{ "cells": [ { "cell_type": "markdown", "id": "273bffe9", "metadata": { "origin_pos": 0 }, "source": [ "# 风格迁移\n", "\n", "摄影爱好者也许接触过滤波器。它能改变照片的颜色风格，从而使风景照更加锐利或者令人像更加美白。但一个滤波器通常只能改变照片的某个方面。如果要照片达到理想中的风格，可能需要尝试大量不同的组合。这个过程的复杂程度不亚于模型调参。\n", "\n", "本节将介绍如何使用卷积神经网络，自动将一个图像中的风格应用在另一图像之上，即*风格迁移*（style transfer） :cite:`Gatys.Ecker.Bethge.2016`。\n", "这里我们需要两张输入图像：一张是*内容图像*，另一张是*风格图像*。\n", "我们将使用神经网络修改内容图像，使其在风格上接近风格图像。\n", "例如， :numref:`fig_style_transfer`中的内容图像为本书作者在西雅图郊区的雷尼尔山国家公园拍摄的风景照，而风格图像则是一幅主题为秋天橡树的油画。\n", "最终输出的合成图像应用了风格图像的油画笔触让整体颜色更加鲜艳，同时保留了内容图像中物体主体的形状。\n", "\n", "![输入内容图像和风格图像，输出风格迁移后的合成图像](../img/style-transfer.svg)\n", ":label:`fig_style_transfer`\n", "\n", "## 方法\n", "\n", " :numref:`fig_style_transfer_model`用简单的例子阐述了基于卷积神经网络的风格迁移方法。\n", "首先，我们初始化合成图像，例如将其初始化为内容图像。\n", "该合成图像是风格迁移过程中唯一需要更新的变量，即风格迁移所需迭代的模型参数。\n", "然后，我们选择一个预训练的卷积神经网络来抽取图像的特征，其中的模型参数在训练中无须更新。\n", "这个深度卷积神经网络凭借多个层逐级抽取图像的特征，我们可以选择其中某些层的输出作为内容特征或风格特征。\n", "以 :numref:`fig_style_transfer_model`为例，这里选取的预训练的神经网络含有3个卷积层，其中第二层输出内容特征，第一层和第三层输出风格特征。\n", "\n", "![基于卷积神经网络的风格迁移。实线箭头和虚线箭头分别表示前向传播和反向传播](../img/neural-style.svg)\n", ":label:`fig_style_transfer_model`\n", "\n", "接下来，我们通过前向传播（实线箭头方向）计算风格迁移的损失函数，并通过反向传播（虚线箭头方向）迭代模型参数，即不断更新合成图像。\n", "风格迁移常用的损失函数由3部分组成：\n", "\n", "1. *内容损失*使合成图像与内容图像在内容特征上接近；\n", "1. *风格损失*使合成图像与风格图像在风格特征上接近；\n", "1. *全变分损失*则有助于减少合成图像中的噪点。\n", "\n", "最后，当模型训练结束时，我们输出风格迁移的模型参数，即得到最终的合成图像。\n", "\n", "在下面，我们将通过代码来进一步了解风格迁移的技术细节。\n", "\n", "## [**阅读内容和风格图像**]\n", "\n", "首先，我们读取内容和风格图像。\n", "从打印出的图像坐标轴可以看出，它们的尺寸并不一样。\n" ] }, { "cell_type": "code", "execution_count": 1, "id": "a0d90f51", "metadata": { "execution": { "iopub.execute_input": "2023-08-18T07:23:26.021505Z", "iopub.status.busy": "2023-08-18T07:23:26.020759Z", "iopub.status.idle": "2023-08-18T07:23:29.597245Z", "shell.execute_reply": "2023-08-18T07:23:29.595990Z" }, "origin_pos": 2, "tab": [ "pytorch" ] }, "outputs": [ { "data": { "image/svg+xml": [ "\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2023-08-18T07:23:29.411028\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.5.1, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "%matplotlib inline\n", "import torch\n", "import torchvision\n", "from torch import nn\n", "from d2l import torch as d2l\n", "\n", "d2l.set_figsize()\n", "content_img = d2l.Image.open('../img/rainier.jpg')\n", "d2l.plt.imshow(content_img);" ] }, { "cell_type": "code", "execution_count": 2, "id": "ec590a65", "metadata": { "execution": { "iopub.execute_input": "2023-08-18T07:23:29.601550Z", "iopub.status.busy": "2023-08-18T07:23:29.600514Z", "iopub.status.idle": "2023-08-18T07:23:30.096132Z", "shell.execute_reply": "2023-08-18T07:23:30.095315Z" }, "origin_pos": 5, "tab": [ "pytorch" ] }, "outputs": [ { "data": { "image/svg+xml": [ "\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2023-08-18T07:23:29.959273\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.5.1, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "style_img = d2l.Image.open('../img/autumn-oak.jpg')\n", "d2l.plt.imshow(style_img);" ] }, { "cell_type": "markdown", "id": "ddc886a6", "metadata": { "origin_pos": 6 }, "source": [ "## [**预处理和后处理**]\n", "\n", "下面，定义图像的预处理函数和后处理函数。\n", "预处理函数`preprocess`对输入图像在RGB三个通道分别做标准化，并将结果变换成卷积神经网络接受的输入格式。\n", "后处理函数`postprocess`则将输出图像中的像素值还原回标准化之前的值。\n", "由于图像打印函数要求每个像素的浮点数值在0～1之间，我们对小于0和大于1的值分别取0和1。