Cycle_GAN_for_Style_Transfer_(Image_to_image_translation).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "U-6zLspJQlb-"
   },
   "source": [
    "# Cycle GAN model for Image Style Transfer (Image to Image Translations)\n",
    "\n",
    "Using summer2winter_yosemite dataset (you can find it as summer2winter_yosemite.zip)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "ExUc1jCt-_zK"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import torch\n",
    "from torch.utils.data import DataLoader\n",
    "from torchvision.utils import make_grid\n",
    "import torchvision.datasets as datasets\n",
    "import torchvision.transforms as transforms\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "qwduc1NE9oBH"
   },
   "outputs": [],
   "source": [
    "class Preprocessor:\n",
    "    def __init__(self, image_class, train_dir, test_dir, image_size, batch_size):\n",
    "        '''\n",
    "        image_class --  str, refers to the class name of the images in which the first order parent containing them is named after\n",
    "        train_dir -- str, refers to the directory containing the image_class directories (2nd order parents) of the training image\n",
    "        test_dir -- str, refers to the directory containing the image_class directories (2nd order parents) of the test image\n",
    "        image_size -- tuple of the size of the image\n",
    "        batch_size -- int, number of images in a single patch\n",
    "        '''\n",
    "        self.image_class = image_class\n",
    "        self.train_dir = train_dir\n",
    "        self.test_dir = test_dir\n",
    "        self.image_size = image_size\n",
    "        self.batch_size = batch_size\n",
    "        self.init_bool = False\n",
    "\n",
    "    def get_data_loaders(self):\n",
    "        # return the loaders if the loaders are already intialized\n",
    "        if self.init_bool:\n",
    "            return self.train_loader, self.test_loader\n",
    "\n",
    "        # get position of images\n",
    "        train_path = './' + self.train_dir\n",
    "        train_path = os.path.join(train_path, self.image_class)\n",
    "        test_path = './' + self.test_dir\n",
    "        #test_path = os.path.join(test_path, self.image_class)\n",
    "\n",
    "        # defining transform\n",
    "        transform = transforms.Compose([transforms.Resize(self.image_size),\n",
    "                                        transforms.ToTensor()])\n",
    "\n",
    "        # define datasets\n",
    "        training_set = datasets.ImageFolder(train_path, transform = transform)\n",
    "        test_set = datasets.ImageFolder(test_path, transform = transform)\n",
    "\n",
    "        # defining data loaders\n",
    "        train_loader = DataLoader(dataset = training_set, batch_size = self.batch_size, shuffle=True)\n",
    "        test_loader = DataLoader(dataset = test_set, batch_size = self.batch_size, shuffle = True)\n",
    "\n",
    "        # declare as initialized\n",
    "        self.train_loader = train_loader\n",
    "        self.test_loader = test_loader\n",
    "        self.init_bool = True\n",
    "\n",
    "        return train_loader, test_loader\n",
    "\n",
    "    def imshow(self, image):\n",
    "        img = image.numpy()\n",
    "        plt.imshow(np.transpose(img, (1,2,0))) # reshaping axes to visualize with np\n",
    "        plt.show();\n",
    "\n",
    "    def show_many_images(self, group='training', fig_size=(12,8)):\n",
    "        '''\n",
    "        show many images of the class\n",
    "        group -- 'training' or 'testing'\n",
    "        fig_size -- figure size\n",
    "        '''\n",
    "        images, _ = next(iter(self.train_loader)) if group == 'training' else\\\n",
    "            next(iter(self.test_loader))\n",
    "        fig = plt.figure(figsize=fig_size)\n",
    "        self.imshow(make_grid(images))\n",
    "\n",
    "    def scale(self,x, feature_scale=(-1,1)):\n",
    "        minn, maxx = feature_scale\n",
    "        x = x*(maxx - minn) - maxx\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "7ugibaxSAxLu"
   },
   "outputs": [],
   "source": [
    "# let X be summer style and Y be winter style\n",
    "X_type = Preprocessor(image_class='summer', \n",
    "                      train_dir='summer2winter_yosemite', \n",
    "                      test_dir='summer2winter_yosemite/test_summer', \n",
    "                      image_size=128, \n",
    "                      batch_size=16)\n",
    "\n",
    "Y_type = Preprocessor(image_class='winter', \n",
    "                      train_dir='summer2winter_yosemite', \n",
    "                      test_dir='summer2winter_yosemite/test_winter', \n",
    "                      image_size=128, \n",
    "                      batch_size=16)\n",
    "\n",
    "train_loader_X, test_loader_X = X_type.get_data_loaders()\n",
    "train_loader_Y, test_loader_Y = Y_type.get_data_loaders()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 220
    },
    "id": "P9QbdfRLGZVK",
    "outputId": "83962994-cda0-4dea-b19e-2f3b4b6defec"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sanity check show many images\n",
    "X_type.show_many_images()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 553
    },
    "id": "n58EBgvBJd6F",
    "outputId": "dc523679-d9db-488d-8b16-739611b55fba"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([3, 128, 128])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sanity check scale and show image\n",
    "img, _ = iter(X_type.train_loader).next()\n",
    "img = img[0]\n",
    "print(img.shape)\n",
    "X_type.imshow(img)\n",
    "\n",
    "scaled_image = X_type.scale(img)\n",
    "X_type.imshow(scaled_image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "udZonVt5L0jE"
   },
   "source": [
    "## Defining the models\n",
    "### 1. Discrimintator\n",
    "\n",
    "Both Discriminators DX and DY have the same properties\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "wWjLiH76L9Uh"
   },
   "outputs": [],
   "source": [
    "from torch import nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class ConvSuper(nn.Module):\n",
    "    '''\n",
    "    This class acts as a base class with a function that is used in all children classes that are convolutional\n",
    "    '''\n",
    "    def _conv(self, in_channels, out_channels, kernel_size, stride=2, padding=1, batch_norm=True):\n",
    "        ''' Creates a convolutional layer with optional batch normalization set to True.\n",
    "            Used to downsample the image.\n",
    "        '''\n",
    "\n",
    "        layers = [nn.Conv2d(in_channels=in_channels, out_channels=out_channels,\n",
    "                               kernel_size=kernel_size, stride=stride, padding=padding, bias=False)]\n",
    "        \n",
    "        if batch_norm:\n",
    "            layers.append(nn.BatchNorm2d(out_channels))\n",
    "\n",
    "        return nn.Sequential(*layers)\n",
    "\n",
    "    def _deconv(self, in_channels, out_channels, kernel_size, stride=2, padding=1, batch_norm=True):\n",
    "        '''Creates a transpose convolutional layer with optional batch normalization.\n",
    "            Used to upsample the image.\n",
    "        '''\n",
    "\n",
    "        layers = [nn.ConvTranspose2d(in_channels, out_channels, kernel_size, stride, \n",
    "                                     padding, bias=False)]\n",
    "\n",
    "        # optional batchnorm\n",
    "        if batch_norm:\n",
    "            layers.append(nn.BatchNorm2d(out_channels))\n",
    "        \n",
    "        return nn.Sequential(*layers)\n",
    "\n",
    "class Discriminator(ConvSuper):\n",
    "    def __init__(self, in_channels=3, conv_dim=64):\n",
    "        super(Discriminator, self).__init__()\n",
    "        \n",
    "        # class attributes\n",
    "        self.conv_dim = conv_dim\n",
    "        self.in_channels = in_channels\n",
    "\n",
    "        # defining the conv layers\n",
    "        ### inspection #1--> depth = 3, w=128, h=128\n",
    "        self.conv1 = self._conv(in_channels, conv_dim, kernel_size=4, batch_norm=False)\n",
    "        ### inspection #2--> depth = 64, w=64, h=64\n",
    "        self.conv2 = self._conv(conv_dim, 2*conv_dim, kernel_size=4)\n",
    "        ### inspection #3--> depth = 128, w=32, h=32\n",
    "        self.conv3 = self._conv(2*conv_dim, 4*conv_dim, kernel_size=4)\n",
    "        ### inspection #4--> depth = 256, w=16, h=16\n",
    "        self.conv4 = self._conv(4*conv_dim, 8*conv_dim, kernel_size=4)\n",
    "        ### inspection #5--> depth = 512, w=8, h=8\n",
    "\n",
    "        # The following conv layer will act as a classification layer\n",
    "        self.conv5 = self._conv(8*conv_dim, 1, kernel_size=4, stride=1, batch_norm=False) \n",
    "        ### inspection #6 (Final) --> depth = 1, w=1, h=1  \n",
    "\n",
    "    def forward(self, X):\n",
    "        # encoders\n",
    "        out = F.relu(self.conv1(X))\n",
    "        out = F.relu(self.conv2(out))\n",
    "        out = F.relu(self.conv3(out))\n",
    "        out = F.relu(self.conv4(out))\n",
    "\n",
    "        # classifier\n",
    "        out = F.relu(self.conv5(out))\n",
    "        return out\n",
    "\n",
    "class ResidualBlock(ConvSuper):\n",
    "    '''Defines a residual block, adds input X to a convolutional layer that is applied to X\n",
    "    with the same size input and output. The resblocks allow the model to learn effective transformation\n",
    "    from a domian to another\n",
    "    '''\n",
    "    def __init__(self, conv_dim):\n",
    "        # initializing super class\n",
    "        super(ResidualBlock,self).__init__()\n",
    "\n",
    "        # class attributes\n",
    "        self.conv_dim = conv_dim\n",
    "\n",
    "        # layers -- No change in dimentions\n",
    "        self.conv1 = self._conv(in_channels=conv_dim, out_channels=conv_dim,\n",
    "                                kernel_size=3, stride=1, padding=1, batch_norm=True)\n",
    "        self.conv2 = self._conv(in_channels=conv_dim, out_channels=conv_dim,\n",
    "                                kernel_size=3, stride=1, padding=1, batch_norm=True)\n",
    "\n",
    "    def forward(self, X):\n",
    "        '''\n",
    "        feed forward function\n",
    "        X -- input data\n",
    "        '''\n",
    "        out = F.relu(self.conv1(X))\n",
    "        out = X + self.conv2(X)\n",
    "        return X\n",
    "\n",
    "class Generator(ConvSuper):\n",
    "    def __init__(self, conv_dim=64, n_res_blocks=6):\n",
    "        super(Generator, self).__init__()\n",
    "\n",
    "        # class attributes\n",
    "        self.conv_dim = conv_dim\n",
    "        self.n_res_blocks = n_res_blocks\n",
    "\n",
    "        ######## encoders ########\n",
    "        ### inspection #1--> depth = 3, w=128, h=128\n",
    "        self.conv1 = self._conv(3,conv_dim,4)\n",
    "        ### inspection #2--> depth = 64, w=64, h=64\n",
    "        self.conv2 = self._conv(conv_dim, conv_dim*2, 4)\n",
    "        ### inspection #3--> depth = 128, w=32, h=32\n",
    "        self.conv3 = self._conv(conv_dim*2, conv_dim*4, 4)\n",
    "        ### inspection #4--> depth = 256, w=16, h=16\n",
    "\n",
    "        ######## residual blocks ########\n",
    "        self.res_blocks = self.__intialize_res(conv_dim*4, n_res_blocks)\n",
    "\n",
    "        ######## decoder ########\n",
    "        ### inspection #5--> depth = 256, w=16, h=16\n",
    "        self.deconv1 = self._deconv(conv_dim*4, conv_dim*2, 4)\n",
    "        ### inspection #6--> depth = 128, w=32, h=32 \n",
    "        self.deconv2 = self._deconv(conv_dim*2, conv_dim, 4)\n",
    "        ### inspection #7--> depth = 64, w=64, h=64 \n",
    "        self.deconv3 = self._deconv(conv_dim, 3, 4, batch_norm=False)  # no batch_norm\n",
    "        ### inspection #8--> depth = 3, w=128, h=128\n",
    "\n",
    "    def forward(self, X):\n",
    "        '''Takes an image as an input and returns the transformed image as an output'''\n",
    "        # encoders\n",
    "        out = F.relu(self.conv1(X))\n",
    "        out = F.relu(self.conv2(out))\n",
    "        out = F.relu(self.conv3(out))\n",
    "\n",
    "        # resblocks\n",
    "        out = self.res_blocks(out)\n",
    "\n",
    "        # decoders\n",
    "        out = F.relu(self.deconv1(out))\n",
    "        out = F.relu(self.deconv2(out))\n",
    "\n",
    "        # tanh applied at the last layer\n",
    "        out = F.tanh(self.deconv3(out))\n",
    "\n",
    "        return out\n",
    "\n",
    "    def __intialize_res(self,conv_dim, n_blocks=1):\n",
    "        layers = []\n",
    "        for i in range(n_blocks):\n",
    "            layers.append(ResidualBlock(conv_dim))\n",
    "\n",
    "        return nn.Sequential(*layers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "y_LeapQPvh1u",
    "outputId": "83ad16b9-55a3-4728-c85b-8ef2c9e76194"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([7, 7])\n"
     ]
    }
   ],
   "source": [
    "# sanity check for the discriminator class\n",
    "DX = Discriminator()\n",
    "out = DX(img.unsqueeze(0)).squeeze().squeeze()\n",
    "print(out.shape)\n",
    "# delete to free memory\n",
    "del DX\n",
    "del out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "MNcCJngqydRC",
    "outputId": "1f047683-852e-4e90-c170-685e22639630"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([3, 128, 128])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:1698: UserWarning: nn.functional.tanh is deprecated. Use torch.tanh instead.\n",
      "  warnings.warn(\"nn.functional.tanh is deprecated. Use torch.tanh instead.\")\n"
     ]
    }
   ],
   "source": [
    "# Sanity check for the generator class \n",
    "GX = Generator()\n",
    "out = GX(img.unsqueeze(0)).squeeze().squeeze();\n",
    "print(out.shape)\n",
    "# delete to free memory\n",
    "del GX\n",
    "del out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "5bE0OvRkd6Kv"
   },
   "source": [
    "## Creating Model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "DR9vbcSud5BV",
    "outputId": "0f10547e-3b3b-4c66-e656-e381278d231b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Configuring on GPU\n",
      "                     G_X2Y                    \n",
      "-----------------------------------------------\n",
      "Generator(\n",
      "  (conv1): Sequential(\n",
      "    (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv2): Sequential(\n",
      "    (0): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv3): Sequential(\n",
      "    (0): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (res_blocks): Sequential(\n",
      "    (0): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (2): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (3): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (4): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (5): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "  )\n",
      "  (deconv1): Sequential(\n",
      "    (0): ConvTranspose2d(256, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (deconv2): Sequential(\n",
      "    (0): ConvTranspose2d(128, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (deconv3): Sequential(\n",
      "    (0): ConvTranspose2d(64, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      ")\n",
      "\n",
      "                     G_Y2X                    \n",
      "-----------------------------------------------\n",
      "Generator(\n",
      "  (conv1): Sequential(\n",
      "    (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv2): Sequential(\n",
      "    (0): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv3): Sequential(\n",
      "    (0): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (res_blocks): Sequential(\n",
      "    (0): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (1): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (2): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (3): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (4): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "    (5): ResidualBlock(\n",
      "      (conv1): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "      (conv2): Sequential(\n",
      "        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "      )\n",
      "    )\n",
      "  )\n",
      "  (deconv1): Sequential(\n",
      "    (0): ConvTranspose2d(256, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (deconv2): Sequential(\n",
      "    (0): ConvTranspose2d(128, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (deconv3): Sequential(\n",
      "    (0): ConvTranspose2d(64, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      ")\n",
      "\n",
      "                      D_X                      \n",
      "-----------------------------------------------\n",
      "Discriminator(\n",
      "  (conv1): Sequential(\n",
      "    (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      "  (conv2): Sequential(\n",
      "    (0): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv3): Sequential(\n",
      "    (0): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv4): Sequential(\n",
      "    (0): Conv2d(256, 512, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv5): Sequential(\n",
      "    (0): Conv2d(512, 1, kernel_size=(4, 4), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "  )\n",
      ")\n",
      "\n",
      "                      D_Y                      \n",
      "-----------------------------------------------\n",
      "Discriminator(\n",
      "  (conv1): Sequential(\n",
      "    (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "  )\n",
      "  (conv2): Sequential(\n",
      "    (0): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv3): Sequential(\n",
      "    (0): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv4): Sequential(\n",
      "    (0): Conv2d(256, 512, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)\n",
      "    (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
      "  )\n",
      "  (conv5): Sequential(\n",
      "    (0): Conv2d(512, 1, kernel_size=(4, 4), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "  )\n",
      ")\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from helpers import save_samples, checkpoint\n",
    "from tqdm import tqdm\n",
    "\n",
    "class CycleGan:\n",
    "    def __init__(self,g_conv_dim=64, d_conv_dim=64, n_res_blocks=6):\n",
    "        self.create_CGAN_models(g_conv_dim, d_conv_dim, n_res_blocks)\n",
    "        self.__print_models()\n",
    "\n",
    "    def create_CGAN_models(self,g_conv_dim, d_conv_dim, n_res_blocks):\n",
    "        '''Builds cycle gan generators and descriminator'''\n",
    "        # Instintiate generators\n",
    "        self.G_X2Y = Generator(g_conv_dim, n_res_blocks)\n",
    "        self.G_Y2X = Generator(g_conv_dim, n_res_blocks)\n",
    "\n",
    "        # Instantiate discriminatros\n",
    "        self.D_X = Discriminator(3,d_conv_dim)\n",
    "        self.D_Y = Discriminator(3,d_conv_dim)\n",
    "\n",
    "        # config on GPU if available\n",
    "        train_on_GPU = torch.cuda.is_available()\n",
    "        if train_on_GPU:\n",
    "            print(\"Configuring on GPU\")\n",
    "        else:\n",
    "            print(\"Configuring on CPU\")\n",
    "\n",
    "        device = 'cuda' if train_on_GPU else 'cpu'\n",
    "        self.G_X2Y.to(device)\n",
    "        self.G_Y2X.to(device)\n",
    "        self.D_X.to(device)\n",
    "        self.D_Y.to(device)\n",
    "\n",
    "        #return self.G_X2Y, self.G_Y2X, self.D_X, self.D_Y\n",
    "\n",
    "    def train(self, trainloader_X, trainloader_Y, testloader_X, testloader_Y,\n",
    "              g_optimizer, d_x_optimizer, d_y_optimizer, lambda_w = 10, epochs=1200, print_every=10, sample_every=100):\n",
    "        '''\n",
    "        trainloader_X: first domain train loader \n",
    "        trainloader_Y: second domain train loader \n",
    "        testloader_X: first domain test loader\n",
    "        testloader_Y: second domain test loader\n",
    "        g_optimizer: generator optimizer\n",
    "        d_x_optimize: first domain discrminator optimizer \n",
    "        d_y_optimizer: second domain discriminator optimizer\n",
    "        lambda_w = 10: lambda weight of the cycle consistency loss with default value 10\n",
    "        epochs=1200: training epochs with default value 1200 \n",
    "        print_every=10: print inspection every (default=10) epochs \n",
    "        sample_every=100: save sample every (default=100) epochs\n",
    "        '''\n",
    "        \n",
    "        # keep track of the training losses through out time for visualization purposes\n",
    "        losses = {\n",
    "            'D_X':[],\n",
    "            'D_Y':[],\n",
    "            'G': []\n",
    "        }\n",
    "\n",
    "        # get some fixed data from both domains X and Y, for sampling purposes\n",
    "        # These images are held constants throughout the training that allows the inspection of the \n",
    "        # model performance\n",
    "        # take 1 image from both batches\n",
    "        fixed_X = X_type.scale(next(iter(testloader_X))[0])\n",
    "        fixed_Y = Y_type.scale(next(iter(testloader_Y))[0])\n",
    "\n",
    "        # batches per epoch\n",
    "        iter_X = iter(trainloader_X)\n",
    "        iter_Y = iter(testloader_Y)\n",
    "        batches_per_epoch = min(len(iter_X), len(iter_Y))\n",
    "\n",
    "        for e in tqdm(range(1, epochs+1)):\n",
    "            # reset iterations for each epoch\n",
    "            if e % batches_per_epoch ==0:\n",
    "                iter_X = iter(trainloader_X)\n",
    "                iter_Y = iter(trainloader_Y)\n",
    "\n",
    "            # get images\n",
    "            X, _ = next(iter_X)\n",
    "            X = X_type.scale(X)\n",
    "\n",
    "            Y, _ = next(iter_Y)\n",
    "            Y = Y_type.scale(Y)\n",
    "\n",
    "            # check if GPU is available and move images to GPU accordingly\n",
    "            device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "            X = X.to(device)\n",
    "            Y = Y.to(device)\n",
    "\n",
    "            ##################################\n",
    "            #### Train the Discriminators ####\n",
    "            ##################################\n",
    "\n",
    "            ## First: D_X real and fake loss components\n",
    "            d_x_optimizer.zero_grad()\n",
    "\n",
    "            # 1. Compute the discriminator losses on real images\n",
    "            real_X = self.D_X(X)\n",
    "            D_X_real_loss = self.__real_mse_loss(real_X)\n",
    "\n",
    "            # 2. Generate fake images that look like the domain X from domain Y\n",
    "            fake_X = self.G_Y2X(Y)\n",
    "\n",
    "            # 3.  Compute the fake losses for D_X\n",
    "            fake_X = self.D_X(fake_X)\n",
    "            D_X_fake_loss = self.__fake_mse_loss(fake_X)\n",
    "\n",
    "            # 4. Compute the total loss and perform backpropagation\n",
    "            D_X_loss = D_X_real_loss + D_X_fake_loss\n",
    "            D_X_loss.backward()\n",
    "            d_x_optimizer.step()\n",
    "\n",
    "\n",
    "            ## Second: D_Y, real and fake components\n",
    "\n",
    "            d_y_optimizer.zero_grad()\n",
    "\n",
    "            # 1. Compute the discriminator losses on real images\n",
    "            real_Y = self.D_Y(Y)\n",
    "            D_Y_real_loss = self.__real_mse_loss(real_Y)\n",
    "\n",
    "            # 2. Generate fake images that look like domain Y based on images from domain X\n",
    "            fake_Y = self.G_X2Y(X)\n",
    "\n",
    "            # 3. Compute fake loss for D_Y\n",
    "            fake_Y = self.D_Y(fake_Y)\n",
    "            D_Y_fake_loss = self.__fake_mse_loss(fake_Y)\n",
    "\n",
    "            # 4. Compute total losses and perform backpropagation\n",
    "            D_Y_loss = D_Y_real_loss + D_Y_fake_loss\n",
    "            D_Y_loss.backward()\n",
    "            d_y_optimizer.step()\n",
    "\n",
    "            ##############################\n",
    "            #### Train the Generators ####\n",
    "            ##############################\n",
    "\n",
    "            g_optimizer.zero_grad()\n",
    "\n",
    "            #### First generate fake X images and reconstructed Y ###\n",
    "\n",
    "            # 1. generate fake images that look like domain X\n",
    "            fake_X = self.G_Y2X(Y)\n",
    "\n",
    "            # 2. Compute the loss based on domain X\n",
    "            out_X = self.D_X(fake_X)\n",
    "            g_Y2X_loss = self.__real_mse_loss(out_X)\n",
    "\n",
    "            # 3. Compute the reconstructed Y\n",
    "            recon_Y = self.G_X2Y(fake_X)\n",
    "\n",
    "            # 4. Compute the cycle consistency loss\n",
    "            recon_Y_loss = self.__cycle_consistency_loss(Y, recon_Y, lambda_w)\n",
    "\n",
    "            #### Second generate fake Y images and reconstructed X ###\n",
    "\n",
    "             # 1. generate fake images that look like domain X\n",
    "            fake_Y = self.G_X2Y(X)\n",
    "\n",
    "            # 2. Compute the loss based on domain X\n",
    "            out_Y = self.D_Y(fake_Y)\n",
    "            g_X2Y_loss = self.__real_mse_loss(out_Y)\n",
    "\n",
    "            # 3. Compute the reconstructed Y\n",
    "            recon_X = self.G_Y2X(fake_Y)\n",
    "\n",
    "            # 4. Compute the cycle consistency loss\n",
    "            recon_X_loss = self.__cycle_consistency_loss(X, recon_X, lambda_w = lambda_w)\n",
    "\n",
    "            ## addup all generator losses and perform backpropagation\n",
    "            g_total_loss = g_X2Y_loss + g_Y2X_loss + recon_X_loss + recon_Y_loss\n",
    "            g_total_loss.backward()\n",
    "            g_optimizer.step()\n",
    "\n",
    "            # Print the log info\n",
    "            if e % print_every == 0:\n",
    "                # append real and fake discriminator losses and the generator loss\n",
    "                losses['D_X'].append(D_X_loss.item())\n",
    "                losses['D_Y'].append(D_Y_loss.item())\n",
    "                losses['G'].append(g_total_loss.item())\n",
    "                #losses.append((D_X_loss.item(), D_Y_loss.item(), g_total_loss.item()))\n",
    "                print('Epoch [{:5d}/{:5d}] | d_X_loss: {:6.4f} | d_Y_loss: {:6.4f} | g_total_loss: {:6.4f}'.format(\n",
    "                        e, epochs, D_X_loss.item(), D_Y_loss.item(), g_total_loss.item()))\n",
    "\n",
    "            # Save the generated samples\n",
    "            if e % sample_every == 0:\n",
    "                self.G_Y2X.eval() # set generators to eval mode for sample generation\n",
    "                self.G_X2Y.eval()\n",
    "                save_samples(e, fixed_Y, fixed_X, self.G_Y2X, self.G_X2Y, batch_size=16)\n",
    "\n",
    "                # return to training mode\n",
    "                self.G_Y2X.train()\n",
    "                self.G_X2Y.train()\n",
    "\n",
    "        return losses\n",
    "\n",
    "    def __real_mse_loss(self, D_out):\n",
    "        return torch.mean((D_out-1)**2)\n",
    "\n",
    "    def __fake_mse_loss(self, D_out):\n",
    "        return torch.mean(D_out**2)\n",
    "\n",
    "    def __cycle_consistency_loss(self, real_im, reconstructed_im, lambda_w):\n",
    "        # calculate reconstruction loss\n",
    "        recon_loss = torch.mean(torch.abs(real_im - reconstructed_im))\n",
    "\n",
    "        # return weighted loss\n",
    "        return lambda_w*recon_loss    \n",
    "\n",
    "    def __print_models(self):\n",
    "        \"\"\"Prints model information for the generators and discriminators.\n",
    "        \"\"\"\n",
    "        print(\"                     G_X2Y                    \")\n",
    "        print(\"-----------------------------------------------\")\n",
    "        print(self.G_X2Y)\n",
    "        print()\n",
    "\n",
    "        print(\"                     G_Y2X                    \")\n",
    "        print(\"-----------------------------------------------\")\n",
    "        print(self.G_Y2X)\n",
    "        print()\n",
    "\n",
    "        print(\"                      D_X                      \")\n",
    "        print(\"-----------------------------------------------\")\n",
    "        print(self.D_X)\n",
    "        print()\n",
    "\n",
    "        print(\"                      D_Y                      \")\n",
    "        print(\"-----------------------------------------------\")\n",
    "        print(self.D_Y)\n",
    "        print()\n",
    "\n",
    "# Creating Models\n",
    "cgan_model = CycleGan()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "XV-m9926f9_X"
   },
   "outputs": [],
   "source": [
    "import torch.optim as optim\n",
    "\n",
    "# hyperparams for Adam optimizer\n",
    "lr=0.0002\n",
    "beta1=0.5\n",
    "beta2=0.999 # default value\n",
    "\n",
    "g_params = list(cgan_model.G_X2Y.parameters()) + list(cgan_model.G_Y2X.parameters())  # Get generator parameters\n",
    "\n",
    "# Create optimizers for the generators and discriminators\n",
    "g_optimizer = optim.Adam(g_params, lr, [beta1, beta2])\n",
    "d_x_optimizer = optim.Adam(cgan_model.D_X.parameters(), lr, [beta1, beta2])\n",
    "d_y_optimizer = optim.Adam(cgan_model.D_Y.parameters(), lr, [beta1, beta2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "mg57edVN4D8_",
    "outputId": "eb186ffe-1c46-4c82-c8c9-69218374e447"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  2%|▏         | 99/4000 [05:32<1:51:04,  1.71s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  100/ 4000] | d_X_loss: 0.4971 | d_Y_loss: 0.5364 | g_total_loss: 4.1819\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000100-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "  2%|▎         | 100/4000 [05:35<2:02:45,  1.89s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000100-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  5%|▍         | 199/4000 [07:00<31:22,  2.02it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  200/ 4000] | d_X_loss: 0.4390 | d_Y_loss: 0.4866 | g_total_loss: 3.9752\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000200-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "  5%|▌         | 200/4000 [07:02<50:18,  1.26it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000200-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  7%|▋         | 299/4000 [07:38<17:55,  3.44it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  300/ 4000] | d_X_loss: 0.5763 | d_Y_loss: 0.4182 | g_total_loss: 3.8054\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000300-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "  8%|▊         | 300/4000 [07:39<39:38,  1.56it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000300-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 10%|▉         | 399/4000 [08:07<15:15,  3.94it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  400/ 4000] | d_X_loss: 0.4292 | d_Y_loss: 0.4628 | g_total_loss: 3.2417\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000400-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 10%|█         | 400/4000 [08:09<37:57,  1.58it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000400-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 12%|█▏        | 499/4000 [08:34<14:55,  3.91it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  500/ 4000] | d_X_loss: 0.4595 | d_Y_loss: 0.5396 | g_total_loss: 3.9933\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000500-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 12%|█▎        | 500/4000 [08:35<35:22,  1.65it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000500-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 15%|█▍        | 599/4000 [09:01<14:31,  3.90it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  600/ 4000] | d_X_loss: 0.3221 | d_Y_loss: 0.3544 | g_total_loss: 3.5405\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000600-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 15%|█▌        | 600/4000 [09:03<36:44,  1.54it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000600-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 17%|█▋        | 699/4000 [09:28<14:10,  3.88it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  700/ 4000] | d_X_loss: 0.4290 | d_Y_loss: 0.4957 | g_total_loss: 3.3518\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000700-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 18%|█▊        | 700/4000 [09:30<34:01,  1.62it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000700-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 20%|█▉        | 799/4000 [09:55<13:48,  3.86it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  800/ 4000] | d_X_loss: 0.3698 | d_Y_loss: 0.1560 | g_total_loss: 3.7453\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000800-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 20%|██        | 800/4000 [09:57<33:08,  1.61it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000800-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 22%|██▏       | 899/4000 [10:22<13:15,  3.90it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [  900/ 4000] | d_X_loss: 1.0000 | d_Y_loss: 0.2101 | g_total_loss: 4.1889\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000900-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 22%|██▎       | 900/4000 [10:24<32:53,  1.57it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-000900-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 25%|██▍       | 999/4000 [10:49<13:00,  3.85it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1000/ 4000] | d_X_loss: 0.5269 | d_Y_loss: 0.3075 | g_total_loss: 3.4095\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001000-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 25%|██▌       | 1000/4000 [10:51<31:48,  1.57it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001000-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 27%|██▋       | 1099/4000 [11:17<12:34,  3.85it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1100/ 4000] | d_X_loss: 0.2832 | d_Y_loss: 0.1870 | g_total_loss: 3.6869\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001100-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 28%|██▊       | 1100/4000 [11:18<25:53,  1.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001100-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 30%|██▉       | 1199/4000 [11:44<12:06,  3.86it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1200/ 4000] | d_X_loss: 0.4055 | d_Y_loss: 0.1847 | g_total_loss: 4.2012\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001200-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 30%|███       | 1200/4000 [11:45<29:49,  1.56it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001200-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 32%|███▏      | 1299/4000 [12:11<11:38,  3.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1300/ 4000] | d_X_loss: 0.2082 | d_Y_loss: 0.6595 | g_total_loss: 3.5393\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001300-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 32%|███▎      | 1300/4000 [12:12<30:04,  1.50it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001300-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 35%|███▍      | 1399/4000 [12:38<11:14,  3.86it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1400/ 4000] | d_X_loss: 0.2286 | d_Y_loss: 0.1810 | g_total_loss: 3.3304\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001400-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 35%|███▌      | 1400/4000 [12:39<23:49,  1.82it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001400-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 37%|███▋      | 1499/4000 [13:05<10:46,  3.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1500/ 4000] | d_X_loss: 0.1922 | d_Y_loss: 0.0367 | g_total_loss: 4.5777\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001500-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 38%|███▊      | 1500/4000 [13:07<26:43,  1.56it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001500-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 40%|███▉      | 1599/4000 [13:32<10:19,  3.88it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1600/ 4000] | d_X_loss: 0.0359 | d_Y_loss: 0.0094 | g_total_loss: 4.0015\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001600-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 40%|████      | 1600/4000 [13:34<25:17,  1.58it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001600-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 42%|████▏     | 1699/4000 [13:59<09:50,  3.90it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1700/ 4000] | d_X_loss: 0.0677 | d_Y_loss: 0.0126 | g_total_loss: 3.7243\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001700-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 42%|████▎     | 1700/4000 [14:01<24:08,  1.59it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001700-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 45%|████▍     | 1799/4000 [14:26<09:21,  3.92it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1800/ 4000] | d_X_loss: 0.0074 | d_Y_loss: 0.0413 | g_total_loss: 4.3573\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001800-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 45%|████▌     | 1800/4000 [14:28<23:32,  1.56it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001800-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 47%|████▋     | 1899/4000 [14:53<09:00,  3.89it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 1900/ 4000] | d_X_loss: 0.0203 | d_Y_loss: 0.0055 | g_total_loss: 4.2073\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001900-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 48%|████▊     | 1900/4000 [14:54<23:53,  1.47it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-001900-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 50%|████▉     | 1999/4000 [15:20<08:26,  3.95it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2000/ 4000] | d_X_loss: 0.0294 | d_Y_loss: 0.0076 | g_total_loss: 4.1601\n",
      "Saved samples_cyclegan/sample-002000-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 50%|█████     | 2000/4000 [15:21<17:50,  1.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002000-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 52%|█████▏    | 2099/4000 [15:46<08:08,  3.90it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2100/ 4000] | d_X_loss: 0.0044 | d_Y_loss: 0.0055 | g_total_loss: 3.7839\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002100-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 52%|█████▎    | 2100/4000 [15:48<20:19,  1.56it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002100-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 55%|█████▍    | 2199/4000 [16:13<07:38,  3.92it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2200/ 4000] | d_X_loss: 0.0453 | d_Y_loss: 0.0081 | g_total_loss: 3.4692\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002200-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 55%|█████▌    | 2200/4000 [16:14<19:23,  1.55it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002200-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 57%|█████▋    | 2299/4000 [16:39<07:20,  3.86it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2300/ 4000] | d_X_loss: 0.0473 | d_Y_loss: 0.0492 | g_total_loss: 3.6430\n",
      "Saved samples_cyclegan/sample-002300-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 57%|█████▊    | 2300/4000 [16:41<15:28,  1.83it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002300-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 60%|█████▉    | 2399/4000 [17:06<06:44,  3.96it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2400/ 4000] | d_X_loss: 0.0042 | d_Y_loss: 0.0102 | g_total_loss: 3.5255\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002400-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 60%|██████    | 2400/4000 [17:07<11:49,  2.26it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002400-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 62%|██████▏   | 2499/4000 [17:32<06:19,  3.96it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2500/ 4000] | d_X_loss: 0.0096 | d_Y_loss: 0.0094 | g_total_loss: 3.5724\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002500-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 62%|██████▎   | 2500/4000 [17:33<10:58,  2.28it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002500-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 65%|██████▍   | 2599/4000 [17:58<05:51,  3.99it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2600/ 4000] | d_X_loss: 0.0102 | d_Y_loss: 0.0325 | g_total_loss: 3.9139\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002600-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 65%|██████▌   | 2600/4000 [17:59<10:17,  2.27it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002600-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 67%|██████▋   | 2699/4000 [18:24<05:32,  3.92it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2700/ 4000] | d_X_loss: 0.5914 | d_Y_loss: 0.0298 | g_total_loss: 3.4092\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002700-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 68%|██████▊   | 2700/4000 [18:25<09:44,  2.22it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002700-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 70%|██████▉   | 2799/4000 [18:50<05:00,  3.99it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2800/ 4000] | d_X_loss: 0.0051 | d_Y_loss: 0.0044 | g_total_loss: 4.2876\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002800-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 70%|███████   | 2800/4000 [18:51<08:52,  2.25it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002800-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 72%|███████▏  | 2899/4000 [19:16<04:35,  4.00it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 2900/ 4000] | d_X_loss: 0.0051 | d_Y_loss: 0.0034 | g_total_loss: 3.6277\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002900-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 72%|███████▎  | 2900/4000 [19:17<07:59,  2.29it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-002900-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 75%|███████▍  | 2999/4000 [19:42<04:14,  3.94it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3000/ 4000] | d_X_loss: 0.0057 | d_Y_loss: 0.4757 | g_total_loss: 3.4144\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003000-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 75%|███████▌  | 3000/4000 [19:43<07:29,  2.22it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003000-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 77%|███████▋  | 3099/4000 [20:08<03:46,  3.97it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3100/ 4000] | d_X_loss: 0.0020 | d_Y_loss: 0.0061 | g_total_loss: 3.7569\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003100-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 78%|███████▊  | 3100/4000 [20:09<06:38,  2.26it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003100-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 80%|███████▉  | 3199/4000 [20:34<03:21,  3.97it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3200/ 4000] | d_X_loss: 0.0036 | d_Y_loss: 0.0026 | g_total_loss: 3.3067\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003200-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 80%|████████  | 3200/4000 [20:35<05:54,  2.26it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003200-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 82%|████████▏ | 3299/4000 [21:00<02:58,  3.92it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3300/ 4000] | d_X_loss: 0.0069 | d_Y_loss: 0.0069 | g_total_loss: 3.6982\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003300-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 82%|████████▎ | 3300/4000 [21:01<05:10,  2.25it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003300-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 85%|████████▍ | 3399/4000 [21:26<02:32,  3.93it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3400/ 4000] | d_X_loss: 0.0051 | d_Y_loss: 0.0027 | g_total_loss: 3.6498\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003400-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 85%|████████▌ | 3400/4000 [21:27<04:21,  2.29it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003400-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 87%|████████▋ | 3499/4000 [21:52<02:05,  3.98it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3500/ 4000] | d_X_loss: 0.0053 | d_Y_loss: 0.0025 | g_total_loss: 3.6344\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003500-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 88%|████████▊ | 3500/4000 [21:53<03:37,  2.30it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003500-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 90%|████████▉ | 3599/4000 [22:18<01:42,  3.91it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3600/ 4000] | d_X_loss: 0.3631 | d_Y_loss: 0.0025 | g_total_loss: 3.6983\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      " 90%|█████████ | 3600/4000 [22:19<02:53,  2.31it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003600-X-Y.png\n",
      "Saved samples_cyclegan/sample-003600-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 92%|█████████▏| 3699/4000 [22:44<01:16,  3.93it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3700/ 4000] | d_X_loss: 0.0060 | d_Y_loss: 0.0046 | g_total_loss: 3.5712\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003700-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 92%|█████████▎| 3700/4000 [22:45<02:14,  2.23it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003700-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 95%|█████████▍| 3799/4000 [23:10<00:50,  3.96it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3800/ 4000] | d_X_loss: 0.0017 | d_Y_loss: 0.0015 | g_total_loss: 3.4518\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003800-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 95%|█████████▌| 3800/4000 [23:11<01:29,  2.23it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003800-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 97%|█████████▋| 3899/4000 [23:36<00:25,  3.98it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 3900/ 4000] | d_X_loss: 0.0044 | d_Y_loss: 0.0027 | g_total_loss: 3.5239\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003900-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 98%|█████████▊| 3900/4000 [23:37<00:44,  2.26it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-003900-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████▉| 3999/4000 [24:02<00:00,  3.97it/s]WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 254.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [ 4000/ 4000] | d_X_loss: 0.0031 | d_Y_loss: 0.0022 | g_total_loss: 3.3868\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-004000-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 4000/4000 [24:03<00:00,  2.77it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved samples_cyclegan/sample-004000-Y-X.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "n_epochs = 4000 # keep this small when testing if a model first works\n",
    "\n",
    "losses = cgan_model.train(train_loader_X, train_loader_Y, \n",
    "                          test_loader_X, test_loader_Y, g_optimizer=g_optimizer, d_x_optimizer=d_x_optimizer,\n",
    "                          d_y_optimizer=d_y_optimizer,lambda_w = 10, epochs=n_epochs, print_every=100, sample_every=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 281
    },
    "id": "u3gOuWYIWK-y",
    "outputId": "03f37a7d-88d9-436f-c717-d2c4a546294e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "def visualize_losses(losses):\n",
    "    '''\n",
    "    losses ob of all the losses in the system.\n",
    "    losses keys[\n",
    "        'D_X'\n",
    "        'D_Y'\n",
    "        'G'\n",
    "    ]\n",
    "    '''\n",
    "    plt.title(\"Losses of the Cycle Gan Model\")\n",
    "    plt.plot(losses['D_X'],label='Discriminator Summer')\n",
    "    plt.plot(losses['D_Y'], label='Discriminator Winter')\n",
    "    plt.plot(losses['G'], label = 'Generator')\n",
    "    plt.legend()\n",
    "    plt.show();\n",
    "\n",
    "visualize_losses(losses)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Vr_zMV5zLIwu",
    "outputId": "f8083e20-b5a0-4141-d50f-ccb74112c869"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "0it [00:00, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Configuring on GPU\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n",
      "WARNING:root:Lossy conversion from float64 to uint8. Range [0.0, 255.0]. Convert image to uint8 prior to saving to suppress this warning.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved test_samples/sample-000000-X-Y.png\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "0it [00:00, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved test_samples/sample-000000-Y-X.png\n",
      "\n",
      "\n",
      "\t\t\t Testing Loss\n",
      "\n",
      "Testing loss Discriminator X\t:\t 0.03791981190443039\n",
      "Testing loss Discriminator Y\t:\t 0.04710119217634201\n",
      "Testing loss Generators\t\t:\t 0.15848776698112488\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    " def test_model(model, batch_size=16):\n",
    "    g_loss = 0\n",
    "    dx_loss = 0\n",
    "    dy_loss = 0\n",
    "    train_on_GPU = torch.cuda.is_available()\n",
    "    if train_on_GPU:\n",
    "        print(\"Configuring on GPU\")\n",
    "    else:\n",
    "        print(\"Configuring on CPU\")\n",
    "\n",
    "    device = 'cuda' if train_on_GPU else 'cpu'\n",
    "    model.D_X.eval()\n",
    "    model.D_Y.eval()\n",
    "    model.G_X2Y.eval()\n",
    "    model.G_Y2X.eval()   \n",
    "    itr = 0\n",
    "    with torch.no_grad(): \n",
    "        for x, y in tqdm(zip(test_loader_X, test_loader_Y)):\n",
    "            \n",
    "            x,_ = x\n",
    "            y,_ = y\n",
    "            x = x.to(device)\n",
    "            y = y.to(device)\n",
    "            x = X_type.scale(x)\n",
    "            y = Y_type.scale(y)\n",
    "            #1. calculate the real error\n",
    "            real_ims_res_x = model.D_X(x)\n",
    "            real_ims_err_x = torch.mean((real_ims_res_x-1)**2)\n",
    "            real_ims_res_y = model.D_Y(y)\n",
    "            real_ims_err_y = torch.mean((real_ims_res_y-1)**2)\n",
    "\n",
    "            #2. generate fake images from the batch\n",
    "            fake_ims_x = model.G_Y2X(y)\n",
    "            fake_ims_y = model.G_X2Y(x)\n",
    "            fake_ims_err_x = torch.mean(model.D_X(fake_ims_y**2))\n",
    "            fake_ims_err_y = torch.mean(model.D_Y(fake_ims_x**2))\n",
    "            \n",
    "            #3. calculate the Discriminator error\n",
    "            d_error_x = fake_ims_err_x + real_ims_err_x\n",
    "            d_error_y = fake_ims_err_x + real_ims_err_y\n",
    "            \n",
    "            #4. calculate the real error with these fakeimages\n",
    "            recon_im_y = model.G_X2Y(fake_ims_x)\n",
    "            recon_im_x = model.G_Y2X(fake_ims_y)\n",
    "            recon_im_y_logits = model.D_Y(recon_im_y)\n",
    "            recon_im_x_logits = model.D_X(recon_im_x)\n",
    "            \n",
    "            #5. calculate real adverserial error on the generated image\n",
    "            adv_real_error_y =  torch.mean((model.D_Y(fake_ims_x)-1)**2)\n",
    "            adv_real_error_x =  torch.mean((model.D_Y(fake_ims_y)-1)**2)\n",
    "            \n",
    "            #6. calculate the error of the generated image in discriminator\n",
    "            recon_im_x_err = torch.mean((recon_im_x_logits-1)**2)\n",
    "            recon_im_y_err = torch.mean((recon_im_y_logits-1)**2)\n",
    "            \n",
    "            #7. calculate cycle loss and add losses to the current loss\n",
    "            dx_loss+= d_error_x/batch_size\n",
    "            dy_loss+= d_error_y/batch_size\n",
    "            ccl_x = torch.mean(torch.abs(x - recon_im_x))\n",
    "            ccl_y = torch.mean(torch.abs(y - recon_im_y))\n",
    "            g_loss += (ccl_x + ccl_y + recon_im_x_err + recon_im_y_err)/batch_size\n",
    "\n",
    "            # save test samples\n",
    "            save_samples(itr, y, x, model.G_Y2X, model.G_X2Y, batch_size=16, sample_dir='test_samples')\n",
    "\n",
    "            # increment iteration\n",
    "            itr+=1\n",
    "\n",
    "            return dx_loss.item(), dy_loss.item(), g_loss.item()\n",
    "\n",
    "\n",
    "loss_dx, loss_dy, loss_g = test_model(cgan_model);\n",
    "\n",
    "print()\n",
    "print()\n",
    "print(\"\\t\\t\\t Testing Loss\")\n",
    "print()\n",
    "print(\"Testing loss Discriminator X\\t:\\t\", loss_dx)\n",
    "print(\"Testing loss Discriminator Y\\t:\\t\", loss_dy)\n",
    "print(\"Testing loss Generators\\t\\t:\\t\", loss_g)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Lr7nW2deNi0g"
   },
   "source": [
    "\n",
    "\n",
    "## Drawbacks \n",
    "1. Up-sampling with low resolution reduces the the quality of the image. May consider some other techniques (mentioned in the the Consideratons section).\n",
    "\n",
    "2. There are some images that are wrongly generated, seems like the model treats any kind of wood as a plant and hence, generates the green leaves in winter to summer generation, and vice versa.\n",
    "\n",
    "## Considerations\n",
    "\n",
    "1. Re-implementing the model with conditional GAN may lead to higher resolution\n",
    "2. Better tuning may lead to better results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "eZ0JibgqQU7Q"
   },
   "source": [
    "#### Note that: Training samples are found in 'samples_cyclegan' folder and the test results are found in the 'test_samples' directory."
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Cycle GAN for Style Transfer (Image to image translation).ipynb",
   "provenance": []
  },
  "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}