3A_Convolutional_Neural_Networks.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to Convolutional Neural Networks (CNNs) in PyTorch"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Representing images digitally\n",
    "\n",
    "While convolutional neural networks (CNNs) see a wide variety of uses, they were originally designed for images, and CNNs are still most commonly used for vision-related tasks.\n",
    "For today, we'll primarily be focusing on CNNs for images.\n",
    "Before we dive into convolutions and neural networks, it's worth prefacing with how images are represented by a computer, as this understanding will inform some of our design choices.\n",
    "\n",
    "Previously, we saw an example of a digitized MNIST handwritten digit.\n",
    "Specifically, we represent it as an $H \\times W$ table, with the value of each element storing the intensity of the corresponding pixel.\n",
    "\n",
    "<img src=\"Figures/mnist_digital.PNG\" alt=\"mnist_digital\" style=\"width: 600px;\"/>\n",
    "\n",
    "With a 2D representation as above, we for the most part can only efficiently represent grayscale images.\n",
    "What if we want color?\n",
    "There are many schemes for storing color, but one of the most common ones is the [RGB color model](https://en.wikipedia.org/wiki/RGB_color_model).\n",
    "In such a system, we store 3 tables of pixel intensities (each called a *channel*), one each for the colors red, green, and blue (hence RGB), resulting in an $H \\times W \\times 3$ tensor.\n",
    "Pixel values for a particular channel indicate how much of the corresponding color the image has at a particular location.\n",
    "\n",
    "Let's load an image and see this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of the image tensor: (620, 1175, 3)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x24fccab9470>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import imageio\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# Read image\n",
    "im = imageio.imread(\"./Figures/chapel.jpg\")\n",
    "print(\"Shape of the image tensor: {}\".format(im.shape))\n",
    "plt.imshow(im)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the image we loaded has height and width of $620 \\times 1175$, with 3 channels corresponding to RGB.\n",
    "\n",
    "We can easily slice out and view individual color channels:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(620, 1175)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x24fccfb3c88>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Red channel\n",
    "im_red = im[:,:,0]\n",
    "print(im_red.shape)\n",
    "plt.imshow(im_red, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(620, 1175)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x24fcd021a20>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Blue channel\n",
    "im_blue = im[:,:,2]\n",
    "print(im_red.shape)\n",
    "plt.imshow(im_blue, cmap='gray')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While we have so far considered only 3 channel RGB images, there are many settings in which we may consider a different number of channels.\n",
    "For example, [hyperspectral imaging](https://en.wikipedia.org/wiki/Hyperspectral_imaging) uses a wide range of the electromagnetic spectrum to characterize a scene.\n",
    "Such modalities may have hundreds of channels or more.\n",
    "Additionally, we'll soon see that certain intermediate representations in a CNN can be considered images with many channels.\n",
    "However, we'll focus much of today's discussion to 1 channel grayscale and 3 channel RGB images as our inputs, as is most commonly the case in computer vision."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Convolutions\n",
    "Convolutional neural networks (CNNs) are a class of neural networks that have convolutional layers.\n",
    "CNNs are particularly effective for data that have spatial structures and correlations (e.g. images).\n",
    "We'll focus on CNNs applied to images in this tutorial.\n",
    "Recall that a multilayer perceptron (MLP) is entirely composed of fully connected layers, which are each a matrix multiply operation (and addition of a bias) followed by a non-linearity (e.g. sigmoid, ReLU). \n",
    "A convolutional layer is similar, except the matrix multiply operation is replaced with a convolution operation (in practice a cross-correlation). \n",
    "Note that a CNN need not be entirely composed of convolutional layers; in fact, many popular CNN architectures end in fully connected layers.\n",
    "\n",
    "As before, since we're building neural networks, let's start by loading PyTorch. We'll find NumPy useful as well, so we'll also import that here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn.functional as F"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Review: Fully connected layer\n",
    "In a fully connected layer, the input $x \\in \\mathbb R^{M \\times C_{in}}$ is a vector (or, rather a batch of vectors), where $M$ is the minibatch size and $C_{in}$ is the dimensionality of the input. \n",
    "We first matrix multiply the input $x$ by a weight matrix $W$.\n",
    "This weight matrix has dimensions $W \\in \\mathbb R^{C_{in} \\times C_{out}}$, where $C_{out}$ is the number of output units.\n",
    "We then add a bias for each output, which we do by adding $b \\in \\mathbb{R}^{C_{out}}$.\n",
    "The output $y \\in \\mathbb{R}^{M \\times C_{out}}$ of the fully connected layer then:\n",
    "\n",
    "\\begin{align*}\n",
    "y = \\text{ReLU}(x W + b)\n",
    "\\end{align*}\n",
    "\n",
    "Remember, the values of $W$ and $b$ are variables that we are trying to learn for our model. \n",
    "Below we have a visualization of what the matrix operation looks like (bias term and activation function omitted).\n",
    "\n",
    "<img src=\"Figures/mnist_matmul.PNG\" width=\"500\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input shape: torch.Size([100, 784])\n",
      "Output shape: torch.Size([100, 10])\n"
     ]
    }
   ],
   "source": [
    "# Create a random flat input vector\n",
    "x_fc = torch.randn(100, 784)\n",
    "\n",
    "# Create weight matrix variable\n",
    "W = torch.randn(784, 10)/np.sqrt(784)\n",
    "W.requires_grad_()\n",
    "\n",
    "# Create bias variable\n",
    "b = torch.zeros(10, requires_grad=True)\n",
    "\n",
    "# Apply fully connected layer\n",
    "y_preact = torch.matmul(x_fc, W) + b\n",
    "y = F.relu(y_preact)\n",
    "\n",
    "# Print input/output shape\n",
    "print(\"Input shape: {}\".format(x_fc.shape))\n",
    "print(\"Output shape: {}\".format(y.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Convolutional layer\n",
    "\n",
    "In a convolutional layer, we convolve the input $x$ with a convolutional kernel (aka filter), which we also call $W$, producing output $y$:\n",
    "\n",
    "\\begin{align*}\n",
    "y = \\text{ReLU}(W*x + b)\n",
    "\\end{align*}\n",
    "\n",
    "In the context of CNNs, the output $y$ is often referred to as feature maps. As with a fully connected layer, the goal is to learn $W$ and $b$ for our model.\n",
    "\n",
    "Unlike the input of a fully connected layer, which is $x \\in \\mathbb R^{M\\times C_{in}}$, the dimensionality of an image input is 4D: $x \\in \\mathbb R^{M \\times C_{in} \\times H_{in} \\times W_{in}}$, where $M$ is still the batch size, $C_{in}$ is the number of channels of the input (e.g. 3 for RGB), and $H_{in}$ and $W_{in}$ are the height and width of the image.\n",
    "\n",
    "The weight parameter $W$ is also different in a convolutional layer.\n",
    "Unlike the 2-D weight matrix for fully connected layers, the kernel is 4-D with dimensions $W \\in \\mathbb R^{C_{out} \\times C_{in} \\times H_K \\times W_K }$, where $H_K$ and $W_K$ are the kernel height and weight, respectively.\n",
    "A common choice for $H_K$ and $W_K$ is $H_K = W_K = 3$ or $5$, but this tends to vary depending on the architecture.\n",
    "Convolving the input with the kernel and adding a bias then gives an output $y \\in \\mathbb R^{M \\times C_{out} \\times H_{out} \\times W_{out}}$.\n",
    "If we use \"same\" padding and a stride of $1$ in our convolution (more on this later), our output will have the same spatial dimensions as the input: $H_{out}=H_{in}$ and $W_{out}=W_{in}$.\n",
    "\n",
    "If you're having trouble visualizing this operation in 4D, it's easier to think about for a single member of the minibatch, one convolutional kernel at a time. \n",
    "Consider a stack of $C_{out}$ number of kernels, each of which are 3D ($C_{in} \\times H_K \\times W_K $). \n",
    "This 3D volume is then slid across the input (which is also 3D: $C_{in} \\times H_{in} \\times W_{in}$) in the two spatial dimensions (along $H_{in}$ and $W_{in}$). \n",
    "The outputs of the multiplication of the kernel and the input at every location creates a single feature map that is $H_{out} \\times W_{out}$. \n",
    "Stacking the feature maps generated by each kernel gives the 3D output $C_{out} \\times H_{out} \\times W_{out} $.\n",
    "Repeat the process for all $M$ inputs in the minibatch, and we get a 4D output $M  \\times C_{out} \\times H_{out} \\times W_{out}$.\n",
    "\n",
    "<img src=\"Figures/conv_filters.png\" alt=\"Convolutional filters\" style=\"width: 400px;\"/>\n",
    "\n",
    "A few more things to note:\n",
    "- Notice the ordering of the dimensions of the input (batch, channels in, height, width).\n",
    "This is commonly referred to as $NCHW$ ordering.\n",
    "Many other languages and libraries (e.g. MATLAB, TensorFlow, the image example at the beginning of this notebook) instead default to the slightly different $NHWC$ ordering.\n",
    "PyTorch defaults to $NCHW$, as it more efficient computationally, especially with CUDA. \n",
    "- An additional argument for the convolution is the *stride*, which controls the how far we slide the convolutional filter as we move it along the input image. \n",
    "The convolutional operator, from its signal processing roots, by default considers a stride length of 1 in all dimensions, but in some situations we would like to consider strides more than 1 (or even less than 1). \n",
    "More on this later.\n",
    "- In the context of signal processing, convolutions usually result in outputs that are larger than the input size, which results from when the kernel \"hangs off the edge\" of the input on both sides. \n",
    "This might not always be desirable.\n",
    "We can control this by controlling the padding of the input.\n",
    "Typically, we use pad the input to ensure the output has the same spatial dimensions as the input (assuming stride of 1); this makes it easier for us to keep track of what the size of our model is.\n",
    "\n",
    "Let's implement this convolution operator in code.\n",
    "There is a convolution implementation in `torch.nn.functional`, which we use here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input shape: torch.Size([100, 1, 28, 28])\n",
      "Convolution output shape: torch.Size([100, 16, 28, 28])\n"
     ]
    }
   ],
   "source": [
    "# Create a random image input tensor\n",
    "x_cnn = torch.randn(100, 1, 28, 28)\n",
    "\n",
    "# Create convolutional kernel variable\n",
    "W1 = torch.randn(16, 1, 3, 3)/np.sqrt(1*3*3)\n",
    "W1.requires_grad_()\n",
    "\n",
    "# Create bias variable\n",
    "b1 = torch.zeros(16, requires_grad=True)\n",
    "\n",
    "# Apply convolutional layer\n",
    "conv1_preact = F.conv2d(x_cnn, W1, bias=b1, stride=1, padding=1)\n",
    "conv1 = F.relu(conv1_preact)\n",
    "\n",
    "# Print input/output shape\n",
    "print(\"Input shape: {}\".format(x_cnn.shape))\n",
    "print(\"Convolution output shape: {}\".format(conv1.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just like in a MLP, we can stack multiple of these convolutional layers. \n",
    "In the *Representing Images Digitally* section, we briefly mentioned considering images with channels more than 3.\n",
    "Observe that the input to the second layer (i.e. the output of the first layer) can be viewed as an \"image\" with $C_{out}$ channels.\n",
    "Instead of each channel representing a color content though, each channel effectively represents how much the original input image activated a particular convolutional kernel.\n",
    "Given $C_{out}$ kernels that are each $C_{in} \\times H_K \\times W_K$, this results in $C_{out}$ channels for the output of the convolution.\n",
    "\n",
    "Note that we need to change the dimensions of the convolutional kernel such that its input channels matches the number of output channels of the previous layer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Second convolution output shape: torch.Size([100, 32, 28, 28])\n"
     ]
    }
   ],
   "source": [
    "# 2nd layer variables\n",
    "W2 = torch.randn(32, 16, 3, 3)/np.sqrt(16*3*3)\n",
    "W2.requires_grad_()\n",
    "b2 = torch.zeros(32, requires_grad=True)\n",
    "\n",
    "# Apply 2nd convolutional layer\n",
    "conv2 = F.relu(F.conv2d(conv1, W2, bias=b2, stride=1, padding=1))\n",
    "\n",
    "# Print output shape\n",
    "print(\"Second convolution output shape: {}\".format(conv2.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In fact, we typically perform these convolution operations many times. \n",
    "Popular CNN architectures for image analysis today can be 100+ layers."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reshaping\n",
    "\n",
    "You'll commonly finding yourself needing to reshape tensors while building CNNs.\n",
    "The PyTorch function for doing so is `view()`. \n",
    "Anyone familiar with NumPy will find it very similar to `np.reshape()`.\n",
    "Importantly, the new dimensions must be chosen so that it is possible to rearrange the input into the shape of the output (i.e. the total number of elements must be the same).\n",
    "As with NumPy, you can optionally replace one of the dimensions with a `-1`, which tells `torch` to infer the missing dimension."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "M = torch.zeros(4, 3)\n",
    "\n",
    "M2 = M.view(1,1,12)\n",
    "M3 = M.view(2,1,2,3)\n",
    "M4 = M.view(-1,2,3)\n",
    "M5 = M.view(-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To get an idea of why reshaping is need in a CNN, let's look at a diagram of a simple CNN.\n",
    "\n",
    "<img src=\"Figures/mnist_cnn_ex.png\" alt=\"mnist_cnn_ex\" style=\"width: 500px;\"/>\n",
    "\n",
    "First of all, the CNN expects a 4D input, with the dimensions corresponding to `[batch, channel, height, width]`.\n",
    "Your data may not come in this format, so you may have to reshape it yourself."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([100, 1, 28, 28])\n"
     ]
    }
   ],
   "source": [
    "# Reshape flat input image into a 4D batched image input\n",
    "x_flat = torch.randn(100, 784)\n",
    "x_reshaped = x_flat.view(-1, 1, 28, 28)\n",
    "\n",
    "# Print input shape\n",
    "print(x_reshaped.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "CNN architectures also commonly contain fully connected layers or a softmax, as we're often interested in classification.\n",
    "Both of these expect 2D inputs with dimensions `[batch, dim]`, so you have to \"flatten\" a CNN's 4D output to 2D.\n",
    "For example, to flatten the convolutional feature maps we created earlier:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([100, 25088])\n"
     ]
    }
   ],
   "source": [
    "# Flatten convolutional feature maps into a vector\n",
    "h_flat = conv2.view(-1, 28*28*32)\n",
    "\n",
    "# Print output shape\n",
    "print(h_flat.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pooling and striding\n",
    "\n",
    "Almost all CNN architectures incorporate either pooling or striding. This is done for a number of reasons, including:\n",
    "- Dimensionality reduction: pooling and striding operations reduces computational complexity by shrinking the number of values passed to the next layer.\n",
    "For example, a 2x2 maxpool reduces the size of the feature maps by a factor of 4.\n",
    "- Translational invariance: Oftentimes in computer vision, we'd prefer that shifting the input by a few pixels doesn't change the output. Pooling and striding reduces sensitivity to exact pixel locations.\n",
    "- Increasing receptive field: by summarizing a window with a single value, subsequent convolutional kernels are seeing a wider swath of the original input image. For example, a max pool on some input followed by a 3x3 convolution results in a kernel \"seeing\" a 6x6 region instead of 3x3.\n",
    "\n",
    "#### Pooling\n",
    "The two most common forms of pooling are max pooling and average pooling. \n",
    "Both reduce values within a window to a single value, on a per-feature-map basis.\n",
    "Max pooling takes the maximum value of the window as the output value; average pooling takes the mean.\n",
    "\n",
    "<img src=\"Figures/maxpool.PNG\" alt=\"avg_vs_max\" style=\"width: 400px;\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of conv2 feature maps before pooling: torch.Size([100, 32, 28, 28])\n",
      "Shape of conv2 feature maps after max pooling: torch.Size([100, 32, 14, 14])\n",
      "Shape of conv2 feature maps after avg pooling: torch.Size([100, 32, 14, 14])\n"
     ]
    }
   ],
   "source": [
    "# Taking the output we've been working with so far, first print its current size\n",
    "print(\"Shape of conv2 feature maps before pooling: {0}\".format(conv2.shape))\n",
    "\n",
    "# Max pool and then print new shape\n",
    "max_pool2 = F.max_pool2d(conv2, kernel_size=2)\n",
    "print(\"Shape of conv2 feature maps after max pooling: {0}\".format(max_pool2.shape))\n",
    "\n",
    "# Average pool and then print new shape\n",
    "avg_pool2 = F.avg_pool2d(conv2, kernel_size=2)\n",
    "print(\"Shape of conv2 feature maps after avg pooling: {0}\".format(avg_pool2.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can confirm that the two types of pooling behave how we'd expect:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature map shape pre-pooling: torch.Size([1, 1, 4, 4])\n",
      "\n",
      "Max pool\n",
      "Shape: torch.Size([1, 1, 2, 2])\n",
      "tensor([[6., 8.],\n",
      "        [3., 4.]])\n",
      "\n",
      "Avg pool\n",
      "Shape: torch.Size([1, 1, 2, 2])\n",
      "tensor([[3.2500, 5.2500],\n",
      "        [2.0000, 2.0000]])\n"
     ]
    }
   ],
   "source": [