\n" ] }, { "cell_type": "code", "execution_count": 3, "id": "6f351192", "metadata": { "execution": { "iopub.execute_input": "2023-08-18T07:23:30.103272Z", "iopub.status.busy": "2023-08-18T07:23:30.102388Z", "iopub.status.idle": "2023-08-18T07:23:30.112076Z", "shell.execute_reply": "2023-08-18T07:23:30.111052Z" }, "origin_pos": 8, "tab": [ "pytorch" ] }, "outputs": [], "source": [ "rgb_mean = torch.tensor([0.485, 0.456, 0.406])\n", "rgb_std = torch.tensor([0.229, 0.224, 0.225])\n", "\n", "def preprocess(img, image_shape):\n", " transforms = torchvision.transforms.Compose([\n", " torchvision.transforms.Resize(image_shape),\n", " torchvision.transforms.ToTensor(),\n", " torchvision.transforms.Normalize(mean=rgb_mean, std=rgb_std)])\n", " return transforms(img).unsqueeze(0)\n", "\n", "def postprocess(img):\n", " img = img[0].to(rgb_std.device)\n", " img = torch.clamp(img.permute(1, 2, 0) * rgb_std + rgb_mean, 0, 1)\n", " return torchvision.transforms.ToPILImage()(img.permute(2, 0, 1))" ] }, { "cell_type": "markdown", "id": "54e9f5d8", "metadata": { "origin_pos": 10 }, "source": [ "## [**抽取图像特征**]\n", "\n", "我们使用基于ImageNet数据集预训练的VGG-19模型来抽取图像特征 :cite:`Gatys.Ecker.Bethge.2016`。\n" ] }, { "cell_type": "code", "execution_count": 4, "id": "f562ee81", "metadata": { "execution": { "iopub.execute_input": "2023-08-18T07:23:30.117700Z", "iopub.status.busy": "2023-08-18T07:23:30.116834Z", "iopub.status.idle": "2023-08-18T07:23:42.885822Z", "shell.execute_reply": "2023-08-18T07:23:42.884582Z" }, "origin_pos": 12, "tab": [ "pytorch" ] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Downloading: \"https://download.pytorch.org/models/vgg19-dcbb9e9d.pth\" to /home/ci/.cache/torch/hub/checkpoints/vgg19-dcbb9e9d.pth\n" ] }, { "data": { "application/json": { "ascii": false, "bar_format": null, "colour": null, "elapsed": 0.007704734802246094, "initial": 0, "n": 0, "ncols": null, "nrows": null, "postfix": null, "prefix": "", "rate": null, "total": 574673361, "unit": "B", "unit_divisor": 1024, "unit_scale": true }, "application/vnd.jupyter.widget-view+json": { "model_id": "0a21ebb04f6e4afe9df09a7d7c6a0fe0", "version_major": 2, "version_minor": 0 }, "text/plain": [ " 0%| | 0.00/548M [00:00\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2023-08-18T07:24:35.138148\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.5.1, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "device, image_shape = d2l.try_gpu(), (300, 450)\n", "net = net.to(device)\n", "content_X, contents_Y = get_contents(image_shape, device)\n", "_, styles_Y = get_styles(image_shape, device)\n", "output = train(content_X, contents_Y, styles_Y, device, 0.3, 500, 50)" ] }, { "cell_type": "markdown", "id": "7f5c2480", "metadata": { "origin_pos": 55 }, "source": [ "我们可以看到，合成图像保留了内容图像的风景和物体，并同时迁移了风格图像的色彩。例如，合成图像具有与风格图像中一样的色彩块，其中一些甚至具有画笔笔触的细微纹理。\n", "\n", "## 小结\n", "\n", "* 风格迁移常用的损失函数由3部分组成：（1）内容损失使合成图像与内容图像在内容特征上接近；（2）风格损失令合成图像与风格图像在风格特征上接近；（3）全变分损失则有助于减少合成图像中的噪点。\n", "* 我们可以通过预训练的卷积神经网络来抽取图像的特征，并通过最小化损失函数来不断更新合成图像来作为模型参数。\n", "* 我们使用格拉姆矩阵表达风格层输出的风格。\n", "\n", "## 练习\n", "\n", "1. 选择不同的内容和风格层，输出有什么变化？\n", "1. 调整损失函数中的权重超参数。输出是否保留更多内容或减少更多噪点？\n", "1. 替换实验中的内容图像和风格图像，能创作出更有趣的合成图像吗？\n", "1. 我们可以对文本使用风格迁移吗？提示:可以参阅调查报告 :cite:`Hu.Lee.Aggarwal.ea.2020`。\n" ] }, { "cell_type": "markdown", "id": "8888edcd", "metadata": { "origin_pos": 57, "tab": [ "pytorch" ] }, "source": [ "[Discussions](https://discuss.d2l.ai/t/3300)\n" ] } ], "metadata": { "language_info": { "name": "python" }, "required_libs": [], "widgets": { "application/vnd.jupyter.widget-state+json": { "state": { "0a21ebb04f6e4afe9df09a7d7c6a0fe0": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "HBoxModel", "state": { "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "2.0.0", "_view_name": "HBoxView", "box_style": "", "children": [ "IPY_MODEL_3386849c303b40d18895bb91db97e325", "IPY_MODEL_155d363cdf40442e8faf86c2f0def49d", "IPY_MODEL_6840bc285801445eafe45e9cfc4a3216" ], "layout": "IPY_MODEL_96dfbfe851a544c09ae13797ba4d4198", "tabbable": null, "tooltip": null } }, "155d363cdf40442e8faf86c2f0def49d": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "FloatProgressModel", "state": { "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "2.0.0", "_view_name": "ProgressView", "bar_style": "success", "description": "", "description_allow_html": false, "layout": "IPY_MODEL_6e8fb617074d452eb01dbfe715d3827f", "max": 574673361.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_a24b158989e04d0e90ba67a8b670ce52", "tabbable": null, "tooltip": null, "value": 574673361.0 } }, "2d45b8677d764d7f9e04f2ab4f38d40b": { "model_module": "@jupyter-widgets/base", "model_module_version": "2.0.0", "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "2.0.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border_bottom": null, "border_left": null, "border_right": null, "border_top": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null } }, "3386849c303b40d18895bb91db97e325": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "HTMLModel", "state": { "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "2.0.0", "_view_name": "HTMLView", "description": "", "description_allow_html": false, "layout": "IPY_MODEL_2d45b8677d764d7f9e04f2ab4f38d40b", "placeholder": "​", "style": "IPY_MODEL_bc6bcbf06ef44f2585d8d82ac182ea56", "tabbable": null, "tooltip": null, "value": "100%" } }, "5223c9213fef443497e360398c21149f": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "HTMLStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "HTMLStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "StyleView", "background": null, "description_width": "", "font_size": null, "text_color": null } }, "57464fe9afd448b0a23eee081a9e085d": { "model_module": "@jupyter-widgets/base", "model_module_version": "2.0.0", "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "2.0.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border_bottom": null, "border_left": null, "border_right": null, "border_top": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null } }, "6840bc285801445eafe45e9cfc4a3216": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "HTMLModel", "state": { "_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "2.0.0", "_view_name": "HTMLView", "description": "", "description_allow_html": false, "layout": "IPY_MODEL_57464fe9afd448b0a23eee081a9e085d", "placeholder": "​", "style": "IPY_MODEL_5223c9213fef443497e360398c21149f", "tabbable": null, "tooltip": null, "value": " 548M/548M [00:10<00:00, 69.9MB/s]" } }, "6e8fb617074d452eb01dbfe715d3827f": { "model_module": "@jupyter-widgets/base", "model_module_version": "2.0.0", "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "2.0.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border_bottom": null, "border_left": null, "border_right": null, "border_top": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null } }, "96dfbfe851a544c09ae13797ba4d4198": { "model_module": "@jupyter-widgets/base", "model_module_version": "2.0.0", "model_name": "LayoutModel", "state": { "_model_module": "@jupyter-widgets/base", "_model_module_version": "2.0.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border_bottom": null, "border_left": null, "border_right": null, "border_top": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null } }, "a24b158989e04d0e90ba67a8b670ce52": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "ProgressStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "StyleView", "bar_color": null, "description_width": "" } }, "bc6bcbf06ef44f2585d8d82ac182ea56": { "model_module": "@jupyter-widgets/controls", "model_module_version": "2.0.0", "model_name": "HTMLStyleModel", "state": { "_model_module": "@jupyter-widgets/controls", "_model_module_version": "2.0.0", "_model_name": "HTMLStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "2.0.0", "_view_name": "StyleView", "background": null, "description_width": "", "font_size": null, "text_color": null } } }, "version_major": 2, "version_minor": 0 } } }, "nbformat": 4, "nbformat_minor": 5 }