torch\n", + "\n", + "google_net = models.googlenet(pretrained=True, )" + ], + "outputs": [ + { + "ename": "ModuleNotFoundError", + "evalue": "No module named 'sympy.printing.pytorch'", + "output_type": "error", + "traceback": [ + "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m", + "\u001B[0;31mModuleNotFoundError\u001B[0m Traceback (most recent call last)", + "Cell \u001B[0;32mIn[8], line 3\u001B[0m\n\u001B[1;32m 1\u001B[0m \u001B[38;5;28;01mimport\u001B[39;00m \u001B[38;5;21;01mtorch\u001B[39;00m\n\u001B[1;32m 2\u001B[0m \u001B[38;5;28;01mimport\u001B[39;00m \u001B[38;5;21;01mtorchvision\u001B[39;00m\u001B[38;5;21;01m.\u001B[39;00m\u001B[38;5;21;01mmodels\u001B[39;00m \u001B[38;5;28;01mas\u001B[39;00m \u001B[38;5;21;01mmodels\u001B[39;00m\n\u001B[0;32m----> 3\u001B[0m \u001B[38;5;28;01mfrom\u001B[39;00m \u001B[38;5;21;01msympy\u001B[39;00m\u001B[38;5;21;01m.\u001B[39;00m\u001B[38;5;21;01mprinting\u001B[39;00m\u001B[38;5;21;01m.\u001B[39;00m\u001B[38;5;21;01mpytorch\u001B[39;00m \u001B[38;5;28;01mimport\u001B[39;00m torch\n\u001B[1;32m 5\u001B[0m google_net \u001B[38;5;241m=\u001B[39m models\u001B[38;5;241m.\u001B[39mgooglenet(pretrained\u001B[38;5;241m=\u001B[39m\u001B[38;5;28;01mTrue\u001B[39;00m, )\n", + "\u001B[0;31mModuleNotFoundError\u001B[0m: No module named 'sympy.printing.pytorch'" + ] + } + ], + "execution_count": 8 + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2025-06-13T08:33:06.406399Z", + "start_time": "2025-06-13T08:33:06.400289Z" + } + }, + "cell_type": "code", + "source": [ + "import torch\n", + "\n", + "# 提取分类器的输入特征数量\n", + "fc_in_features = google_net.fc.in_features\n", + "print(\"fc_in_features: \", fc_in_features)\n", + "\n", + "# 查看分类层的输出参数\n", + "fc_out_features = google_net.fc.out_features\n", + "print(\"fc_out_features: \", fc_out_features)\n", + "\n", + "# 修改分类器\n", + "google_net.fc = torch.nn.Linear(fc_in_features, 10)" + ], + "id": "d5763cbec91c47c", + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "fc_in_features: 1024\n", + "fc_out_features: 1000\n" + ] + } + ], + "execution_count": 9 + }, + { + "metadata": { + "ExecuteTime": { + "end_time": "2025-06-13T08:50:49.075617Z", + "start_time": "2025-06-13T08:50:48.639134Z" + } + }, + "cell_type": "code", + "source": [ + "import torchvision\n", + "from torchvision import datasets, transforms\n", + "from torch.utils.data import DataLoader\n", + "\n", + "# 加载mnist数据集\n", + "mnist_dataset = datasets.MNIST(root=\"./data\", train=True, transform=transforms.ToTensor(), download=True,\n", + " target_transform=None)\n", + "\n", + "# 取32张图片\n", + "tensor_loader = DataLoader(dataset=mnist_dataset, batch_size=32)\n", + "\n", + "data_iter = iter(tensor_loader)\n", + "img_tensor, label_tensor = next(data_iter)\n", + "print(img_tensor.shape)\n", + "\n", + "grid_tensor = torchvision.utils.make_grid(img_tensor, nrow=8, padding=2)\n", + "grid_img = transforms.ToPILImage()(grid_tensor)\n", + "display(grid_img)\n", + "\n", + "print(grid_tensor.shape)\n", + "\n", + "torchvision.utils.save_image(grid_tensor, \"./mnist_grid.png\")" + ], + "id": "ad1ab89dd8285a8c", + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "torch.Size([32, 1, 28, 28])\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ], + "image/png": "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", + "image/jpeg": "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" + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "torch.Size([3, 122, 242])\n" + ] + } + ], + "execution_count": 24 + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 2 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython2", + "version": "2.7.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +}