gan_mnist.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6a688681",
   "metadata": {},
   "source": [
    "# GANs - MNIST Digit Dataset\n",
    "\n",
    "![gan structure](https://miro.medium.com/max/1400/1*ZKUo2QtHasnr8-RiqeJ_YA.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "82b7cf67",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "626d19c9",
   "metadata": {},
   "source": [
    "### Helper Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6b41fd9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "### SCALE & NORMALIZE MNIST dataset ###\n",
    "# Scale data from range [0-255] to [0-1]\n",
    "# Normalize data to range [-1, 1]\n",
    "def scale_norm(x):\n",
    "    # Scale\n",
    "    x = (x/255)\n",
    "    # Normalize | Without unsqueeze_, Normalize() will throw a size error\n",
    "    x = Normalize(mean=(0.5,), std=(0.5,))(x.unsqueeze_(0))\n",
    "    x = x.squeeze_(0)   # Reverse engineering\n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "bd50c268",
   "metadata": {},
   "outputs": [],
   "source": [
    "### DENORMALIZE input ###\n",
    "# \n",
    "def denorm(x):\n",
    "    out = (x + 1) / 2\n",
    "    return out.clamp(0, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e8321f3d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "2931b24c",
   "metadata": {},
   "source": [
    "### Import dependencies and dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a8f92a41",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "from torchvision.datasets import MNIST\n",
    "from torchvision.transforms import ToTensor, Normalize, Compose, Lambda\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c9f1b443",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_set = MNIST(root='./data', \n",
    "                  train=True, \n",
    "                  download=True,\n",
    "                   transform=Compose([ToTensor(), Normalize(mean=(0.5,), std=(0.5,))])\n",
    "                 )\n",
    "\n",
    "test_set = MNIST(root='./data', \n",
    "                  train=False, \n",
    "                  download=True,\n",
    "                  transform=Compose([ToTensor(), Normalize(mean=(0.5,), std=(0.5,))])\n",
    "                )\n",
    "\n",
    "##########################################################\n",
    "### We don't need to scale the data becase when we     ###\n",
    "### sample batch from DataLoader, it is already scaled ###\n",
    "### (But I'm keeping the code, it might comes handy)   ###\n",
    "##########################################################\n",
    "\n",
    "# print('Before Pre-processing\\n======Train=========')\n",
    "# print(f'Max: {torch.max(train_set.data)}, Min: {torch.min(train_set.data)}')\n",
    "# print('======Test=========')\n",
    "# print(f'Max: {torch.max(test_set.data)}, Min: {torch.min(test_set.data)}')\n",
    "\n",
    "# train_set.data = scale_norm(train_set.data)\n",
    "# test_set.data  = scale_norm(test_set.data)\n",
    "\n",
    "# print('\\nAfter Pre-processing\\n======Train=========')\n",
    "# print(f'Max: {torch.max(train_set.data)}, Min: {torch.min(train_set.data)}')\n",
    "# print('======Test=========')\n",
    "# print(f'Max: {torch.max(test_set.data)}, Min: {torch.min(test_set.data)}')\n",
    "\n",
    "# print(f'\\nTraining set: {train_set.data.shape}')\n",
    "# print(f'Training set: {test_set.data.shape}')\n",
    "\n",
    "batch_size = 64\n",
    "\n",
    "train_dl = DataLoader(train_set, batch_size=batch_size, shuffle=True)\n",
    "test_dl = DataLoader(test_set, batch_size=batch_size, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "af0e3c17",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "639ed963",
   "metadata": {},
   "source": [
    "### Plot Some Training Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ec2694d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 100 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(5, 5))\n",
    "rows, columns = 10, 10\n",
    "for i in range(1, columns*rows +1):\n",
    "    fig.add_subplot(rows, columns, i)\n",
    "    plt.imshow(train_set.data[i-1], cmap='gray')\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a8ed178d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using cuda device\n"
     ]
    }
   ],
   "source": [
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "print(f'Using {device} device')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6772b0c2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "b82ee3bd",
   "metadata": {},
   "source": [
    "## Discriminator Model (D)\n",
    "\n",
    "The discriminator takes an image as input, and tries to classify it as \"real\" or \"generated\". <br />\n",
    "We can use a CNN for the discriminator, but for simplicity, for now we'll a feedforward network with 3 linear layers to keep things since. <br />\n",
    "Every image is 28x28 matrix which can be flatten to a vector of size 784. <br />\n",
    "`Input:784 -> NN:784 x hidden -> LeakyRelu(0.2) -> NN:hidden x hidden -> LeakyRelu(0.2) -> NN:hidden x 1 -> Sigmoid (output)` <br /> <br />\n",
    "*'LeakyRelu' because we want the gradient to flow to the negative inputs as well*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7a3bf4ae",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sequential(\n",
      "  (0): Linear(in_features=784, out_features=256, bias=True)\n",
      "  (1): LeakyReLU(negative_slope=0.2)\n",
      "  (2): Linear(in_features=256, out_features=256, bias=True)\n",
      "  (3): LeakyReLU(negative_slope=0.2)\n",
      "  (4): Linear(in_features=256, out_features=1, bias=True)\n",
      "  (5): Sigmoid()\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "hidden_size = 256\n",
    "\n",
    "D = nn.Sequential(\n",
    "        nn.Linear(784, hidden_size),\n",
    "        nn.LeakyReLU(0.2),\n",
    "        nn.Linear(hidden_size, hidden_size),\n",
    "        nn.LeakyReLU(0.2),\n",
    "        nn.Linear(hidden_size, 1),\n",
    "        nn.Sigmoid()\n",
    "    )\n",
    "D.to(device)\n",
    "print(D)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc32cd69",
   "metadata": {},
   "source": [
    "## Generator Model (G)\n",
    "\n",
    "The input to the generator is typically a vector or a matrix which is used as a seed for generating an image. <br />\n",
    "Once again, to keep things simple, we'll use a feedfoward neural network with 3 layers, and the output will be a vector of size 784, which can be transformed to a 28x28 matrix image. <br />\n",
    "`Input:latent -> NN:latent x hidden -> Relu -> NN:hidden x hidden -> Relu -> NN: hidden x 784 -> TanH (output)` <br /> <br />\n",
    "*'TanH' is a function's output ranging from -1 to 1 (which is exactly our data range)*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2b516367",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sequential(\n",
      "  (0): Linear(in_features=256, out_features=256, bias=True)\n",
      "  (1): ReLU()\n",
      "  (2): Linear(in_features=256, out_features=256, bias=True)\n",
      "  (3): ReLU()\n",
      "  (4): Linear(in_features=256, out_features=784, bias=True)\n",
      "  (5): Tanh()\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "latent_size = 256\n",
    "\n",
    "G = nn.Sequential(\n",
    "        nn.Linear(latent_size, hidden_size),\n",
    "        nn.ReLU(),\n",
    "        nn.Linear(hidden_size, hidden_size),\n",
    "        nn.ReLU(),\n",
    "        nn.Linear(hidden_size, 784),\n",
    "        nn.Tanh()\n",
    "    )\n",
    "\n",
    "G.to(device)\n",
    "print(G)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68886852",
   "metadata": {},
   "source": [
    "### Test the Generator Model\n",
    "- Random 5 noises (Seeds) \n",
    "- Send to GPU `(because we sent G model to the gpu in the above cell)`\n",
    "- denorm() the generated image `from [-1, 1] -> [0, 1]`\n",
    "- torch view() is the same as reshape() in numpy\n",
    "- detach() split the image (input) from its gradient\n",
    "- Send back to CPU to plot `because matplotlib can't access data on the gpu`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9441e9ae",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([5, 784])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x360 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "noise = torch.rand(5, latent_size).to(device)\n",
    "gen_img = G(noise)\n",
    "print(gen_img.shape)\n",
    "gen_img = denorm(gen_img)\n",
    "gen_img = gen_img.view(-1, 28, 28)\n",
    "\n",
    "fig = plt.figure(figsize=(13, 5))\n",
    "rows, columns = 1, 5\n",
    "for i in range(1, columns*rows + 1):\n",
    "    fig.add_subplot(rows, columns, i)\n",
    "    plt.imshow(gen_img[i-1].detach().cpu(), cmap='gray')\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0935cb1c",
   "metadata": {},
   "source": [
    "## How The Model Will Be Trained\n",
    "* ### Loss/Objective Function (Binary Cross Entropy)\n",
    "\n",
    "#### For Discriminator\n",
    "*The loss of Discriminator is the sum of the loss of giving prediction on real images and loss of giving prediction on fake images*\n",
    "![BCE_real](./bce1.PNG)\n",
    "*For Real Images (y=1):* <br />\n",
    "`Here, y_hat is the prediction from the Discriminator Model (D)`\n",
    "> - We want **y_hat = 1** (**D** is doing its job well) because then Loss = log(1) = 0 (no punishment) <br />\n",
    "> - If **y_hat = 0.1** (**D** only has confidence of 10% of saying an image is real), then Loss = -log(0.1) = 2.3 (Punishment) <br />\n",
    "\n",
    "---\n",
    "![BCE_fake](./bce0.PNG)\n",
    "*For Fake Images (y=0):*\n",
    "> - We want **y_hat = 0** (**D** is confidence in predicting fake images)<br />\n",
    "> - If **y_hat = 0.99** then Loss = -log(0.01) = 4.6 (Punishment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d56e02c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "criterion = nn.BCELoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8d3248b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "d_optimizer = torch.optim.Adam(D.parameters(), lr=0.0003)\n",
    "g_optimizer = torch.optim.Adam(G.parameters(), lr=0.0003)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "252bee4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def reset_gradient():\n",
    "    d_optimizer.zero_grad()\n",
    "    g_optimizer.zero_grad()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "418c2a51",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_discriminator(real_imgs):\n",
    "    curr_batch_size = len(real_imgs)\n",
    "    \n",
    "    ##### Create labels (y) for real and fake batch #####\n",
    "    label_real = torch.ones(curr_batch_size).to(device)   # y = 1\n",
    "    label_fake = torch.zeros(curr_batch_size).to(device)  # y = 0\n",
    "    \n",
    "    reset_gradient()\n",
    "    \n",
    "    ##### Train on real images #####\n",
    "    real_imgs = real_imgs.view(curr_batch_size, -1).to(device)\n",
    "    pred_real_imgs = D(real_imgs)   # D's prediction on the real images\n",
    "    pred_real_imgs = pred_real_imgs.view(-1)\n",
    "    loss_real_imgs = criterion(pred_real_imgs, label_real) # Want y_hat = 1\n",
    "    \n",
    "    ##### Train on fake images #####\n",
    "    noise = torch.randn(curr_batch_size, latent_size).to(device)\n",
    "    fake_imgs = G(noise)   # Generate fake images\n",
    "    pred_fake_imgs = D(fake_imgs) # D's prediction on fake images\n",
    "    pred_fake_imgs = pred_fake_imgs.view(-1)\n",
    "    loss_fake_imgs = criterion(pred_fake_imgs, label_fake) # We want y_hat = 0\n",
    "    \n",
    "    d_loss = loss_fake_imgs + loss_real_imgs\n",
    "    d_loss.backward()     # Compute gradient w.r.t the Discriminator's parameters\n",
    "    d_optimizer.step()\n",
    "    return (d_loss, pred_real_imgs, pred_fake_imgs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "50b00fd2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_generator():\n",
    "    noise = torch.randn(batch_size, latent_size).to(device)\n",
    "    reset_gradient()\n",
    "    fake_imgs = G(noise)   # Generate fake images\n",
    "    pred = D(fake_imgs)           # Get some predictions from D \n",
    "    pred = pred.view(-1)\n",
    "    label = torch.ones(batch_size).to(device)\n",
    "    \n",
    "    ### The reason we use label (y = 1) is because we want\n",
    "    ### G to generate images to fool D to predict that the fake image is 1\n",
    "    g_loss = criterion(pred, label)\n",
    "    g_loss.backward()\n",
    "    g_optimizer.step()\n",
    "    return (g_loss, fake_imgs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fe99634b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "f8dbcb23",
   "metadata": {},
   "source": [
    "## Save Progess While Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "64d49eb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "sample_directory = './mnist_gans/samples'\n",
    "if not os.path.exists(sample_directory):\n",
    "    os.makedirs(sample_directory)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a6a04be5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "from torchvision.utils import save_image\n",
    "\n",
    "# Save a batch of real images (8 imgs each row)\n",
    "for imgs, _ in train_dl:\n",
    "    save_image(denorm(imgs), os.path.join(sample_directory, 'real_images.png'), nrow=8)\n",
    "    break\n",
    "\n",
    "# Display Image\n",
    "Image(os.path.join(sample_directory, 'real_images.png'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a9645e63",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0000.png\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "noise = torch.randn(batch_size, latent_size).to(device)\n",
    "\n",
    "def save_fake_images(index):\n",
    "    fake_imgs = G(noise)\n",
    "    fake_imgs = fake_imgs.reshape(len(fake_imgs), 1, 28, 28)\n",
    "    fake_img_name = 'fake_images-{0:0=4d}.png'.format(index)\n",
    "    print('Saving', fake_img_name)\n",
    "    save_image(denorm(fake_imgs), os.path.join(sample_directory, fake_img_name), nrow=8)\n",
    "    \n",
    "# # Before training\n",
    "save_fake_images(0)\n",
    "Image(os.path.join(sample_directory, 'fake_images-0000.png'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "43c6ce43",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "70993171",
   "metadata": {},
   "outputs": [],
   "source": [
    "# D returns (d_loss, pred_real_imgs, pred_fake_imgs)\n",
    "# G returns (g_loss, fake_imgs)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c4408ef",
   "metadata": {},
   "source": [
    "## Train GANs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8fd049f7",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [1/300], Step [200/938], d_loss: 0.0444, g_loss: 4.7488, D(x): 0.98, D(G(z)): 0.03\n",
      "Epoch [1/300], Step [400/938], d_loss: 0.0612, g_loss: 5.8051, D(x): 0.98, D(G(z)): 0.03\n",
      "Epoch [1/300], Step [600/938], d_loss: 0.2040, g_loss: 3.6361, D(x): 0.98, D(G(z)): 0.13\n",
      "Epoch [1/300], Step [800/938], d_loss: 0.2738, g_loss: 3.6840, D(x): 0.91, D(G(z)): 0.09\n",
      "Saving fake_images-0001.png\n",
      "Epoch [2/300], Step [200/938], d_loss: 0.8857, g_loss: 2.9657, D(x): 0.89, D(G(z)): 0.44\n",
      "Epoch [2/300], Step [400/938], d_loss: 0.7412, g_loss: 4.1553, D(x): 0.83, D(G(z)): 0.24\n",
      "Epoch [2/300], Step [600/938], d_loss: 1.5734, g_loss: 1.4295, D(x): 0.60, D(G(z)): 0.42\n",
      "Epoch [2/300], Step [800/938], d_loss: 0.5440, g_loss: 3.5656, D(x): 0.82, D(G(z)): 0.18\n",
      "Saving fake_images-0002.png\n",
      "Epoch [3/300], Step [200/938], d_loss: 0.7123, g_loss: 2.3861, D(x): 0.81, D(G(z)): 0.26\n",
      "Epoch [3/300], Step [400/938], d_loss: 0.9122, g_loss: 2.2467, D(x): 0.71, D(G(z)): 0.21\n",
      "Epoch [3/300], Step [600/938], d_loss: 1.1340, g_loss: 2.3626, D(x): 0.64, D(G(z)): 0.22\n",
      "Epoch [3/300], Step [800/938], d_loss: 0.8693, g_loss: 2.6743, D(x): 0.74, D(G(z)): 0.18\n",
      "Saving fake_images-0003.png\n",
      "Epoch [4/300], Step [200/938], d_loss: 0.2171, g_loss: 3.1388, D(x): 0.95, D(G(z)): 0.13\n",
      "Epoch [4/300], Step [400/938], d_loss: 0.4533, g_loss: 2.2474, D(x): 0.90, D(G(z)): 0.24\n",
      "Epoch [4/300], Step [600/938], d_loss: 0.3990, g_loss: 3.1388, D(x): 0.89, D(G(z)): 0.15\n",
      "Epoch [4/300], Step [800/938], d_loss: 0.5429, g_loss: 3.1612, D(x): 0.85, D(G(z)): 0.09\n",
      "Saving fake_images-0004.png\n",
      "Epoch [5/300], Step [200/938], d_loss: 0.4686, g_loss: 3.3300, D(x): 0.86, D(G(z)): 0.13\n",
      "Epoch [5/300], Step [400/938], d_loss: 0.2115, g_loss: 3.6066, D(x): 0.94, D(G(z)): 0.11\n",
      "Epoch [5/300], Step [600/938], d_loss: 0.1836, g_loss: 4.8003, D(x): 0.92, D(G(z)): 0.04\n",
      "Epoch [5/300], Step [800/938], d_loss: 0.4024, g_loss: 2.9093, D(x): 0.94, D(G(z)): 0.26\n",
      "Saving fake_images-0005.png\n",
      "Epoch [6/300], Step [200/938], d_loss: 0.3395, g_loss: 4.8135, D(x): 0.88, D(G(z)): 0.07\n",
      "Epoch [6/300], Step [400/938], d_loss: 0.1078, g_loss: 5.1512, D(x): 0.97, D(G(z)): 0.06\n",
      "Epoch [6/300], Step [600/938], d_loss: 0.2900, g_loss: 5.7890, D(x): 0.88, D(G(z)): 0.07\n",
      "Epoch [6/300], Step [800/938], d_loss: 0.3200, g_loss: 3.7968, D(x): 0.91, D(G(z)): 0.06\n",
      "Saving fake_images-0006.png\n",
      "Epoch [7/300], Step [200/938], d_loss: 0.6249, g_loss: 4.2644, D(x): 0.78, D(G(z)): 0.14\n",
      "Epoch [7/300], Step [400/938], d_loss: 0.3479, g_loss: 3.5135, D(x): 0.92, D(G(z)): 0.11\n",
      "Epoch [7/300], Step [600/938], d_loss: 0.1738, g_loss: 6.9504, D(x): 0.90, D(G(z)): 0.01\n",
      "Epoch [7/300], Step [800/938], d_loss: 0.5431, g_loss: 3.8726, D(x): 0.83, D(G(z)): 0.12\n",
      "Saving fake_images-0007.png\n",
      "Epoch [8/300], Step [200/938], d_loss: 0.2834, g_loss: 5.0600, D(x): 0.91, D(G(z)): 0.08\n",
      "Epoch [8/300], Step [400/938], d_loss: 0.6120, g_loss: 2.4512, D(x): 0.79, D(G(z)): 0.09\n",
      "Epoch [8/300], Step [600/938], d_loss: 0.1706, g_loss: 3.3898, D(x): 0.94, D(G(z)): 0.07\n",
      "Epoch [8/300], Step [800/938], d_loss: 0.6400, g_loss: 2.7689, D(x): 0.91, D(G(z)): 0.30\n",
      "Saving fake_images-0008.png\n",
      "Epoch [9/300], Step [200/938], d_loss: 0.4656, g_loss: 4.8179, D(x): 0.89, D(G(z)): 0.21\n",
      "Epoch [9/300], Step [400/938], d_loss: 0.3628, g_loss: 3.8143, D(x): 0.88, D(G(z)): 0.09\n",
      "Epoch [9/300], Step [600/938], d_loss: 0.4573, g_loss: 3.2672, D(x): 0.92, D(G(z)): 0.24\n",
      "Epoch [9/300], Step [800/938], d_loss: 0.7600, g_loss: 5.0673, D(x): 0.77, D(G(z)): 0.04\n",
      "Saving fake_images-0009.png\n",
      "Epoch [10/300], Step [200/938], d_loss: 0.2029, g_loss: 3.9180, D(x): 0.91, D(G(z)): 0.06\n",
      "Epoch [10/300], Step [400/938], d_loss: 0.2948, g_loss: 4.5768, D(x): 0.97, D(G(z)): 0.17\n",
      "Epoch [10/300], Step [600/938], d_loss: 0.7389, g_loss: 3.8423, D(x): 0.85, D(G(z)): 0.21\n",
      "Epoch [10/300], Step [800/938], d_loss: 0.8677, g_loss: 1.6955, D(x): 0.73, D(G(z)): 0.18\n",
      "Saving fake_images-0010.png\n",
      "Epoch [11/300], Step [200/938], d_loss: 0.3840, g_loss: 3.9113, D(x): 0.90, D(G(z)): 0.18\n",
      "Epoch [11/300], Step [400/938], d_loss: 0.3967, g_loss: 3.1963, D(x): 0.93, D(G(z)): 0.22\n",
      "Epoch [11/300], Step [600/938], d_loss: 0.2687, g_loss: 3.1694, D(x): 0.94, D(G(z)): 0.15\n",
      "Epoch [11/300], Step [800/938], d_loss: 0.3454, g_loss: 3.2897, D(x): 0.86, D(G(z)): 0.10\n",
      "Saving fake_images-0011.png\n",
      "\t*** SAVED MODEL FOR EPOCH 10 ***\n",
      "Epoch [12/300], Step [200/938], d_loss: 0.6536, g_loss: 3.4627, D(x): 0.82, D(G(z)): 0.20\n",
      "Epoch [12/300], Step [400/938], d_loss: 1.1203, g_loss: 2.3307, D(x): 0.68, D(G(z)): 0.14\n",
      "Epoch [12/300], Step [600/938], d_loss: 0.3723, g_loss: 3.2023, D(x): 0.90, D(G(z)): 0.15\n",
      "Epoch [12/300], Step [800/938], d_loss: 0.5806, g_loss: 3.2864, D(x): 0.81, D(G(z)): 0.12\n",
      "Saving fake_images-0012.png\n",
      "Epoch [13/300], Step [200/938], d_loss: 0.2469, g_loss: 3.3306, D(x): 0.90, D(G(z)): 0.07\n",
      "Epoch [13/300], Step [400/938], d_loss: 0.5088, g_loss: 3.6225, D(x): 0.85, D(G(z)): 0.09\n",
      "Epoch [13/300], Step [600/938], d_loss: 0.3149, g_loss: 4.2865, D(x): 0.89, D(G(z)): 0.07\n",
      "Epoch [13/300], Step [800/938], d_loss: 0.2737, g_loss: 4.1851, D(x): 0.89, D(G(z)): 0.06\n",
      "Saving fake_images-0013.png\n",
      "Epoch [14/300], Step [200/938], d_loss: 0.2847, g_loss: 4.0843, D(x): 0.95, D(G(z)): 0.13\n",
      "Epoch [14/300], Step [400/938], d_loss: 0.2085, g_loss: 4.4550, D(x): 0.96, D(G(z)): 0.11\n",
      "Epoch [14/300], Step [600/938], d_loss: 0.5779, g_loss: 2.4932, D(x): 0.97, D(G(z)): 0.31\n",
      "Epoch [14/300], Step [800/938], d_loss: 0.3373, g_loss: 3.6579, D(x): 0.92, D(G(z)): 0.14\n",
      "Saving fake_images-0014.png\n",
      "Epoch [15/300], Step [200/938], d_loss: 0.2635, g_loss: 2.8466, D(x): 0.95, D(G(z)): 0.11\n",
      "Epoch [15/300], Step [400/938], d_loss: 0.3432, g_loss: 3.7510, D(x): 0.90, D(G(z)): 0.12\n",
      "Epoch [15/300], Step [600/938], d_loss: 0.3715, g_loss: 3.7601, D(x): 0.92, D(G(z)): 0.15\n",
      "Epoch [15/300], Step [800/938], d_loss: 0.2748, g_loss: 3.6511, D(x): 0.90, D(G(z)): 0.09\n",
      "Saving fake_images-0015.png\n",
      "Epoch [16/300], Step [200/938], d_loss: 0.5202, g_loss: 3.4561, D(x): 0.85, D(G(z)): 0.13\n",
      "Epoch [16/300], Step [400/938], d_loss: 0.3525, g_loss: 3.1251, D(x): 0.91, D(G(z)): 0.16\n",
      "Epoch [16/300], Step [600/938], d_loss: 0.4829, g_loss: 2.4811, D(x): 0.86, D(G(z)): 0.12\n",
      "Epoch [16/300], Step [800/938], d_loss: 0.3396, g_loss: 3.2312, D(x): 0.94, D(G(z)): 0.19\n",
      "Saving fake_images-0016.png\n",
      "Epoch [17/300], Step [200/938], d_loss: 0.4158, g_loss: 2.7490, D(x): 0.91, D(G(z)): 0.18\n",
      "Epoch [17/300], Step [400/938], d_loss: 0.4357, g_loss: 2.8739, D(x): 0.89, D(G(z)): 0.19\n",
      "Epoch [17/300], Step [600/938], d_loss: 0.5948, g_loss: 1.8871, D(x): 0.82, D(G(z)): 0.17\n",
      "Epoch [17/300], Step [800/938], d_loss: 0.7878, g_loss: 3.1238, D(x): 0.76, D(G(z)): 0.10\n",
      "Saving fake_images-0017.png\n",
      "Epoch [18/300], Step [200/938], d_loss: 0.4702, g_loss: 4.0058, D(x): 0.84, D(G(z)): 0.14\n",
      "Epoch [18/300], Step [400/938], d_loss: 0.2722, g_loss: 3.4964, D(x): 0.90, D(G(z)): 0.09\n",
      "Epoch [18/300], Step [600/938], d_loss: 0.5070, g_loss: 2.0010, D(x): 0.80, D(G(z)): 0.13\n",
      "Epoch [18/300], Step [800/938], d_loss: 0.5827, g_loss: 3.6347, D(x): 0.80, D(G(z)): 0.14\n",
      "Saving fake_images-0018.png\n",
      "Epoch [19/300], Step [200/938], d_loss: 0.4368, g_loss: 4.6495, D(x): 0.80, D(G(z)): 0.07\n",
      "Epoch [19/300], Step [400/938], d_loss: 0.4896, g_loss: 2.3500, D(x): 0.89, D(G(z)): 0.21\n",
      "Epoch [19/300], Step [600/938], d_loss: 0.3502, g_loss: 4.5639, D(x): 0.93, D(G(z)): 0.15\n",
      "Epoch [19/300], Step [800/938], d_loss: 0.4356, g_loss: 4.2327, D(x): 0.82, D(G(z)): 0.08\n",
      "Saving fake_images-0019.png\n",
      "Epoch [20/300], Step [200/938], d_loss: 0.5747, g_loss: 2.1778, D(x): 0.83, D(G(z)): 0.18\n",
      "Epoch [20/300], Step [400/938], d_loss: 0.7955, g_loss: 2.6454, D(x): 0.71, D(G(z)): 0.09\n",
      "Epoch [20/300], Step [600/938], d_loss: 0.6346, g_loss: 3.6113, D(x): 0.78, D(G(z)): 0.16\n",
      "Epoch [20/300], Step [800/938], d_loss: 0.5494, g_loss: 2.6950, D(x): 0.84, D(G(z)): 0.17\n",
      "Saving fake_images-0020.png\n",
      "Epoch [21/300], Step [200/938], d_loss: 0.3274, g_loss: 2.5988, D(x): 0.96, D(G(z)): 0.20\n",
      "Epoch [21/300], Step [400/938], d_loss: 0.8150, g_loss: 2.7774, D(x): 0.74, D(G(z)): 0.08\n",
      "Epoch [21/300], Step [600/938], d_loss: 0.4769, g_loss: 2.7306, D(x): 0.84, D(G(z)): 0.16\n",
      "Epoch [21/300], Step [800/938], d_loss: 0.4647, g_loss: 3.2618, D(x): 0.89, D(G(z)): 0.18\n",
      "Saving fake_images-0021.png\n",
      "Epoch [22/300], Step [200/938], d_loss: 0.4571, g_loss: 2.7383, D(x): 0.92, D(G(z)): 0.20\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [22/300], Step [400/938], d_loss: 0.5780, g_loss: 3.8159, D(x): 0.81, D(G(z)): 0.12\n",
      "Epoch [22/300], Step [600/938], d_loss: 0.6113, g_loss: 2.2835, D(x): 0.84, D(G(z)): 0.23\n",
      "Epoch [22/300], Step [800/938], d_loss: 0.4202, g_loss: 3.5791, D(x): 0.90, D(G(z)): 0.18\n",
      "Saving fake_images-0022.png\n",
      "Epoch [23/300], Step [200/938], d_loss: 0.4261, g_loss: 3.8974, D(x): 0.84, D(G(z)): 0.11\n",
      "Epoch [23/300], Step [400/938], d_loss: 0.4911, g_loss: 3.7804, D(x): 0.83, D(G(z)): 0.13\n",
      "Epoch [23/300], Step [600/938], d_loss: 0.4727, g_loss: 3.3977, D(x): 0.85, D(G(z)): 0.16\n",
      "Epoch [23/300], Step [800/938], d_loss: 0.5444, g_loss: 3.6154, D(x): 0.87, D(G(z)): 0.19\n",
      "Saving fake_images-0023.png\n",
      "Epoch [24/300], Step [200/938], d_loss: 0.5936, g_loss: 2.3122, D(x): 0.90, D(G(z)): 0.24\n",
      "Epoch [24/300], Step [400/938], d_loss: 0.3682, g_loss: 4.5823, D(x): 0.84, D(G(z)): 0.06\n",
      "Epoch [24/300], Step [600/938], d_loss: 0.3018, g_loss: 3.5664, D(x): 0.86, D(G(z)): 0.08\n",
      "Epoch [24/300], Step [800/938], d_loss: 0.6361, g_loss: 2.7453, D(x): 0.81, D(G(z)): 0.18\n",
      "Saving fake_images-0024.png\n",
      "Epoch [25/300], Step [200/938], d_loss: 0.8405, g_loss: 2.8532, D(x): 0.70, D(G(z)): 0.13\n",
      "Epoch [25/300], Step [400/938], d_loss: 0.5961, g_loss: 2.5108, D(x): 0.77, D(G(z)): 0.13\n",
      "Epoch [25/300], Step [600/938], d_loss: 0.4281, g_loss: 3.2939, D(x): 0.83, D(G(z)): 0.12\n",
      "Epoch [25/300], Step [800/938], d_loss: 0.5610, g_loss: 1.7070, D(x): 0.85, D(G(z)): 0.25\n",
      "Saving fake_images-0025.png\n",
      "Epoch [26/300], Step [200/938], d_loss: 0.5443, g_loss: 2.1664, D(x): 0.91, D(G(z)): 0.25\n",
      "Epoch [26/300], Step [400/938], d_loss: 0.7622, g_loss: 2.5044, D(x): 0.68, D(G(z)): 0.11\n",
      "Epoch [26/300], Step [600/938], d_loss: 0.7145, g_loss: 2.2127, D(x): 0.75, D(G(z)): 0.16\n",
      "Epoch [26/300], Step [800/938], d_loss: 0.5429, g_loss: 2.6022, D(x): 0.81, D(G(z)): 0.17\n",
      "Saving fake_images-0026.png\n",
      "Epoch [27/300], Step [200/938], d_loss: 0.6927, g_loss: 2.9767, D(x): 0.74, D(G(z)): 0.14\n",
      "Epoch [27/300], Step [400/938], d_loss: 0.6377, g_loss: 2.1213, D(x): 0.87, D(G(z)): 0.29\n",
      "Epoch [27/300], Step [600/938], d_loss: 0.8475, g_loss: 2.1934, D(x): 0.69, D(G(z)): 0.20\n",
      "Epoch [27/300], Step [800/938], d_loss: 0.7224, g_loss: 3.0540, D(x): 0.77, D(G(z)): 0.23\n",
      "Saving fake_images-0027.png\n",
      "Epoch [28/300], Step [200/938], d_loss: 0.5860, g_loss: 2.2199, D(x): 0.84, D(G(z)): 0.22\n",
      "Epoch [28/300], Step [400/938], d_loss: 0.7027, g_loss: 2.8309, D(x): 0.76, D(G(z)): 0.11\n",
      "Epoch [28/300], Step [600/938], d_loss: 0.5319, g_loss: 2.8762, D(x): 0.87, D(G(z)): 0.21\n",
      "Epoch [28/300], Step [800/938], d_loss: 0.2513, g_loss: 3.0367, D(x): 0.91, D(G(z)): 0.10\n",
      "Saving fake_images-0028.png\n",
      "Epoch [29/300], Step [200/938], d_loss: 0.6343, g_loss: 3.0839, D(x): 0.73, D(G(z)): 0.07\n",
      "Epoch [29/300], Step [400/938], d_loss: 0.8648, g_loss: 2.7928, D(x): 0.79, D(G(z)): 0.32\n",
      "Epoch [29/300], Step [600/938], d_loss: 0.7464, g_loss: 3.0427, D(x): 0.82, D(G(z)): 0.26\n",
      "Epoch [29/300], Step [800/938], d_loss: 0.3819, g_loss: 2.2505, D(x): 0.90, D(G(z)): 0.19\n",
      "Saving fake_images-0029.png\n",
      "Epoch [30/300], Step [200/938], d_loss: 0.8108, g_loss: 2.7275, D(x): 0.79, D(G(z)): 0.26\n",
      "Epoch [30/300], Step [400/938], d_loss: 0.7104, g_loss: 2.6195, D(x): 0.79, D(G(z)): 0.25\n",
      "Epoch [30/300], Step [600/938], d_loss: 0.7049, g_loss: 2.6577, D(x): 0.76, D(G(z)): 0.20\n",
      "Epoch [30/300], Step [800/938], d_loss: 0.5597, g_loss: 2.9631, D(x): 0.83, D(G(z)): 0.21\n",
      "Saving fake_images-0030.png\n",
      "Epoch [31/300], Step [200/938], d_loss: 0.7940, g_loss: 2.5947, D(x): 0.74, D(G(z)): 0.13\n",
      "Epoch [31/300], Step [400/938], d_loss: 0.6123, g_loss: 2.3214, D(x): 0.79, D(G(z)): 0.20\n",
      "Epoch [31/300], Step [600/938], d_loss: 0.4656, g_loss: 2.3206, D(x): 0.88, D(G(z)): 0.22\n",
      "Epoch [31/300], Step [800/938], d_loss: 0.7421, g_loss: 2.0791, D(x): 0.75, D(G(z)): 0.18\n",
      "Saving fake_images-0031.png\n",
      "Epoch [32/300], Step [200/938], d_loss: 0.8055, g_loss: 2.2517, D(x): 0.76, D(G(z)): 0.28\n",
      "Epoch [32/300], Step [400/938], d_loss: 0.5560, g_loss: 2.6462, D(x): 0.78, D(G(z)): 0.13\n",
      "Epoch [32/300], Step [600/938], d_loss: 0.6958, g_loss: 2.4736, D(x): 0.76, D(G(z)): 0.16\n",
      "Epoch [32/300], Step [800/938], d_loss: 0.8430, g_loss: 2.0465, D(x): 0.82, D(G(z)): 0.30\n",
      "Saving fake_images-0032.png\n",
      "Epoch [33/300], Step [200/938], d_loss: 1.0134, g_loss: 2.1041, D(x): 0.69, D(G(z)): 0.28\n",
      "Epoch [33/300], Step [400/938], d_loss: 0.8138, g_loss: 2.1046, D(x): 0.72, D(G(z)): 0.19\n",
      "Epoch [33/300], Step [600/938], d_loss: 0.5068, g_loss: 2.7228, D(x): 0.84, D(G(z)): 0.18\n",
      "Epoch [33/300], Step [800/938], d_loss: 0.8612, g_loss: 2.4817, D(x): 0.65, D(G(z)): 0.14\n",
      "Saving fake_images-0033.png\n",
      "Epoch [34/300], Step [200/938], d_loss: 0.6198, g_loss: 2.1124, D(x): 0.83, D(G(z)): 0.24\n",
      "Epoch [34/300], Step [400/938], d_loss: 0.7031, g_loss: 1.8785, D(x): 0.90, D(G(z)): 0.36\n",
      "Epoch [34/300], Step [600/938], d_loss: 0.6748, g_loss: 1.8810, D(x): 0.75, D(G(z)): 0.18\n",
      "Epoch [34/300], Step [800/938], d_loss: 0.8540, g_loss: 2.0153, D(x): 0.74, D(G(z)): 0.23\n",
      "Saving fake_images-0034.png\n",
      "Epoch [35/300], Step [200/938], d_loss: 0.8286, g_loss: 2.1801, D(x): 0.71, D(G(z)): 0.21\n",
      "Epoch [35/300], Step [400/938], d_loss: 0.7199, g_loss: 1.4761, D(x): 0.79, D(G(z)): 0.25\n",
      "Epoch [35/300], Step [600/938], d_loss: 0.7305, g_loss: 2.0807, D(x): 0.77, D(G(z)): 0.27\n",
      "Epoch [35/300], Step [800/938], d_loss: 0.5731, g_loss: 2.6400, D(x): 0.75, D(G(z)): 0.06\n",
      "Saving fake_images-0035.png\n",
      "Epoch [36/300], Step [200/938], d_loss: 0.7715, g_loss: 1.8690, D(x): 0.75, D(G(z)): 0.24\n",
      "Epoch [36/300], Step [400/938], d_loss: 0.7486, g_loss: 2.2174, D(x): 0.79, D(G(z)): 0.21\n",
      "Epoch [36/300], Step [600/938], d_loss: 0.8239, g_loss: 1.9961, D(x): 0.74, D(G(z)): 0.26\n",
      "Epoch [36/300], Step [800/938], d_loss: 0.9133, g_loss: 1.9130, D(x): 0.67, D(G(z)): 0.10\n",
      "Saving fake_images-0036.png\n",
      "Epoch [37/300], Step [200/938], d_loss: 0.8041, g_loss: 1.7564, D(x): 0.77, D(G(z)): 0.27\n",
      "Epoch [37/300], Step [400/938], d_loss: 0.6522, g_loss: 2.1302, D(x): 0.86, D(G(z)): 0.29\n",
      "Epoch [37/300], Step [600/938], d_loss: 0.6240, g_loss: 2.2076, D(x): 0.80, D(G(z)): 0.22\n",
      "Epoch [37/300], Step [800/938], d_loss: 0.8948, g_loss: 2.1623, D(x): 0.63, D(G(z)): 0.17\n",
      "Saving fake_images-0037.png\n",
      "Epoch [38/300], Step [200/938], d_loss: 0.6362, g_loss: 1.6846, D(x): 0.74, D(G(z)): 0.15\n",
      "Epoch [38/300], Step [400/938], d_loss: 0.8802, g_loss: 1.9275, D(x): 0.77, D(G(z)): 0.32\n",
      "Epoch [38/300], Step [600/938], d_loss: 0.7118, g_loss: 2.6708, D(x): 0.75, D(G(z)): 0.22\n",
      "Epoch [38/300], Step [800/938], d_loss: 0.7498, g_loss: 2.1374, D(x): 0.72, D(G(z)): 0.18\n",
      "Saving fake_images-0038.png\n",
      "Epoch [39/300], Step [200/938], d_loss: 0.8724, g_loss: 1.7322, D(x): 0.80, D(G(z)): 0.34\n",
      "Epoch [39/300], Step [400/938], d_loss: 0.5901, g_loss: 2.1068, D(x): 0.79, D(G(z)): 0.22\n",
      "Epoch [39/300], Step [600/938], d_loss: 0.6039, g_loss: 1.6866, D(x): 0.87, D(G(z)): 0.29\n",
      "Epoch [39/300], Step [800/938], d_loss: 0.6726, g_loss: 1.5712, D(x): 0.75, D(G(z)): 0.20\n",
      "Saving fake_images-0039.png\n",
      "Epoch [40/300], Step [200/938], d_loss: 0.5847, g_loss: 2.1789, D(x): 0.77, D(G(z)): 0.17\n",
      "Epoch [40/300], Step [400/938], d_loss: 0.8965, g_loss: 2.4669, D(x): 0.71, D(G(z)): 0.22\n",
      "Epoch [40/300], Step [600/938], d_loss: 0.8538, g_loss: 2.8661, D(x): 0.72, D(G(z)): 0.21\n",
      "Epoch [40/300], Step [800/938], d_loss: 0.6978, g_loss: 2.3113, D(x): 0.72, D(G(z)): 0.18\n",
      "Saving fake_images-0040.png\n",
      "Epoch [41/300], Step [200/938], d_loss: 0.7559, g_loss: 2.1085, D(x): 0.77, D(G(z)): 0.28\n",
      "Epoch [41/300], Step [400/938], d_loss: 0.8752, g_loss: 2.2246, D(x): 0.68, D(G(z)): 0.24\n",
      "Epoch [41/300], Step [600/938], d_loss: 0.8400, g_loss: 2.3342, D(x): 0.82, D(G(z)): 0.34\n",
      "Epoch [41/300], Step [800/938], d_loss: 1.0210, g_loss: 2.4656, D(x): 0.59, D(G(z)): 0.13\n",
      "Saving fake_images-0041.png\n",
      "Epoch [42/300], Step [200/938], d_loss: 0.7984, g_loss: 2.2967, D(x): 0.71, D(G(z)): 0.20\n",
      "Epoch [42/300], Step [400/938], d_loss: 0.4692, g_loss: 2.6222, D(x): 0.84, D(G(z)): 0.20\n",
      "Epoch [42/300], Step [600/938], d_loss: 0.5670, g_loss: 2.1127, D(x): 0.91, D(G(z)): 0.27\n",
      "Epoch [42/300], Step [800/938], d_loss: 0.6838, g_loss: 1.7148, D(x): 0.86, D(G(z)): 0.30\n",
      "Saving fake_images-0042.png\n",
      "Epoch [43/300], Step [200/938], d_loss: 0.7545, g_loss: 2.2819, D(x): 0.71, D(G(z)): 0.17\n",
      "Epoch [43/300], Step [400/938], d_loss: 0.7677, g_loss: 2.5671, D(x): 0.73, D(G(z)): 0.17\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [43/300], Step [600/938], d_loss: 0.6952, g_loss: 2.0246, D(x): 0.75, D(G(z)): 0.21\n",
      "Epoch [43/300], Step [800/938], d_loss: 0.7689, g_loss: 1.9951, D(x): 0.74, D(G(z)): 0.26\n",
      "Saving fake_images-0043.png\n",
      "Epoch [44/300], Step [200/938], d_loss: 1.0196, g_loss: 2.0506, D(x): 0.65, D(G(z)): 0.26\n",
      "Epoch [44/300], Step [400/938], d_loss: 0.6738, g_loss: 2.4033, D(x): 0.80, D(G(z)): 0.26\n",
      "Epoch [44/300], Step [600/938], d_loss: 0.7770, g_loss: 2.1862, D(x): 0.75, D(G(z)): 0.22\n",
      "Epoch [44/300], Step [800/938], d_loss: 0.9163, g_loss: 1.9183, D(x): 0.72, D(G(z)): 0.30\n",
      "Saving fake_images-0044.png\n",
      "Epoch [45/300], Step [200/938], d_loss: 1.0915, g_loss: 1.4329, D(x): 0.71, D(G(z)): 0.34\n",
      "Epoch [45/300], Step [400/938], d_loss: 0.6428, g_loss: 2.6248, D(x): 0.76, D(G(z)): 0.21\n",
      "Epoch [45/300], Step [600/938], d_loss: 0.7026, g_loss: 2.4090, D(x): 0.71, D(G(z)): 0.19\n",
      "Epoch [45/300], Step [800/938], d_loss: 0.9480, g_loss: 1.5432, D(x): 0.66, D(G(z)): 0.25\n",
      "Saving fake_images-0045.png\n",
      "Epoch [46/300], Step [200/938], d_loss: 0.9828, g_loss: 1.5821, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [46/300], Step [400/938], d_loss: 0.9325, g_loss: 1.7315, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [46/300], Step [600/938], d_loss: 1.0666, g_loss: 2.3747, D(x): 0.67, D(G(z)): 0.28\n",
      "Epoch [46/300], Step [800/938], d_loss: 1.0108, g_loss: 1.8356, D(x): 0.70, D(G(z)): 0.32\n",
      "Saving fake_images-0046.png\n",
      "Epoch [47/300], Step [200/938], d_loss: 0.6907, g_loss: 2.3428, D(x): 0.76, D(G(z)): 0.23\n",
      "Epoch [47/300], Step [400/938], d_loss: 0.7661, g_loss: 1.6360, D(x): 0.81, D(G(z)): 0.29\n",
      "Epoch [47/300], Step [600/938], d_loss: 0.8106, g_loss: 1.6375, D(x): 0.86, D(G(z)): 0.40\n",
      "Epoch [47/300], Step [800/938], d_loss: 0.8328, g_loss: 2.0722, D(x): 0.79, D(G(z)): 0.33\n",
      "Saving fake_images-0047.png\n",
      "Epoch [48/300], Step [200/938], d_loss: 0.7797, g_loss: 1.9486, D(x): 0.74, D(G(z)): 0.24\n",
      "Epoch [48/300], Step [400/938], d_loss: 0.9968, g_loss: 1.6496, D(x): 0.65, D(G(z)): 0.27\n",
      "Epoch [48/300], Step [600/938], d_loss: 1.0068, g_loss: 1.2140, D(x): 0.77, D(G(z)): 0.37\n",
      "Epoch [48/300], Step [800/938], d_loss: 0.6934, g_loss: 1.7274, D(x): 0.80, D(G(z)): 0.29\n",
      "Saving fake_images-0048.png\n",
      "Epoch [49/300], Step [200/938], d_loss: 0.7535, g_loss: 2.6307, D(x): 0.78, D(G(z)): 0.25\n",
      "Epoch [49/300], Step [400/938], d_loss: 0.8457, g_loss: 2.6014, D(x): 0.83, D(G(z)): 0.38\n",
      "Epoch [49/300], Step [600/938], d_loss: 0.9061, g_loss: 1.9944, D(x): 0.67, D(G(z)): 0.24\n",
      "Epoch [49/300], Step [800/938], d_loss: 0.5441, g_loss: 2.4635, D(x): 0.79, D(G(z)): 0.16\n",
      "Saving fake_images-0049.png\n",
      "Epoch [50/300], Step [200/938], d_loss: 0.8934, g_loss: 1.9818, D(x): 0.74, D(G(z)): 0.27\n",
      "Epoch [50/300], Step [400/938], d_loss: 1.0887, g_loss: 1.9764, D(x): 0.62, D(G(z)): 0.26\n",
      "Epoch [50/300], Step [600/938], d_loss: 0.9561, g_loss: 1.5552, D(x): 0.73, D(G(z)): 0.33\n",
      "Epoch [50/300], Step [800/938], d_loss: 0.7786, g_loss: 2.2073, D(x): 0.69, D(G(z)): 0.16\n",
      "Saving fake_images-0050.png\n",
      "Epoch [51/300], Step [200/938], d_loss: 0.7230, g_loss: 1.6984, D(x): 0.76, D(G(z)): 0.26\n",
      "Epoch [51/300], Step [400/938], d_loss: 0.6918, g_loss: 1.8850, D(x): 0.73, D(G(z)): 0.18\n",
      "Epoch [51/300], Step [600/938], d_loss: 0.8384, g_loss: 1.7273, D(x): 0.68, D(G(z)): 0.22\n",
      "Epoch [51/300], Step [800/938], d_loss: 0.7081, g_loss: 1.9761, D(x): 0.79, D(G(z)): 0.27\n",
      "Saving fake_images-0051.png\n",
      "\t*** SAVED MODEL FOR EPOCH 50 ***\n",
      "Epoch [52/300], Step [200/938], d_loss: 0.8834, g_loss: 2.3997, D(x): 0.68, D(G(z)): 0.19\n",
      "Epoch [52/300], Step [400/938], d_loss: 0.8134, g_loss: 1.3995, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [52/300], Step [600/938], d_loss: 0.8185, g_loss: 1.9760, D(x): 0.70, D(G(z)): 0.22\n",
      "Epoch [52/300], Step [800/938], d_loss: 0.6625, g_loss: 1.5265, D(x): 0.77, D(G(z)): 0.23\n",
      "Saving fake_images-0052.png\n",
      "Epoch [53/300], Step [200/938], d_loss: 0.8307, g_loss: 1.5586, D(x): 0.72, D(G(z)): 0.26\n",
      "Epoch [53/300], Step [400/938], d_loss: 0.6118, g_loss: 1.5167, D(x): 0.75, D(G(z)): 0.21\n",
      "Epoch [53/300], Step [600/938], d_loss: 0.9734, g_loss: 2.2639, D(x): 0.63, D(G(z)): 0.18\n",
      "Epoch [53/300], Step [800/938], d_loss: 0.8415, g_loss: 1.6531, D(x): 0.75, D(G(z)): 0.26\n",
      "Saving fake_images-0053.png\n",
      "Epoch [54/300], Step [200/938], d_loss: 1.0043, g_loss: 1.9473, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [54/300], Step [400/938], d_loss: 0.8254, g_loss: 1.5837, D(x): 0.79, D(G(z)): 0.32\n",
      "Epoch [54/300], Step [600/938], d_loss: 0.7480, g_loss: 1.3330, D(x): 0.76, D(G(z)): 0.26\n",
      "Epoch [54/300], Step [800/938], d_loss: 0.7891, g_loss: 2.2269, D(x): 0.71, D(G(z)): 0.23\n",
      "Saving fake_images-0054.png\n",
      "Epoch [55/300], Step [200/938], d_loss: 0.9395, g_loss: 2.1960, D(x): 0.68, D(G(z)): 0.27\n",
      "Epoch [55/300], Step [400/938], d_loss: 0.7918, g_loss: 1.9989, D(x): 0.71, D(G(z)): 0.21\n",
      "Epoch [55/300], Step [600/938], d_loss: 0.7985, g_loss: 2.3309, D(x): 0.73, D(G(z)): 0.23\n",
      "Epoch [55/300], Step [800/938], d_loss: 0.9039, g_loss: 1.6896, D(x): 0.74, D(G(z)): 0.35\n",
      "Saving fake_images-0055.png\n",
      "Epoch [56/300], Step [200/938], d_loss: 0.7702, g_loss: 1.6196, D(x): 0.79, D(G(z)): 0.29\n",
      "Epoch [56/300], Step [400/938], d_loss: 0.9337, g_loss: 2.2337, D(x): 0.65, D(G(z)): 0.25\n",
      "Epoch [56/300], Step [600/938], d_loss: 0.6912, g_loss: 1.5553, D(x): 0.82, D(G(z)): 0.29\n",
      "Epoch [56/300], Step [800/938], d_loss: 0.7629, g_loss: 2.2107, D(x): 0.74, D(G(z)): 0.28\n",
      "Saving fake_images-0056.png\n",
      "Epoch [57/300], Step [200/938], d_loss: 0.7062, g_loss: 1.9826, D(x): 0.79, D(G(z)): 0.25\n",
      "Epoch [57/300], Step [400/938], d_loss: 0.8270, g_loss: 1.9092, D(x): 0.71, D(G(z)): 0.25\n",
      "Epoch [57/300], Step [600/938], d_loss: 0.9749, g_loss: 2.2330, D(x): 0.63, D(G(z)): 0.20\n",
      "Epoch [57/300], Step [800/938], d_loss: 1.0253, g_loss: 1.4620, D(x): 0.58, D(G(z)): 0.21\n",
      "Saving fake_images-0057.png\n",
      "Epoch [58/300], Step [200/938], d_loss: 0.7349, g_loss: 2.0907, D(x): 0.72, D(G(z)): 0.23\n",
      "Epoch [58/300], Step [400/938], d_loss: 0.8808, g_loss: 2.0528, D(x): 0.66, D(G(z)): 0.23\n",
      "Epoch [58/300], Step [600/938], d_loss: 0.6971, g_loss: 1.7137, D(x): 0.80, D(G(z)): 0.27\n",
      "Epoch [58/300], Step [800/938], d_loss: 0.8742, g_loss: 1.7489, D(x): 0.71, D(G(z)): 0.30\n",
      "Saving fake_images-0058.png\n",
      "Epoch [59/300], Step [200/938], d_loss: 0.8993, g_loss: 1.7318, D(x): 0.67, D(G(z)): 0.26\n",
      "Epoch [59/300], Step [400/938], d_loss: 0.6964, g_loss: 2.1267, D(x): 0.71, D(G(z)): 0.18\n",
      "Epoch [59/300], Step [600/938], d_loss: 0.7772, g_loss: 2.0401, D(x): 0.66, D(G(z)): 0.19\n",
      "Epoch [59/300], Step [800/938], d_loss: 0.9280, g_loss: 1.5272, D(x): 0.70, D(G(z)): 0.31\n",
      "Saving fake_images-0059.png\n",
      "Epoch [60/300], Step [200/938], d_loss: 0.9815, g_loss: 2.0748, D(x): 0.73, D(G(z)): 0.35\n",
      "Epoch [60/300], Step [400/938], d_loss: 0.8970, g_loss: 1.6268, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [60/300], Step [600/938], d_loss: 0.7966, g_loss: 2.2944, D(x): 0.71, D(G(z)): 0.22\n",
      "Epoch [60/300], Step [800/938], d_loss: 0.9651, g_loss: 2.6236, D(x): 0.68, D(G(z)): 0.21\n",
      "Saving fake_images-0060.png\n",
      "Epoch [61/300], Step [200/938], d_loss: 0.6722, g_loss: 1.5152, D(x): 0.77, D(G(z)): 0.27\n",
      "Epoch [61/300], Step [400/938], d_loss: 1.0683, g_loss: 1.2823, D(x): 0.79, D(G(z)): 0.42\n",
      "Epoch [61/300], Step [600/938], d_loss: 0.6940, g_loss: 1.7710, D(x): 0.76, D(G(z)): 0.23\n",
      "Epoch [61/300], Step [800/938], d_loss: 0.6972, g_loss: 1.8969, D(x): 0.83, D(G(z)): 0.29\n",
      "Saving fake_images-0061.png\n",
      "Epoch [62/300], Step [200/938], d_loss: 0.7430, g_loss: 1.9620, D(x): 0.68, D(G(z)): 0.19\n",
      "Epoch [62/300], Step [400/938], d_loss: 0.8142, g_loss: 1.6873, D(x): 0.71, D(G(z)): 0.27\n",
      "Epoch [62/300], Step [600/938], d_loss: 0.9390, g_loss: 1.6626, D(x): 0.71, D(G(z)): 0.29\n",
      "Epoch [62/300], Step [800/938], d_loss: 0.8267, g_loss: 2.2846, D(x): 0.65, D(G(z)): 0.17\n",
      "Saving fake_images-0062.png\n",
      "Epoch [63/300], Step [200/938], d_loss: 0.7995, g_loss: 2.6711, D(x): 0.67, D(G(z)): 0.17\n",
      "Epoch [63/300], Step [400/938], d_loss: 1.0039, g_loss: 1.2971, D(x): 0.73, D(G(z)): 0.37\n",
      "Epoch [63/300], Step [600/938], d_loss: 0.7872, g_loss: 1.7961, D(x): 0.65, D(G(z)): 0.19\n",
      "Epoch [63/300], Step [800/938], d_loss: 1.1048, g_loss: 0.9665, D(x): 0.88, D(G(z)): 0.52\n",
      "Saving fake_images-0063.png\n",
      "Epoch [64/300], Step [200/938], d_loss: 0.9589, g_loss: 1.3826, D(x): 0.73, D(G(z)): 0.38\n",
      "Epoch [64/300], Step [400/938], d_loss: 0.6921, g_loss: 2.1608, D(x): 0.73, D(G(z)): 0.22\n",
      "Epoch [64/300], Step [600/938], d_loss: 0.8036, g_loss: 1.7742, D(x): 0.77, D(G(z)): 0.29\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch [64/300], Step [800/938], d_loss: 0.8079, g_loss: 1.3747, D(x): 0.74, D(G(z)): 0.30\n",
      "Saving fake_images-0064.png\n",
      "Epoch [65/300], Step [200/938], d_loss: 0.8173, g_loss: 2.1980, D(x): 0.75, D(G(z)): 0.29\n",
      "Epoch [65/300], Step [400/938], d_loss: 0.8323, g_loss: 1.9246, D(x): 0.68, D(G(z)): 0.23\n",
      "Epoch [65/300], Step [600/938], d_loss: 0.7948, g_loss: 2.3050, D(x): 0.77, D(G(z)): 0.31\n",
      "Epoch [65/300], Step [800/938], d_loss: 0.9243, g_loss: 2.0798, D(x): 0.75, D(G(z)): 0.35\n",
      "Saving fake_images-0065.png\n",
      "Epoch [66/300], Step [200/938], d_loss: 0.8410, g_loss: 1.3303, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [66/300], Step [400/938], d_loss: 0.9556, g_loss: 1.5951, D(x): 0.71, D(G(z)): 0.33\n",
      "Epoch [66/300], Step [600/938], d_loss: 0.9740, g_loss: 1.7662, D(x): 0.70, D(G(z)): 0.32\n",
      "Epoch [66/300], Step [800/938], d_loss: 0.9233, g_loss: 1.5183, D(x): 0.71, D(G(z)): 0.29\n",
      "Saving fake_images-0066.png\n",
      "Epoch [67/300], Step [200/938], d_loss: 0.7575, g_loss: 2.7177, D(x): 0.71, D(G(z)): 0.20\n",
      "Epoch [67/300], Step [400/938], d_loss: 0.9053, g_loss: 1.5529, D(x): 0.76, D(G(z)): 0.37\n",
      "Epoch [67/300], Step [600/938], d_loss: 0.9763, g_loss: 1.9948, D(x): 0.64, D(G(z)): 0.22\n",
      "Epoch [67/300], Step [800/938], d_loss: 0.7412, g_loss: 2.4270, D(x): 0.71, D(G(z)): 0.20\n",
      "Saving fake_images-0067.png\n",
      "Epoch [68/300], Step [200/938], d_loss: 0.7314, g_loss: 1.8631, D(x): 0.72, D(G(z)): 0.23\n",
      "Epoch [68/300], Step [400/938], d_loss: 0.8478, g_loss: 1.4896, D(x): 0.72, D(G(z)): 0.28\n",
      "Epoch [68/300], Step [600/938], d_loss: 0.9289, g_loss: 1.7033, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [68/300], Step [800/938], d_loss: 0.8208, g_loss: 1.2339, D(x): 0.70, D(G(z)): 0.27\n",
      "Saving fake_images-0068.png\n",
      "Epoch [69/300], Step [200/938], d_loss: 0.9833, g_loss: 1.5460, D(x): 0.66, D(G(z)): 0.29\n",
      "Epoch [69/300], Step [400/938], d_loss: 0.7623, g_loss: 1.2715, D(x): 0.72, D(G(z)): 0.24\n",
      "Epoch [69/300], Step [600/938], d_loss: 0.9307, g_loss: 1.9269, D(x): 0.63, D(G(z)): 0.22\n",
      "Epoch [69/300], Step [800/938], d_loss: 0.6210, g_loss: 1.4337, D(x): 0.80, D(G(z)): 0.26\n",
      "Saving fake_images-0069.png\n",
      "Epoch [70/300], Step [200/938], d_loss: 0.9570, g_loss: 1.2344, D(x): 0.80, D(G(z)): 0.38\n",
      "Epoch [70/300], Step [400/938], d_loss: 1.1133, g_loss: 1.7831, D(x): 0.58, D(G(z)): 0.24\n",
      "Epoch [70/300], Step [600/938], d_loss: 0.6877, g_loss: 1.6310, D(x): 0.77, D(G(z)): 0.26\n",
      "Epoch [70/300], Step [800/938], d_loss: 0.8423, g_loss: 1.8547, D(x): 0.72, D(G(z)): 0.27\n",
      "Saving fake_images-0070.png\n",
      "Epoch [71/300], Step [200/938], d_loss: 0.9360, g_loss: 1.2276, D(x): 0.70, D(G(z)): 0.32\n",
      "Epoch [71/300], Step [400/938], d_loss: 0.9730, g_loss: 1.8533, D(x): 0.71, D(G(z)): 0.32\n",
      "Epoch [71/300], Step [600/938], d_loss: 1.1201, g_loss: 1.0420, D(x): 0.71, D(G(z)): 0.40\n",
      "Epoch [71/300], Step [800/938], d_loss: 0.8879, g_loss: 1.9461, D(x): 0.64, D(G(z)): 0.22\n",
      "Saving fake_images-0071.png\n",
      "Epoch [72/300], Step [200/938], d_loss: 0.9111, g_loss: 1.2823, D(x): 0.69, D(G(z)): 0.27\n",
      "Epoch [72/300], Step [400/938], d_loss: 0.7463, g_loss: 1.8379, D(x): 0.72, D(G(z)): 0.23\n",
      "Epoch [72/300], Step [600/938], d_loss: 0.6606, g_loss: 2.0181, D(x): 0.79, D(G(z)): 0.27\n",
      "Epoch [72/300], Step [800/938], d_loss: 0.9649, g_loss: 1.3992, D(x): 0.68, D(G(z)): 0.30\n",
      "Saving fake_images-0072.png\n",
      "Epoch [73/300], Step [200/938], d_loss: 0.8195, g_loss: 1.5766, D(x): 0.73, D(G(z)): 0.27\n",
      "Epoch [73/300], Step [400/938], d_loss: 0.8519, g_loss: 2.4342, D(x): 0.64, D(G(z)): 0.18\n",
      "Epoch [73/300], Step [600/938], d_loss: 0.9532, g_loss: 1.6367, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [73/300], Step [800/938], d_loss: 0.8132, g_loss: 1.3005, D(x): 0.70, D(G(z)): 0.26\n",
      "Saving fake_images-0073.png\n",
      "Epoch [74/300], Step [200/938], d_loss: 0.8343, g_loss: 1.1841, D(x): 0.81, D(G(z)): 0.36\n",
      "Epoch [74/300], Step [400/938], d_loss: 0.7314, g_loss: 1.6616, D(x): 0.68, D(G(z)): 0.17\n",
      "Epoch [74/300], Step [600/938], d_loss: 1.0481, g_loss: 1.5999, D(x): 0.71, D(G(z)): 0.39\n",
      "Epoch [74/300], Step [800/938], d_loss: 1.1474, g_loss: 1.6949, D(x): 0.63, D(G(z)): 0.33\n",
      "Saving fake_images-0074.png\n",
      "Epoch [75/300], Step [200/938], d_loss: 0.8773, g_loss: 1.5520, D(x): 0.76, D(G(z)): 0.34\n",
      "Epoch [75/300], Step [400/938], d_loss: 0.8301, g_loss: 1.7650, D(x): 0.66, D(G(z)): 0.23\n",
      "Epoch [75/300], Step [600/938], d_loss: 0.8545, g_loss: 2.1358, D(x): 0.63, D(G(z)): 0.19\n",
      "Epoch [75/300], Step [800/938], d_loss: 0.9890, g_loss: 1.4270, D(x): 0.70, D(G(z)): 0.34\n",
      "Saving fake_images-0075.png\n",
      "Epoch [76/300], Step [200/938], d_loss: 0.6835, g_loss: 1.9285, D(x): 0.72, D(G(z)): 0.23\n",
      "Epoch [76/300], Step [400/938], d_loss: 0.8360, g_loss: 1.5356, D(x): 0.76, D(G(z)): 0.33\n",
      "Epoch [76/300], Step [600/938], d_loss: 0.6851, g_loss: 1.7796, D(x): 0.79, D(G(z)): 0.29\n",
      "Epoch [76/300], Step [800/938], d_loss: 0.7139, g_loss: 1.9733, D(x): 0.76, D(G(z)): 0.25\n",
      "Saving fake_images-0076.png\n",
      "Epoch [77/300], Step [200/938], d_loss: 0.8759, g_loss: 1.7325, D(x): 0.77, D(G(z)): 0.37\n",
      "Epoch [77/300], Step [400/938], d_loss: 1.1922, g_loss: 1.5484, D(x): 0.73, D(G(z)): 0.44\n",
      "Epoch [77/300], Step [600/938], d_loss: 0.8010, g_loss: 1.8860, D(x): 0.75, D(G(z)): 0.29\n",
      "Epoch [77/300], Step [800/938], d_loss: 0.9773, g_loss: 2.0876, D(x): 0.62, D(G(z)): 0.23\n",
      "Saving fake_images-0077.png\n",
      "Epoch [78/300], Step [200/938], d_loss: 0.8772, g_loss: 1.5504, D(x): 0.68, D(G(z)): 0.25\n",
      "Epoch [78/300], Step [400/938], d_loss: 0.8763, g_loss: 1.8075, D(x): 0.71, D(G(z)): 0.28\n",
      "Epoch [78/300], Step [600/938], d_loss: 0.6987, g_loss: 1.6959, D(x): 0.76, D(G(z)): 0.24\n",
      "Epoch [78/300], Step [800/938], d_loss: 1.0277, g_loss: 2.2236, D(x): 0.55, D(G(z)): 0.12\n",
      "Saving fake_images-0078.png\n",
      "Epoch [79/300], Step [200/938], d_loss: 0.9281, g_loss: 1.8472, D(x): 0.69, D(G(z)): 0.27\n",
      "Epoch [79/300], Step [400/938], d_loss: 0.9189, g_loss: 2.0569, D(x): 0.70, D(G(z)): 0.27\n",
      "Epoch [79/300], Step [600/938], d_loss: 0.8157, g_loss: 1.8584, D(x): 0.72, D(G(z)): 0.29\n",
      "Epoch [79/300], Step [800/938], d_loss: 0.8980, g_loss: 1.6190, D(x): 0.64, D(G(z)): 0.23\n",
      "Saving fake_images-0079.png\n",
      "Epoch [80/300], Step [200/938], d_loss: 1.2504, g_loss: 1.3339, D(x): 0.51, D(G(z)): 0.25\n",
      "Epoch [80/300], Step [400/938], d_loss: 0.9761, g_loss: 1.9054, D(x): 0.62, D(G(z)): 0.23\n",
      "Epoch [80/300], Step [600/938], d_loss: 0.9595, g_loss: 1.6285, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [80/300], Step [800/938], d_loss: 0.9859, g_loss: 1.5318, D(x): 0.72, D(G(z)): 0.36\n",
      "Saving fake_images-0080.png\n",
      "Epoch [81/300], Step [200/938], d_loss: 1.0605, g_loss: 1.2224, D(x): 0.71, D(G(z)): 0.39\n",
      "Epoch [81/300], Step [400/938], d_loss: 0.9986, g_loss: 1.5917, D(x): 0.65, D(G(z)): 0.31\n",
      "Epoch [81/300], Step [600/938], d_loss: 0.9438, g_loss: 1.3471, D(x): 0.67, D(G(z)): 0.30\n",
      "Epoch [81/300], Step [800/938], d_loss: 0.9979, g_loss: 1.4201, D(x): 0.66, D(G(z)): 0.28\n",
      "Saving fake_images-0081.png\n",
      "Epoch [82/300], Step [200/938], d_loss: 0.8832, g_loss: 1.5612, D(x): 0.73, D(G(z)): 0.32\n",
      "Epoch [82/300], Step [400/938], d_loss: 0.9314, g_loss: 1.4769, D(x): 0.68, D(G(z)): 0.28\n",
      "Epoch [82/300], Step [600/938], d_loss: 0.9531, g_loss: 1.5504, D(x): 0.64, D(G(z)): 0.26\n",
      "Epoch [82/300], Step [800/938], d_loss: 1.0150, g_loss: 1.5315, D(x): 0.76, D(G(z)): 0.42\n",
      "Saving fake_images-0082.png\n",
      "Epoch [83/300], Step [200/938], d_loss: 0.7960, g_loss: 1.7739, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [83/300], Step [400/938], d_loss: 0.8138, g_loss: 1.8043, D(x): 0.73, D(G(z)): 0.30\n",
      "Epoch [83/300], Step [600/938], d_loss: 0.8886, g_loss: 2.1264, D(x): 0.64, D(G(z)): 0.20\n",
      "Epoch [83/300], Step [800/938], d_loss: 1.0231, g_loss: 1.4571, D(x): 0.68, D(G(z)): 0.31\n",
      "Saving fake_images-0083.png\n",
      "Epoch [84/300], Step [200/938], d_loss: 0.9889, g_loss: 1.3248, D(x): 0.72, D(G(z)): 0.37\n",
      "Epoch [84/300], Step [400/938], d_loss: 1.0050, g_loss: 1.1803, D(x): 0.70, D(G(z)): 0.37\n",
      "Epoch [84/300], Step [600/938], d_loss: 0.8018, g_loss: 2.2050, D(x): 0.73, D(G(z)): 0.27\n",
      "Epoch [84/300], Step [800/938], d_loss: 0.8568, g_loss: 1.6031, D(x): 0.66, D(G(z)): 0.26\n",
      "Saving fake_images-0084.png\n",
      "Epoch [85/300], Step [200/938], d_loss: 0.8519, g_loss: 1.6913, D(x): 0.74, D(G(z)): 0.31\n",
      "Epoch [85/300], Step [400/938], d_loss: 0.9595, g_loss: 1.4365, D(x): 0.74, D(G(z)): 0.37\n",
      "Epoch [85/300], Step [600/938], d_loss: 0.9845, g_loss: 2.1294, D(x): 0.62, D(G(z)): 0.23\n",
      "Epoch [85/300], Step [800/938], d_loss: 1.0068, g_loss: 1.3875, D(x): 0.71, D(G(z)): 0.37\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0085.png\n",
      "Epoch [86/300], Step [200/938], d_loss: 0.8122, g_loss: 1.7091, D(x): 0.73, D(G(z)): 0.25\n",
      "Epoch [86/300], Step [400/938], d_loss: 0.8741, g_loss: 1.2393, D(x): 0.73, D(G(z)): 0.33\n",
      "Epoch [86/300], Step [600/938], d_loss: 0.7514, g_loss: 1.5040, D(x): 0.74, D(G(z)): 0.25\n",
      "Epoch [86/300], Step [800/938], d_loss: 0.7397, g_loss: 1.4982, D(x): 0.76, D(G(z)): 0.28\n",
      "Saving fake_images-0086.png\n",
      "Epoch [87/300], Step [200/938], d_loss: 1.0871, g_loss: 1.3999, D(x): 0.72, D(G(z)): 0.42\n",
      "Epoch [87/300], Step [400/938], d_loss: 0.8160, g_loss: 1.9741, D(x): 0.66, D(G(z)): 0.23\n",
      "Epoch [87/300], Step [600/938], d_loss: 0.9819, g_loss: 1.6223, D(x): 0.69, D(G(z)): 0.26\n",
      "Epoch [87/300], Step [800/938], d_loss: 1.1761, g_loss: 1.2126, D(x): 0.70, D(G(z)): 0.40\n",
      "Saving fake_images-0087.png\n",
      "Epoch [88/300], Step [200/938], d_loss: 1.0773, g_loss: 1.8032, D(x): 0.70, D(G(z)): 0.36\n",
      "Epoch [88/300], Step [400/938], d_loss: 0.9467, g_loss: 1.6457, D(x): 0.63, D(G(z)): 0.22\n",
      "Epoch [88/300], Step [600/938], d_loss: 0.8450, g_loss: 1.5088, D(x): 0.74, D(G(z)): 0.31\n",
      "Epoch [88/300], Step [800/938], d_loss: 0.9856, g_loss: 1.9260, D(x): 0.71, D(G(z)): 0.36\n",
      "Saving fake_images-0088.png\n",
      "Epoch [89/300], Step [200/938], d_loss: 1.1229, g_loss: 1.5586, D(x): 0.63, D(G(z)): 0.34\n",
      "Epoch [89/300], Step [400/938], d_loss: 1.2214, g_loss: 1.9331, D(x): 0.54, D(G(z)): 0.22\n",
      "Epoch [89/300], Step [600/938], d_loss: 0.8922, g_loss: 1.9252, D(x): 0.71, D(G(z)): 0.29\n",
      "Epoch [89/300], Step [800/938], d_loss: 0.8817, g_loss: 1.7074, D(x): 0.73, D(G(z)): 0.32\n",
      "Saving fake_images-0089.png\n",
      "Epoch [90/300], Step [200/938], d_loss: 0.9539, g_loss: 1.7335, D(x): 0.62, D(G(z)): 0.27\n",
      "Epoch [90/300], Step [400/938], d_loss: 0.9026, g_loss: 1.9171, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [90/300], Step [600/938], d_loss: 0.9435, g_loss: 2.0996, D(x): 0.65, D(G(z)): 0.25\n",
      "Epoch [90/300], Step [800/938], d_loss: 0.9465, g_loss: 1.2937, D(x): 0.62, D(G(z)): 0.21\n",
      "Saving fake_images-0090.png\n",
      "Epoch [91/300], Step [200/938], d_loss: 0.8249, g_loss: 1.6537, D(x): 0.72, D(G(z)): 0.29\n",
      "Epoch [91/300], Step [400/938], d_loss: 0.9874, g_loss: 1.8874, D(x): 0.60, D(G(z)): 0.26\n",
      "Epoch [91/300], Step [600/938], d_loss: 0.6862, g_loss: 2.1831, D(x): 0.74, D(G(z)): 0.23\n",
      "Epoch [91/300], Step [800/938], d_loss: 1.0385, g_loss: 1.8708, D(x): 0.68, D(G(z)): 0.30\n",
      "Saving fake_images-0091.png\n",
      "Epoch [92/300], Step [200/938], d_loss: 0.7289, g_loss: 1.2828, D(x): 0.75, D(G(z)): 0.28\n",
      "Epoch [92/300], Step [400/938], d_loss: 0.8355, g_loss: 1.5017, D(x): 0.68, D(G(z)): 0.24\n",
      "Epoch [92/300], Step [600/938], d_loss: 0.8798, g_loss: 1.8914, D(x): 0.74, D(G(z)): 0.29\n",
      "Epoch [92/300], Step [800/938], d_loss: 0.9282, g_loss: 1.4909, D(x): 0.67, D(G(z)): 0.29\n",
      "Saving fake_images-0092.png\n",
      "Epoch [93/300], Step [200/938], d_loss: 0.8700, g_loss: 1.3178, D(x): 0.70, D(G(z)): 0.30\n",
      "Epoch [93/300], Step [400/938], d_loss: 0.9397, g_loss: 1.5067, D(x): 0.73, D(G(z)): 0.33\n",
      "Epoch [93/300], Step [600/938], d_loss: 0.8745, g_loss: 1.4968, D(x): 0.71, D(G(z)): 0.30\n",
      "Epoch [93/300], Step [800/938], d_loss: 0.8770, g_loss: 1.2360, D(x): 0.77, D(G(z)): 0.38\n",
      "Saving fake_images-0093.png\n",
      "Epoch [94/300], Step [200/938], d_loss: 0.9294, g_loss: 2.1218, D(x): 0.70, D(G(z)): 0.30\n",
      "Epoch [94/300], Step [400/938], d_loss: 1.0163, g_loss: 1.2977, D(x): 0.68, D(G(z)): 0.33\n",
      "Epoch [94/300], Step [600/938], d_loss: 0.8485, g_loss: 1.9819, D(x): 0.70, D(G(z)): 0.26\n",
      "Epoch [94/300], Step [800/938], d_loss: 0.9843, g_loss: 1.3763, D(x): 0.72, D(G(z)): 0.38\n",
      "Saving fake_images-0094.png\n",
      "Epoch [95/300], Step [200/938], d_loss: 0.7935, g_loss: 1.6287, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [95/300], Step [400/938], d_loss: 1.1597, g_loss: 1.4931, D(x): 0.58, D(G(z)): 0.30\n",
      "Epoch [95/300], Step [600/938], d_loss: 0.8000, g_loss: 1.3901, D(x): 0.73, D(G(z)): 0.28\n",
      "Epoch [95/300], Step [800/938], d_loss: 0.9620, g_loss: 1.4111, D(x): 0.74, D(G(z)): 0.35\n",
      "Saving fake_images-0095.png\n",
      "Epoch [96/300], Step [200/938], d_loss: 0.8033, g_loss: 1.7935, D(x): 0.71, D(G(z)): 0.26\n",
      "Epoch [96/300], Step [400/938], d_loss: 1.1406, g_loss: 1.3475, D(x): 0.60, D(G(z)): 0.28\n",
      "Epoch [96/300], Step [600/938], d_loss: 0.8113, g_loss: 1.6037, D(x): 0.70, D(G(z)): 0.27\n",
      "Epoch [96/300], Step [800/938], d_loss: 0.8892, g_loss: 2.1553, D(x): 0.64, D(G(z)): 0.20\n",
      "Saving fake_images-0096.png\n",
      "Epoch [97/300], Step [200/938], d_loss: 0.6469, g_loss: 1.5950, D(x): 0.72, D(G(z)): 0.16\n",
      "Epoch [97/300], Step [400/938], d_loss: 1.2432, g_loss: 1.4827, D(x): 0.60, D(G(z)): 0.30\n",
      "Epoch [97/300], Step [600/938], d_loss: 0.9121, g_loss: 1.6605, D(x): 0.68, D(G(z)): 0.27\n",
      "Epoch [97/300], Step [800/938], d_loss: 0.7910, g_loss: 1.8244, D(x): 0.86, D(G(z)): 0.39\n",
      "Saving fake_images-0097.png\n",
      "Epoch [98/300], Step [200/938], d_loss: 1.1342, g_loss: 1.9162, D(x): 0.70, D(G(z)): 0.40\n",
      "Epoch [98/300], Step [400/938], d_loss: 0.9726, g_loss: 1.2933, D(x): 0.71, D(G(z)): 0.34\n",
      "Epoch [98/300], Step [600/938], d_loss: 1.4154, g_loss: 1.6111, D(x): 0.47, D(G(z)): 0.24\n",
      "Epoch [98/300], Step [800/938], d_loss: 0.9879, g_loss: 1.5302, D(x): 0.72, D(G(z)): 0.34\n",
      "Saving fake_images-0098.png\n",
      "Epoch [99/300], Step [200/938], d_loss: 0.8786, g_loss: 1.3981, D(x): 0.79, D(G(z)): 0.36\n",
      "Epoch [99/300], Step [400/938], d_loss: 0.8771, g_loss: 1.8258, D(x): 0.66, D(G(z)): 0.24\n",
      "Epoch [99/300], Step [600/938], d_loss: 1.1357, g_loss: 1.1384, D(x): 0.62, D(G(z)): 0.31\n",
      "Epoch [99/300], Step [800/938], d_loss: 0.9257, g_loss: 1.0168, D(x): 0.69, D(G(z)): 0.32\n",
      "Saving fake_images-0099.png\n",
      "Epoch [100/300], Step [200/938], d_loss: 0.9141, g_loss: 1.2196, D(x): 0.74, D(G(z)): 0.34\n",
      "Epoch [100/300], Step [400/938], d_loss: 0.9899, g_loss: 1.8547, D(x): 0.58, D(G(z)): 0.19\n",
      "Epoch [100/300], Step [600/938], d_loss: 0.8147, g_loss: 1.4746, D(x): 0.69, D(G(z)): 0.26\n",
      "Epoch [100/300], Step [800/938], d_loss: 0.8958, g_loss: 1.1923, D(x): 0.72, D(G(z)): 0.31\n",
      "Saving fake_images-0100.png\n",
      "Epoch [101/300], Step [200/938], d_loss: 0.8877, g_loss: 1.4662, D(x): 0.63, D(G(z)): 0.24\n",
      "Epoch [101/300], Step [400/938], d_loss: 0.9175, g_loss: 1.4785, D(x): 0.76, D(G(z)): 0.35\n",
      "Epoch [101/300], Step [600/938], d_loss: 1.0524, g_loss: 1.9555, D(x): 0.59, D(G(z)): 0.21\n",
      "Epoch [101/300], Step [800/938], d_loss: 1.2139, g_loss: 1.1945, D(x): 0.70, D(G(z)): 0.39\n",
      "Saving fake_images-0101.png\n",
      "\t*** SAVED MODEL FOR EPOCH 100 ***\n",
      "Epoch [102/300], Step [200/938], d_loss: 0.9437, g_loss: 1.2513, D(x): 0.61, D(G(z)): 0.24\n",
      "Epoch [102/300], Step [400/938], d_loss: 0.7133, g_loss: 2.1468, D(x): 0.71, D(G(z)): 0.21\n",
      "Epoch [102/300], Step [600/938], d_loss: 1.0252, g_loss: 1.2666, D(x): 0.67, D(G(z)): 0.34\n",
      "Epoch [102/300], Step [800/938], d_loss: 0.8277, g_loss: 1.6965, D(x): 0.68, D(G(z)): 0.25\n",
      "Saving fake_images-0102.png\n",
      "Epoch [103/300], Step [200/938], d_loss: 1.0231, g_loss: 1.5439, D(x): 0.67, D(G(z)): 0.31\n",
      "Epoch [103/300], Step [400/938], d_loss: 0.8781, g_loss: 1.3895, D(x): 0.66, D(G(z)): 0.22\n",
      "Epoch [103/300], Step [600/938], d_loss: 0.9406, g_loss: 1.4751, D(x): 0.71, D(G(z)): 0.34\n",
      "Epoch [103/300], Step [800/938], d_loss: 0.8866, g_loss: 2.1184, D(x): 0.67, D(G(z)): 0.28\n",
      "Saving fake_images-0103.png\n",
      "Epoch [104/300], Step [200/938], d_loss: 0.7855, g_loss: 1.5654, D(x): 0.72, D(G(z)): 0.28\n",
      "Epoch [104/300], Step [400/938], d_loss: 1.1090, g_loss: 1.6555, D(x): 0.70, D(G(z)): 0.36\n",
      "Epoch [104/300], Step [600/938], d_loss: 0.8988, g_loss: 1.8170, D(x): 0.63, D(G(z)): 0.24\n",
      "Epoch [104/300], Step [800/938], d_loss: 1.0726, g_loss: 1.4168, D(x): 0.66, D(G(z)): 0.34\n",
      "Saving fake_images-0104.png\n",
      "Epoch [105/300], Step [200/938], d_loss: 1.0212, g_loss: 1.7734, D(x): 0.64, D(G(z)): 0.29\n",
      "Epoch [105/300], Step [400/938], d_loss: 0.9916, g_loss: 1.7243, D(x): 0.65, D(G(z)): 0.30\n",
      "Epoch [105/300], Step [600/938], d_loss: 0.7765, g_loss: 1.7480, D(x): 0.78, D(G(z)): 0.34\n",
      "Epoch [105/300], Step [800/938], d_loss: 0.7436, g_loss: 1.7047, D(x): 0.80, D(G(z)): 0.31\n",
      "Saving fake_images-0105.png\n",
      "Epoch [106/300], Step [200/938], d_loss: 0.9520, g_loss: 1.5341, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [106/300], Step [400/938], d_loss: 0.9065, g_loss: 1.3876, D(x): 0.74, D(G(z)): 0.35\n",
      "Epoch [106/300], Step [600/938], d_loss: 0.9049, g_loss: 1.5573, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [106/300], Step [800/938], d_loss: 0.9913, g_loss: 1.2305, D(x): 0.73, D(G(z)): 0.39\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0106.png\n",
      "Epoch [107/300], Step [200/938], d_loss: 0.8327, g_loss: 1.4625, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [107/300], Step [400/938], d_loss: 0.9632, g_loss: 2.3820, D(x): 0.64, D(G(z)): 0.21\n",
      "Epoch [107/300], Step [600/938], d_loss: 0.9030, g_loss: 1.5405, D(x): 0.63, D(G(z)): 0.24\n",
      "Epoch [107/300], Step [800/938], d_loss: 0.9025, g_loss: 1.3753, D(x): 0.71, D(G(z)): 0.33\n",
      "Saving fake_images-0107.png\n",
      "Epoch [108/300], Step [200/938], d_loss: 0.8509, g_loss: 1.4386, D(x): 0.66, D(G(z)): 0.26\n",
      "Epoch [108/300], Step [400/938], d_loss: 0.8349, g_loss: 1.2334, D(x): 0.77, D(G(z)): 0.34\n",
      "Epoch [108/300], Step [600/938], d_loss: 0.9854, g_loss: 1.4537, D(x): 0.71, D(G(z)): 0.34\n",
      "Epoch [108/300], Step [800/938], d_loss: 0.8418, g_loss: 1.9966, D(x): 0.72, D(G(z)): 0.30\n",
      "Saving fake_images-0108.png\n",
      "Epoch [109/300], Step [200/938], d_loss: 0.9238, g_loss: 1.3103, D(x): 0.71, D(G(z)): 0.34\n",
      "Epoch [109/300], Step [400/938], d_loss: 0.8935, g_loss: 1.8771, D(x): 0.68, D(G(z)): 0.31\n",
      "Epoch [109/300], Step [600/938], d_loss: 1.0153, g_loss: 2.0904, D(x): 0.61, D(G(z)): 0.20\n",
      "Epoch [109/300], Step [800/938], d_loss: 0.9083, g_loss: 1.4788, D(x): 0.69, D(G(z)): 0.30\n",
      "Saving fake_images-0109.png\n",
      "Epoch [110/300], Step [200/938], d_loss: 1.0354, g_loss: 1.3816, D(x): 0.66, D(G(z)): 0.37\n",
      "Epoch [110/300], Step [400/938], d_loss: 1.1073, g_loss: 1.2926, D(x): 0.67, D(G(z)): 0.37\n",
      "Epoch [110/300], Step [600/938], d_loss: 0.7197, g_loss: 1.6337, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [110/300], Step [800/938], d_loss: 1.0288, g_loss: 1.6838, D(x): 0.62, D(G(z)): 0.26\n",
      "Saving fake_images-0110.png\n",
      "Epoch [111/300], Step [200/938], d_loss: 0.8312, g_loss: 1.1381, D(x): 0.72, D(G(z)): 0.29\n",
      "Epoch [111/300], Step [400/938], d_loss: 0.9267, g_loss: 1.6317, D(x): 0.67, D(G(z)): 0.27\n",
      "Epoch [111/300], Step [600/938], d_loss: 0.9011, g_loss: 1.4957, D(x): 0.63, D(G(z)): 0.23\n",
      "Epoch [111/300], Step [800/938], d_loss: 0.9161, g_loss: 1.3968, D(x): 0.72, D(G(z)): 0.33\n",
      "Saving fake_images-0111.png\n",
      "Epoch [112/300], Step [200/938], d_loss: 0.7428, g_loss: 2.2738, D(x): 0.75, D(G(z)): 0.25\n",
      "Epoch [112/300], Step [400/938], d_loss: 0.8024, g_loss: 1.5572, D(x): 0.69, D(G(z)): 0.25\n",
      "Epoch [112/300], Step [600/938], d_loss: 0.9940, g_loss: 1.6435, D(x): 0.65, D(G(z)): 0.29\n",
      "Epoch [112/300], Step [800/938], d_loss: 0.9497, g_loss: 1.2927, D(x): 0.81, D(G(z)): 0.43\n",
      "Saving fake_images-0112.png\n",
      "Epoch [113/300], Step [200/938], d_loss: 0.9957, g_loss: 1.2264, D(x): 0.81, D(G(z)): 0.41\n",
      "Epoch [113/300], Step [400/938], d_loss: 0.8491, g_loss: 1.9583, D(x): 0.73, D(G(z)): 0.29\n",
      "Epoch [113/300], Step [600/938], d_loss: 0.8375, g_loss: 1.6479, D(x): 0.73, D(G(z)): 0.29\n",
      "Epoch [113/300], Step [800/938], d_loss: 0.7169, g_loss: 1.9146, D(x): 0.77, D(G(z)): 0.24\n",
      "Saving fake_images-0113.png\n",
      "Epoch [114/300], Step [200/938], d_loss: 0.8632, g_loss: 1.4208, D(x): 0.72, D(G(z)): 0.30\n",
      "Epoch [114/300], Step [400/938], d_loss: 0.9178, g_loss: 1.6098, D(x): 0.66, D(G(z)): 0.27\n",
      "Epoch [114/300], Step [600/938], d_loss: 1.0707, g_loss: 1.2868, D(x): 0.65, D(G(z)): 0.29\n",
      "Epoch [114/300], Step [800/938], d_loss: 1.0676, g_loss: 1.9073, D(x): 0.61, D(G(z)): 0.29\n",
      "Saving fake_images-0114.png\n",
      "Epoch [115/300], Step [200/938], d_loss: 0.7457, g_loss: 1.1116, D(x): 0.74, D(G(z)): 0.30\n",
      "Epoch [115/300], Step [400/938], d_loss: 1.0475, g_loss: 1.4719, D(x): 0.69, D(G(z)): 0.37\n",
      "Epoch [115/300], Step [600/938], d_loss: 0.9015, g_loss: 1.5258, D(x): 0.71, D(G(z)): 0.32\n",
      "Epoch [115/300], Step [800/938], d_loss: 1.0761, g_loss: 1.5363, D(x): 0.63, D(G(z)): 0.31\n",
      "Saving fake_images-0115.png\n",
      "Epoch [116/300], Step [200/938], d_loss: 0.9656, g_loss: 1.7001, D(x): 0.66, D(G(z)): 0.29\n",
      "Epoch [116/300], Step [400/938], d_loss: 1.2076, g_loss: 2.2504, D(x): 0.53, D(G(z)): 0.17\n",