    "# Recreate values in pooling figure and make it 4D\n",
    "feature_map_fig = torch.tensor(np.array([[1,1,2,4],\n",
    "                                        [5,6,7,8],\n",
    "                                        [3,2,1,0],\n",
    "                                        [1,2,3,4]], dtype=np.float32))\n",
    "fmap_fig = feature_map_fig.view(1,1,4,4)\n",
    "print(\"Feature map shape pre-pooling: {}\".format(fmap_fig.shape))\n",
    "\n",
    "# Maxpool\n",
    "max_pool_fig = F.max_pool2d(fmap_fig, kernel_size=2)\n",
    "print(\"\\nMax pool\")\n",
    "print(\"Shape: {}\".format(max_pool_fig.shape))\n",
    "print(torch.squeeze(max_pool_fig))\n",
    "\n",
    "# Avgpool\n",
    "avg_pool_fig = F.avg_pool2d(fmap_fig, kernel_size=2)\n",
    "print(\"\\nAvg pool\")\n",
    "print(\"Shape: {}\".format(avg_pool_fig.shape))\n",
    "print(torch.squeeze(avg_pool_fig))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Striding\n",
    "One might expect that pixels in an image have high correlation with neighboring pixels, so we can save computation by skipping positions while sliding the convolutional kernel. \n",
    "By default, a CNN slides across the input one pixel at a time, which we call a stride of 1.\n",
    "By instead striding by 2, we skip calculating 75% of the values of the output feature map, which yields a feature map that's half the size in each spatial direction.\n",
    "Note, while pooling is an operation done after the convolution, striding is part of the convolution operation itself."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of conv1 feature maps: torch.Size([100, 16, 28, 28])\n",
      "Shape of conv2 feature maps with stride of 2: torch.Size([100, 32, 14, 14])\n"
     ]
    }
   ],
   "source": [
    "# Since striding is part of the convolution operation, we'll start with the feature maps before the 2nd convolution\n",
    "print(\"Shape of conv1 feature maps: {0}\".format(conv1.shape))\n",
    "\n",
    "# Apply 2nd convolutional layer, with striding of 2\n",
    "conv2_strided = F.relu(F.conv2d(conv1, W2, bias=b2, stride=2, padding=1))\n",
    "\n",
    "# Print output shape\n",
    "print(\"Shape of conv2 feature maps with stride of 2: {0}\".format(conv2_strided.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building a custom CNN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's revisit MNIST digit classification, but this time, we'll use the following CNN as our classifier: $5 \\times 5$ convolution -> $2 \\times 2$ max pool -> $5 \\times 5$ convolution -> $2 \\times 2$ max pool -> fully connected to $\\mathbb R^{256}$ -> fully connected to $\\mathbb R^{10}$ (prediction). \n",
    "ReLU activation functions will be used to impose non-linearities.\n",
    "Remember, convolutions produce 4-D outputs, and fully connected layers expect 2-D inputs, so tensors must be reshaped when transitioning from one to the other.\n",
    "\n",
    "We can build this CNN with the components introduced before, but as with the logistic regression example, it may prove helpful to instead organize our model with a `nn.Module`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch.nn as nn\n",
    "\n",
    "class MNIST_CNN(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.conv1 = nn.Conv2d(1, 32, kernel_size=5, padding=2)\n",
    "        self.conv2 = nn.Conv2d(32, 64, kernel_size=5, padding=2)\n",
    "        self.fc1 = nn.Linear(7*7*64, 256)\n",
    "        self.fc2 = nn.Linear(256, 10)\n",
    "\n",
    "    def forward(self, x):\n",
    "        # conv layer 1\n",
    "        x = self.conv1(x)\n",
    "        x = F.relu(x)\n",
    "        x = F.max_pool2d(x, kernel_size=2)\n",
    "        \n",
    "        # conv layer 2\n",
    "        x = self.conv2(x)\n",
    "        x = F.relu(x)\n",
    "        x = F.max_pool2d(x, kernel_size=2)\n",
    "        \n",
    "        # fc layer 1\n",
    "        x = x.view(-1, 7*7*64)\n",
    "        x = self.fc1(x)\n",
    "        x = F.relu(x)\n",
    "        \n",
    "        # fc layer 2\n",
    "        x = self.fc2(x)\n",
    "        return x        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice how our `nn.Module` contains several operation chained together.\n",
    "The code for submodule initialization, which creates all the stateful parameters associated with each operation, is placed in the `__init__()` function, where it is run once during object instantiation.\n",
    "Meanwhile, the code describing the forward pass, which is used every time the model is run, is placed in the `forward()` method.\n",
    "Printing an instantiated model shows the model summary:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MNIST_CNN(\n",
      "  (conv1): Conv2d(1, 32, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
      "  (conv2): Conv2d(32, 64, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
      "  (fc1): Linear(in_features=3136, out_features=256, bias=True)\n",
      "  (fc2): Linear(in_features=256, out_features=10, bias=True)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "model = MNIST_CNN()\n",
    "print(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can drop this model into our logistic regression training code, with few modifications beyond changing the model itself.\n",
    "A few other changes:\n",
    "- CNNs expect a 4-D input, so we no longer have to reshape the images before feeding them to our neural network.\n",
    "- Since CNNs are a little more complex than models we've worked with before, we're going to increase the number of epochs (complete passes through the training data) during training.\n",
    "- We switch from a vanilla stochastic gradient descent optimizer to the [Adam](https://arxiv.org/abs/1412.6980) optimizer, which tends to do well for neural networks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b55d6f340ec0428eac1c4ba7262fda73",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=3), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2da8ddac80d74ca3ab5fd5efcefbd0b0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=600), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fce41ea4b3b74dc2ab55a27f6578cbbf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=600), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e59571b38ab04469b7e1af29747dd20d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=600), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7cd916a53b1a42caa0d8fab7855e43bc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Test accuracy: 0.9908000230789185\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets, transforms\n",
    "from tqdm.notebook import tqdm, trange\n",
    "\n",
    "# Load the data\n",
    "mnist_train = datasets.MNIST(root=\"./datasets\", train=True, transform=transforms.ToTensor(), download=True)\n",
    "mnist_test = datasets.MNIST(root=\"./datasets\", train=False, transform=transforms.ToTensor(), download=True)\n",
    "train_loader = torch.utils.data.DataLoader(mnist_train, batch_size=100, shuffle=True)\n",
    "test_loader = torch.utils.data.DataLoader(mnist_test, batch_size=100, shuffle=False)\n",
    "\n",
    "## Training\n",
    "# Instantiate model  \n",
    "model = MNIST_CNN()  # <---- change here\n",
    "\n",
    "# Loss and Optimizer\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=0.001)  # <---- change here\n",
    "\n",
    "# Iterate through train set minibatchs \n",
    "for epoch in trange(3):  # <---- change here\n",
    "    for images, labels in tqdm(train_loader):\n",
    "        # Zero out the gradients\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        # Forward pass\n",
    "        x = images  # <---- change here \n",
    "        y = model(x)\n",
    "        loss = criterion(y, labels)\n",
    "        # Backward pass\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "## Testing\n",
    "correct = 0\n",
    "total = len(mnist_test)\n",
    "\n",
    "with torch.no_grad():\n",
    "    # Iterate through test set minibatchs \n",
    "    for images, labels in tqdm(test_loader):\n",
    "        # Forward pass\n",
    "        x = images  # <---- change here \n",
    "        y = model(x)\n",
    "\n",
    "        predictions = torch.argmax(y, dim=1)\n",
    "        correct += torch.sum((predictions == labels).float())\n",
    "\n",
    "print('Test accuracy: {}'.format(correct/total))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you are running this notebook on CPU, training this CNN might take a while.\n",
    "On the other hand, if you use a GPU, this model should train in seconds.\n",
    "This is why we usually prefer to use GPUs when we have them."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Torchvision"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Datasets and transforms\n",
    "\n",
    "As any experienced ML practioner will say, data wrangling is often half (sometimes even 90%) of the battle when building a model.\n",
    "Often, we have to write significant code to handle downloading, organizing, formatting, shuffling, pre-processing, augmenting, and batching examples. \n",
    "For popular datasets, we'd like to standardize data handling so that the comparisons we make are specific to the models themselves.\n",
    "\n",
    "Enter [Torchvision](https://pytorch.org/docs/stable/torchvision/index.html).\n",
    "Torchvision includes easy-to-use APIs for downloading and loading many popular vision datasets.\n",
    "We've previously seen this in action for downloading the MNIST dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import datasets\n",
    "\n",
    "mnist_train = datasets.MNIST(root=\"./datasets\", train=True, transform=transforms.ToTensor(), download=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Of course, there's [many more](https://pytorch.org/docs/stable/torchvision/datasets.html).\n",
    "Currently, datasets for image classification (e.g. MNIST, CIFAR, ImageNet), object detection (VOC, COCO, Cityscapes), and video action recognition (UCF101, Kinetics) are included.\n",
    "\n",
    "For formatting, pre-processing, and augmenting, [transforms](https://pytorch.org/docs/stable/torchvision/transforms.html) can come in handy.\n",
    "Again, we've seen this before (see above), when we used a transform to convert the MNIST data from PIL images to PyTorch tensors.\n",
    "However, transforms can be used for much more. \n",
    "Preprocessing steps like data whitening are common before feeding the data into the model.\n",
    "Also, in many cases, we use data augmentations to artificially inflate our dataset and learn invariances.\n",
    "Transforms are a versatile tool for all of these."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Leveraging popular convolutional neural networks\n",
    "\n",
    "While you certainly can build your own custom CNNs like we did above, more often than not, it's better to use one of the popular existing architectures. \n",
    "The Torchvision documentation has a [list of supported CNNs](https://pytorch.org/docs/stable/torchvision/models.html), as well as some performance characteristics. \n",
    "There's a number of reasons for using one of these CNNs instead of designing your own.\n",
    "\n",
    "First, for image datasets larger and more complex than MNIST (which is basically all of them), a fair amount network depth and width is often necessary.\n",
    "For example, some of the popular CNNs can be over 100 layers deep, with several tricks and details beyond what we've covered in this notebook.\n",
    "Coding all of this yourself has a high potential for error, especially when you're first getting started.\n",
    "Instead, you can create the CNN architecture using Torchvision, using a couple lines:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "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): BasicBlock(\n",
      "      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "    )\n",
      "  )\n",
      "  (layer2): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "    )\n",
      "  )\n",
      "  (layer3): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "      (downsample): Sequential(\n",
      "        (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): BasicBlock(\n",
      "      (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "    )\n",
      "  )\n",
      "  (layer4): Sequential(\n",
      "    (0): BasicBlock(\n",
      "      (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "      (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): BasicBlock(\n",
      "      (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      (relu): ReLU(inplace=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",
      "    )\n",
      "  )\n",
      "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
      "  (fc): Linear(in_features=512, out_features=1000, bias=True)\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "import torchvision.models as models\n",
    "resnet18 = models.resnet18()\n",
    "print(resnet18)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Loading a working CNN architecture in a couple lines can save a significant amount of time both implementing and debugging.\n",
    "\n",
    "The second, perhaps even more important, reason to use one of these existing architectures is the ability to use pre-trained weights.\n",
    "Early on in the recent resurgence of deep learning, people discovered that the weights of a CNN trained for ImageNet classification were highly transferable. \n",
    "For example, it is common to use the weights of an ImageNet-trained CNN as a weight initialization for other vision tasks, or even to freeze the bulk of the weights and only re-train the final classification layer(s) on a new task.\n",
    "This is significant, as in most settings, we rarely have enough labeled data to train a powerful CNN from scratch without overfitting.\n",
    "Loading pre-trained CNN is also pretty simple, involving an additional argument to the previous cell block:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# resnet18 = models.resnet18(pretrained=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font size=\"1\">*Note that this line of code is commented out, as running it will initiate a download of the pre-trained weights, which is a fairly large file.*</font>\n",
    "\n",
    "A full tutorial on using pre-trained CNNs is a little beyond the scope of this notebook.\n",
    "See [this tutorial](https://pytorch.org/tutorials/beginner/transfer_learning_tutorial.html) for an example."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Other computer vision tasks\n",
    "The base CNN architectures were often designed for image classification, but the same CNNs are often used as the backbone of most modern computer vision models.\n",
    "These other models often take this base CNN and include additional networks or make other architecture changes to adapt them to other tasks, such as object detection.\n",
    "Torchvision contains a few models (and pre-trained weights) for object detection, segmentation, and video action recognition.\n",
    "For example, to load a [Faster R-CNN](https://arxiv.org/abs/1506.01497) with a [ResNet50](https://arxiv.org/abs/1512.03385) convolutional feature extractor with [Feature Pyramid Networks](https://arxiv.org/abs/1612.03144) pre-trained on [MS COCO](http://cocodataset.org/#home):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# object_detector = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font size=\"1\">*Again, this line has been commented out to prevent loading a large network for this demo.*</font>\n",
    "\n",
    "Torchvision's selection of non-classification models is relatively light, and not particularly flexible.\n",
    "A number of other libraries are available, depending on the task.\n",
    "For example, for object detection and segmentation, Facebook AI Research's [Detectron2](https://github.com/facebookresearch/detectron2) is highly recommend."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "03c821f7cc304647bd7dd129ece95022": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_a9254a18982444839701b4f3c20b1113",
       "max": 600,
       "style": "IPY_MODEL_feba469114dd47aaa7a13bc21b134099",
       "value": 600
      }
     },
     "06865ecae49c4d328421c474fea799a0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "09e99596521e40948b044aea1dd49ceb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "0a271956e4984a0a9ba69c1f66809bfb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6ca5eb76ff8441cc996cdbc8bcfb5ab3",
       "style": "IPY_MODEL_eb483872c21d43d4a42b34091b4eb8c1",
       "value": " 600/600 [05:31&lt;00:00,  1.81it/s]"
      }
     },
     "0edf80c192134246a102f49c1ac4a77e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "13e3067366c34bfdbd817690ebeb5685": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "15ada75a5a524e0d80956489b169679b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1d856c2263ec4b1f97f46727b8d85f12": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "1e7d4fd24b054b3daa60ae7dd2726ec5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_6933964202a74264b0ed487f706297a3",
        "IPY_MODEL_b8590f98675a453ba38b85163a285eb4"
       ],
       "layout": "IPY_MODEL_ba35f1bbfe5549f394fcb2a1b544584b"
      }
     },
     "234c9d9fbbaa4aaa84dfd3598983e96a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "26e0d143239647a2832a3d656f56fd12": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "2779485b2912403f80088f678d5a01c0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "28a7ac6f810e4b4a94503dbda7ebf78c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "2d0476b37d43454585abb625c1c50181": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "2da8ddac80d74ca3ab5fd5efcefbd0b0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_c047c4ef8c3c4f1fa8508267806dc5c2",
        "IPY_MODEL_35e061d7f6aa4817a081a3ce2416bc7d"
       ],
       "layout": "IPY_MODEL_62ef7f1b4c0f4a008c366a21cec33c17"
      }
     },
     "2e67a47c98ae4b19a913886efd194566": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_bb81062e385c414abc8c3f8a70716327",
       "max": 600,
       "style": "IPY_MODEL_731297e272154e4ab71e7fd6a6859309",
       "value": 600
      }
     },
     "3152dae7252a4ee3958217f48a9bb0c7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "322ecba10f114730888b6b15a8694fea": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "34484a12bd9f494a9bb6c6e479524246": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "35e061d7f6aa4817a081a3ce2416bc7d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_5dfd065b0af94b72b0a797c1659d58e4",
       "style": "IPY_MODEL_13e3067366c34bfdbd817690ebeb5685",
       "value": " 600/600 [04:52&lt;00:00,  2.05it/s]"
      }
     },
     "371bf011ef8c488fac1fae1c63dc7dae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_6f06679d39f84b75a08a05e478e0bc89",
       "style": "IPY_MODEL_cadc3c445e7444f5af275f20c674019b",
       "value": " 100/100 [00:18&lt;00:00,  5.43it/s]"
      }
     },
     "40143d1b96fa4cbdb4ea4f915ac0e7c7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_34484a12bd9f494a9bb6c6e479524246",
       "style": "IPY_MODEL_d274246f37a643ba879a390b788b3c5b",
       "value": " 100/100 [00:18&lt;00:00,  5.32it/s]"
      }
     },
     "456c3368617d4acd9b81809d2643f847": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "526b7dd6def34ac68dfd667d9ecc7373": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "55697ad8b13f40eb900098274954edff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_15ada75a5a524e0d80956489b169679b",
       "style": "IPY_MODEL_783576334fac49bbb171283f192725f2",
       "value": " 3/3 [28:40&lt;00:00, 573.52s/it]"
      }
     },
     "5c35d2a0736845aca0c743382fb21435": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_968de7cc30cc49219edb47c77ac19835",
       "style": "IPY_MODEL_26e0d143239647a2832a3d656f56fd12",
       "value": " 600/600 [06:08&lt;00:00,  1.63it/s]"
      }
     },
     "5c778265dd2e4f55990114a410be07e7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_1d856c2263ec4b1f97f46727b8d85f12",
       "style": "IPY_MODEL_3152dae7252a4ee3958217f48a9bb0c7",
       "value": 100
      }
     },
     "5dfd065b0af94b72b0a797c1659d58e4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5dfea9beebf347babc96499b6ef6e55b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "5fa779faf29448bda9ecb19f04db2363": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "62ef7f1b4c0f4a008c366a21cec33c17": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "68181c92926e4b209ebddc72e0a18a90": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6933964202a74264b0ed487f706297a3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": "  0%",
       "layout": "IPY_MODEL_88946e6e9e3d45c2a58c6712f7b78fef",
       "max": 3,
       "style": "IPY_MODEL_28a7ac6f810e4b4a94503dbda7ebf78c"
      }
     },
     "6ca5eb76ff8441cc996cdbc8bcfb5ab3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6ebc572251db40378e419fa603857e39": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "6f06679d39f84b75a08a05e478e0bc89": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "6faf5a48daed4ceb8b42b918f5890080": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "70bfccb7e1144fc3a04cd0ca72ae63eb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_f67d8b42eb0b4156929a609efc6e9708",
        "IPY_MODEL_55697ad8b13f40eb900098274954edff"
       ],
       "layout": "IPY_MODEL_6faf5a48daed4ceb8b42b918f5890080"
      }
     },
     "731297e272154e4ab71e7fd6a6859309": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "783576334fac49bbb171283f192725f2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "78f433d8b75e4a08a36465e145079cf1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "7adead9aa0e54151b5bae6180e87eaf2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7cd916a53b1a42caa0d8fab7855e43bc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_5c778265dd2e4f55990114a410be07e7",
        "IPY_MODEL_371bf011ef8c488fac1fae1c63dc7dae"
       ],
       "layout": "IPY_MODEL_d1536ee4c2b94f5d819c52777db82fd1"
      }
     },
     "7d6bdfc513274cbc997b980fd0bf6d4b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_b4d97b0af743462fa7ab900547f28cec",
       "max": 3,
       "style": "IPY_MODEL_b2374c1841914c56a1c0a0ce0fdd8f52",
       "value": 3
      }
     },
     "7e775f18ba894cad88b398c06546947d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "7f1e0c9662724b0a94f6412f23d4f48b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_526b7dd6def34ac68dfd667d9ecc7373",
       "style": "IPY_MODEL_7adead9aa0e54151b5bae6180e87eaf2",
       "value": 100
      }
     },
     "806fdae35ae54e4398bc198978f1143a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_06865ecae49c4d328421c474fea799a0",
       "style": "IPY_MODEL_6ebc572251db40378e419fa603857e39",
       "value": " 600/600 [17:51&lt;00:00,  1.79s/it]"
      }
     },
     "8334eba58bc34ceeb9c15edb42401d3f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_2e67a47c98ae4b19a913886efd194566",
        "IPY_MODEL_9301e0239a174dfc9b221ec2a02af44f"
       ],
       "layout": "IPY_MODEL_5dfea9beebf347babc96499b6ef6e55b"
      }
     },
     "88946e6e9e3d45c2a58c6712f7b78fef": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "88e7221416214664b30773fed9a88614": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "8cecc0e1039c4eee90dfa2db1afab3c1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "90722b21a44e4541a5333511f9e1e3c8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "9301e0239a174dfc9b221ec2a02af44f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_d9151e052ab545159f94594f8d155fd7",
       "style": "IPY_MODEL_e45c675d8d044d7c9550a20e40b0539e",
       "value": " 600/600 [05:17&lt;00:00,  1.89it/s]"
      }
     },
     "968de7cc30cc49219edb47c77ac19835": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "96c234a72edc42ca80e536e3a54c2acc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a229f6b4ca20475ea3a5745d18d7131c": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_322ecba10f114730888b6b15a8694fea",
       "style": "IPY_MODEL_a8a4eb83499a453787b42fc46b93a0ba",
       "value": " 223/600 [01:52&lt;03:02,  2.06it/s]"
      }
     },
     "a83b638605be4a9ba18c4529729de544": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "a8a4eb83499a453787b42fc46b93a0ba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "a9254a18982444839701b4f3c20b1113": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "abc221956ffb4578aec132dddb6ef921": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "acb0ac60dd0643c6b1ae204792f442d4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "acc0927c8ae7415f9df9b49feeadfded": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "description": " 37%",
       "layout": "IPY_MODEL_2d0476b37d43454585abb625c1c50181",
       "max": 600,
       "style": "IPY_MODEL_88e7221416214664b30773fed9a88614",
       "value": 223
      }
     },
     "aff6325150904f36b4dae74956e87bff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_03c821f7cc304647bd7dd129ece95022",
        "IPY_MODEL_806fdae35ae54e4398bc198978f1143a"
       ],
       "layout": "IPY_MODEL_78f433d8b75e4a08a36465e145079cf1"
      }
     },
     "b18a79c6165442b7ac686c25eb915836": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "b1ce822ea1f44e69b04dfabf5d6b176b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_7f1e0c9662724b0a94f6412f23d4f48b",
        "IPY_MODEL_40143d1b96fa4cbdb4ea4f915ac0e7c7"
       ],
       "layout": "IPY_MODEL_8cecc0e1039c4eee90dfa2db1afab3c1"
      }
     },
     "b2374c1841914c56a1c0a0ce0fdd8f52": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "b4d97b0af743462fa7ab900547f28cec": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b55d6f340ec0428eac1c4ba7262fda73": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_7d6bdfc513274cbc997b980fd0bf6d4b",
        "IPY_MODEL_c4a75cbcd5784be48fdfb09dbacbd3a3"
       ],
       "layout": "IPY_MODEL_acb0ac60dd0643c6b1ae204792f442d4"
      }
     },
     "b7e465f909304b8da8f19b67743c925f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "b8590f98675a453ba38b85163a285eb4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_decefda30d7f4c59a15bfd79f082045b",
       "style": "IPY_MODEL_cf367f49b29745fe88c33e2b2f46f8e4",
       "value": " 0/3 [00:00&lt;?, ?it/s]"
      }
     },
     "b9c879ca534d4691a94bd015b53d7967": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "ba35f1bbfe5549f394fcb2a1b544584b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "bb81062e385c414abc8c3f8a70716327": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "c047c4ef8c3c4f1fa8508267806dc5c2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_456c3368617d4acd9b81809d2643f847",
       "max": 600,
       "style": "IPY_MODEL_dccbf0d594854417975e06c2251f4f92",
       "value": 600
      }
     },
     "c18eea10e4ff4974b07ff2253f5f779d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_09e99596521e40948b044aea1dd49ceb",
       "max": 600,
       "style": "IPY_MODEL_d6c628ea9a2347579e8b04a34dae7e83",
       "value": 600
      }
     },
     "c4a75cbcd5784be48fdfb09dbacbd3a3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_90722b21a44e4541a5333511f9e1e3c8",
       "style": "IPY_MODEL_b18a79c6165442b7ac686c25eb915836",
       "value": " 3/3 [16:33&lt;00:00, 331.17s/it]"
      }
     },
     "c6a40c98b57d4d99ad89de894e76aafb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_a83b638605be4a9ba18c4529729de544",
       "max": 600,
       "style": "IPY_MODEL_abc221956ffb4578aec132dddb6ef921",
       "value": 600
      }
     },
     "cadc3c445e7444f5af275f20c674019b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "cf367f49b29745fe88c33e2b2f46f8e4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d0160fcf3a1c4409aa417bc54dcd2029": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_f9ae08e19e6540a2ab502af5ff456067",
        "IPY_MODEL_fe492931e45e4f9995c1302483fde1ba"
       ],
       "layout": "IPY_MODEL_b9c879ca534d4691a94bd015b53d7967"
      }
     },
     "d1536ee4c2b94f5d819c52777db82fd1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "d274246f37a643ba879a390b788b3c5b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "d34d970aa5e84bd8b30f3c0af6851044": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_acc0927c8ae7415f9df9b49feeadfded",
        "IPY_MODEL_a229f6b4ca20475ea3a5745d18d7131c"
       ],
       "layout": "IPY_MODEL_68181c92926e4b209ebddc72e0a18a90"
      }
     },
     "d6c628ea9a2347579e8b04a34dae7e83": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "d9151e052ab545159f94594f8d155fd7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "dccbf0d594854417975e06c2251f4f92": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     },
     "de12b4d028d240d88666934d195132db": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "decefda30d7f4c59a15bfd79f082045b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {}
     },
     "e45c675d8d044d7c9550a20e40b0539e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "e59571b38ab04469b7e1af29747dd20d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_c6a40c98b57d4d99ad89de894e76aafb",
        "IPY_MODEL_0a271956e4984a0a9ba69c1f66809bfb"
       ],
       "layout": "IPY_MODEL_0edf80c192134246a102f49c1ac4a77e"
      }
     },
     "eb483872c21d43d4a42b34091b4eb8c1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "description_width": ""
      }
     },
     "f67d8b42eb0b4156929a609efc6e9708": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_de12b4d028d240d88666934d195132db",
       "max": 3,
       "style": "IPY_MODEL_2779485b2912403f80088f678d5a01c0",
       "value": 3
      }
     },
     "f9ae08e19e6540a2ab502af5ff456067": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "IntProgressModel",
      "state": {
       "bar_style": "success",
       "description": "100%",
       "layout": "IPY_MODEL_234c9d9fbbaa4aaa84dfd3598983e96a",
       "max": 600,
       "style": "IPY_MODEL_7e775f18ba894cad88b398c06546947d",
       "value": 600
      }
     },
     "fce41ea4b3b74dc2ab55a27f6578cbbf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "children": [
        "IPY_MODEL_c18eea10e4ff4974b07ff2253f5f779d",
        "IPY_MODEL_5c35d2a0736845aca0c743382fb21435"
       ],
       "layout": "IPY_MODEL_b7e465f909304b8da8f19b67743c925f"
      }
     },
     "fe492931e45e4f9995c1302483fde1ba": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "layout": "IPY_MODEL_96c234a72edc42ca80e536e3a54c2acc",
       "style": "IPY_MODEL_5fa779faf29448bda9ecb19f04db2363",
       "value": " 600/600 [05:32&lt;00:00,  1.81it/s]"
      }
     },
     "feba469114dd47aaa7a13bc21b134099": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "description_width": "initial"
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}