      "Epoch [116/300], Step [600/938], d_loss: 0.8429, g_loss: 1.7992, D(x): 0.66, D(G(z)): 0.23\n",
      "Epoch [116/300], Step [800/938], d_loss: 0.8686, g_loss: 1.4492, D(x): 0.76, D(G(z)): 0.33\n",
      "Saving fake_images-0116.png\n",
      "Epoch [117/300], Step [200/938], d_loss: 1.0360, g_loss: 1.3300, D(x): 0.68, D(G(z)): 0.35\n",
      "Epoch [117/300], Step [400/938], d_loss: 0.7108, g_loss: 1.4693, D(x): 0.77, D(G(z)): 0.29\n",
      "Epoch [117/300], Step [600/938], d_loss: 1.0177, g_loss: 1.8739, D(x): 0.60, D(G(z)): 0.20\n",
      "Epoch [117/300], Step [800/938], d_loss: 0.9887, g_loss: 1.3386, D(x): 0.63, D(G(z)): 0.29\n",
      "Saving fake_images-0117.png\n",
      "Epoch [118/300], Step [200/938], d_loss: 0.8523, g_loss: 1.2745, D(x): 0.72, D(G(z)): 0.33\n",
      "Epoch [118/300], Step [400/938], d_loss: 0.9693, g_loss: 2.3904, D(x): 0.61, D(G(z)): 0.23\n",
      "Epoch [118/300], Step [600/938], d_loss: 0.9964, g_loss: 1.8091, D(x): 0.60, D(G(z)): 0.22\n",
      "Epoch [118/300], Step [800/938], d_loss: 0.9161, g_loss: 1.4778, D(x): 0.64, D(G(z)): 0.25\n",
      "Saving fake_images-0118.png\n",
      "Epoch [119/300], Step [200/938], d_loss: 1.1223, g_loss: 1.1976, D(x): 0.67, D(G(z)): 0.35\n",
      "Epoch [119/300], Step [400/938], d_loss: 0.7568, g_loss: 1.5331, D(x): 0.77, D(G(z)): 0.32\n",
      "Epoch [119/300], Step [600/938], d_loss: 1.0394, g_loss: 1.4757, D(x): 0.63, D(G(z)): 0.29\n",
      "Epoch [119/300], Step [800/938], d_loss: 0.8003, g_loss: 1.3963, D(x): 0.68, D(G(z)): 0.24\n",
      "Saving fake_images-0119.png\n",
      "Epoch [120/300], Step [200/938], d_loss: 1.1189, g_loss: 1.3191, D(x): 0.66, D(G(z)): 0.38\n",
      "Epoch [120/300], Step [400/938], d_loss: 0.8463, g_loss: 1.7676, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [120/300], Step [600/938], d_loss: 1.0480, g_loss: 1.2995, D(x): 0.70, D(G(z)): 0.38\n",
      "Epoch [120/300], Step [800/938], d_loss: 0.8643, g_loss: 1.3505, D(x): 0.68, D(G(z)): 0.27\n",
      "Saving fake_images-0120.png\n",
      "Epoch [121/300], Step [200/938], d_loss: 1.0211, g_loss: 1.2995, D(x): 0.69, D(G(z)): 0.37\n",
      "Epoch [121/300], Step [400/938], d_loss: 0.9203, g_loss: 1.4197, D(x): 0.72, D(G(z)): 0.35\n",
      "Epoch [121/300], Step [600/938], d_loss: 0.8789, g_loss: 1.8656, D(x): 0.71, D(G(z)): 0.30\n",
      "Epoch [121/300], Step [800/938], d_loss: 0.9446, g_loss: 1.5532, D(x): 0.68, D(G(z)): 0.29\n",
      "Saving fake_images-0121.png\n",
      "Epoch [122/300], Step [200/938], d_loss: 1.0126, g_loss: 1.1501, D(x): 0.75, D(G(z)): 0.43\n",
      "Epoch [122/300], Step [400/938], d_loss: 0.8018, g_loss: 1.4339, D(x): 0.69, D(G(z)): 0.23\n",
      "Epoch [122/300], Step [600/938], d_loss: 1.0255, g_loss: 1.3847, D(x): 0.61, D(G(z)): 0.26\n",
      "Epoch [122/300], Step [800/938], d_loss: 1.0416, g_loss: 1.3621, D(x): 0.69, D(G(z)): 0.37\n",
      "Saving fake_images-0122.png\n",
      "Epoch [123/300], Step [200/938], d_loss: 0.9515, g_loss: 1.4873, D(x): 0.79, D(G(z)): 0.42\n",
      "Epoch [123/300], Step [400/938], d_loss: 1.0029, g_loss: 1.7723, D(x): 0.62, D(G(z)): 0.27\n",
      "Epoch [123/300], Step [600/938], d_loss: 1.0307, g_loss: 1.7852, D(x): 0.72, D(G(z)): 0.37\n",
      "Epoch [123/300], Step [800/938], d_loss: 0.9775, g_loss: 1.4037, D(x): 0.60, D(G(z)): 0.25\n",
      "Saving fake_images-0123.png\n",
      "Epoch [124/300], Step [200/938], d_loss: 1.1464, g_loss: 1.3783, D(x): 0.72, D(G(z)): 0.42\n",
      "Epoch [124/300], Step [400/938], d_loss: 0.8147, g_loss: 1.4502, D(x): 0.74, D(G(z)): 0.31\n",
      "Epoch [124/300], Step [600/938], d_loss: 0.8937, g_loss: 1.3728, D(x): 0.71, D(G(z)): 0.32\n",
      "Epoch [124/300], Step [800/938], d_loss: 0.9655, g_loss: 1.4924, D(x): 0.65, D(G(z)): 0.27\n",
      "Saving fake_images-0124.png\n",
      "Epoch [125/300], Step [200/938], d_loss: 0.8673, g_loss: 1.4137, D(x): 0.73, D(G(z)): 0.32\n",
      "Epoch [125/300], Step [400/938], d_loss: 0.9992, g_loss: 1.6753, D(x): 0.62, D(G(z)): 0.24\n",
      "Epoch [125/300], Step [600/938], d_loss: 0.9880, g_loss: 1.4472, D(x): 0.64, D(G(z)): 0.28\n",
      "Epoch [125/300], Step [800/938], d_loss: 1.0436, g_loss: 1.6758, D(x): 0.67, D(G(z)): 0.34\n",
      "Saving fake_images-0125.png\n",
      "Epoch [126/300], Step [200/938], d_loss: 1.0563, g_loss: 1.1553, D(x): 0.69, D(G(z)): 0.37\n",
      "Epoch [126/300], Step [400/938], d_loss: 1.0691, g_loss: 1.1698, D(x): 0.65, D(G(z)): 0.33\n",
      "Epoch [126/300], Step [600/938], d_loss: 1.0554, g_loss: 1.3547, D(x): 0.77, D(G(z)): 0.42\n",
      "Epoch [126/300], Step [800/938], d_loss: 0.9094, g_loss: 1.5663, D(x): 0.70, D(G(z)): 0.28\n",
      "Saving fake_images-0126.png\n",
      "Epoch [127/300], Step [200/938], d_loss: 0.9432, g_loss: 1.2102, D(x): 0.80, D(G(z)): 0.38\n",
      "Epoch [127/300], Step [400/938], d_loss: 1.0113, g_loss: 1.2610, D(x): 0.78, D(G(z)): 0.41\n",
      "Epoch [127/300], Step [600/938], d_loss: 0.7142, g_loss: 1.6741, D(x): 0.78, D(G(z)): 0.27\n",
      "Epoch [127/300], Step [800/938], d_loss: 0.9369, g_loss: 1.2773, D(x): 0.77, D(G(z)): 0.38\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0127.png\n",
      "Epoch [128/300], Step [200/938], d_loss: 0.8152, g_loss: 1.6839, D(x): 0.65, D(G(z)): 0.19\n",
      "Epoch [128/300], Step [400/938], d_loss: 0.7260, g_loss: 1.5621, D(x): 0.74, D(G(z)): 0.27\n",
      "Epoch [128/300], Step [600/938], d_loss: 1.0253, g_loss: 1.4101, D(x): 0.64, D(G(z)): 0.33\n",
      "Epoch [128/300], Step [800/938], d_loss: 1.1014, g_loss: 1.7743, D(x): 0.62, D(G(z)): 0.30\n",
      "Saving fake_images-0128.png\n",
      "Epoch [129/300], Step [200/938], d_loss: 0.9833, g_loss: 1.6462, D(x): 0.61, D(G(z)): 0.25\n",
      "Epoch [129/300], Step [400/938], d_loss: 0.7857, g_loss: 1.7621, D(x): 0.71, D(G(z)): 0.25\n",
      "Epoch [129/300], Step [600/938], d_loss: 0.8152, g_loss: 1.5951, D(x): 0.75, D(G(z)): 0.31\n",
      "Epoch [129/300], Step [800/938], d_loss: 1.0692, g_loss: 1.4732, D(x): 0.73, D(G(z)): 0.40\n",
      "Saving fake_images-0129.png\n",
      "Epoch [130/300], Step [200/938], d_loss: 0.9605, g_loss: 1.6154, D(x): 0.65, D(G(z)): 0.28\n",
      "Epoch [130/300], Step [400/938], d_loss: 1.1087, g_loss: 1.6312, D(x): 0.71, D(G(z)): 0.41\n",
      "Epoch [130/300], Step [600/938], d_loss: 1.0574, g_loss: 2.0377, D(x): 0.61, D(G(z)): 0.23\n",
      "Epoch [130/300], Step [800/938], d_loss: 1.1712, g_loss: 1.4150, D(x): 0.74, D(G(z)): 0.44\n",
      "Saving fake_images-0130.png\n",
      "Epoch [131/300], Step [200/938], d_loss: 1.0526, g_loss: 1.6197, D(x): 0.60, D(G(z)): 0.23\n",
      "Epoch [131/300], Step [400/938], d_loss: 0.8846, g_loss: 1.4801, D(x): 0.72, D(G(z)): 0.31\n",
      "Epoch [131/300], Step [600/938], d_loss: 0.9384, g_loss: 1.8021, D(x): 0.71, D(G(z)): 0.30\n",
      "Epoch [131/300], Step [800/938], d_loss: 0.9467, g_loss: 1.5153, D(x): 0.66, D(G(z)): 0.24\n",
      "Saving fake_images-0131.png\n",
      "Epoch [132/300], Step [200/938], d_loss: 0.8628, g_loss: 1.5128, D(x): 0.67, D(G(z)): 0.27\n",
      "Epoch [132/300], Step [400/938], d_loss: 0.8058, g_loss: 1.5216, D(x): 0.74, D(G(z)): 0.28\n",
      "Epoch [132/300], Step [600/938], d_loss: 0.8486, g_loss: 1.3516, D(x): 0.80, D(G(z)): 0.40\n",
      "Epoch [132/300], Step [800/938], d_loss: 0.9097, g_loss: 1.3899, D(x): 0.74, D(G(z)): 0.37\n",
      "Saving fake_images-0132.png\n",
      "Epoch [133/300], Step [200/938], d_loss: 0.8987, g_loss: 2.0554, D(x): 0.76, D(G(z)): 0.31\n",
      "Epoch [133/300], Step [400/938], d_loss: 1.0576, g_loss: 1.4611, D(x): 0.61, D(G(z)): 0.30\n",
      "Epoch [133/300], Step [600/938], d_loss: 0.6789, g_loss: 1.8412, D(x): 0.74, D(G(z)): 0.23\n",
      "Epoch [133/300], Step [800/938], d_loss: 0.8245, g_loss: 1.7991, D(x): 0.71, D(G(z)): 0.29\n",
      "Saving fake_images-0133.png\n",
      "Epoch [134/300], Step [200/938], d_loss: 0.7748, g_loss: 1.6393, D(x): 0.74, D(G(z)): 0.27\n",
      "Epoch [134/300], Step [400/938], d_loss: 0.8991, g_loss: 1.6635, D(x): 0.69, D(G(z)): 0.29\n",
      "Epoch [134/300], Step [600/938], d_loss: 1.2183, g_loss: 1.9331, D(x): 0.60, D(G(z)): 0.33\n",
      "Epoch [134/300], Step [800/938], d_loss: 0.9887, g_loss: 1.1706, D(x): 0.74, D(G(z)): 0.41\n",
      "Saving fake_images-0134.png\n",
      "Epoch [135/300], Step [200/938], d_loss: 1.0483, g_loss: 1.4870, D(x): 0.63, D(G(z)): 0.30\n",
      "Epoch [135/300], Step [400/938], d_loss: 0.9347, g_loss: 1.9748, D(x): 0.62, D(G(z)): 0.22\n",
      "Epoch [135/300], Step [600/938], d_loss: 1.1087, g_loss: 1.8538, D(x): 0.61, D(G(z)): 0.32\n",
      "Epoch [135/300], Step [800/938], d_loss: 1.0027, g_loss: 1.3153, D(x): 0.58, D(G(z)): 0.24\n",
      "Saving fake_images-0135.png\n",
      "Epoch [136/300], Step [200/938], d_loss: 0.9111, g_loss: 1.9450, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [136/300], Step [400/938], d_loss: 0.8418, g_loss: 1.8489, D(x): 0.65, D(G(z)): 0.19\n",
      "Epoch [136/300], Step [600/938], d_loss: 1.0756, g_loss: 1.6575, D(x): 0.66, D(G(z)): 0.36\n",
      "Epoch [136/300], Step [800/938], d_loss: 1.0020, g_loss: 1.6656, D(x): 0.62, D(G(z)): 0.24\n",
      "Saving fake_images-0136.png\n",
      "Epoch [137/300], Step [200/938], d_loss: 1.0590, g_loss: 1.4722, D(x): 0.68, D(G(z)): 0.36\n",
      "Epoch [137/300], Step [400/938], d_loss: 0.9071, g_loss: 1.7938, D(x): 0.69, D(G(z)): 0.30\n",
      "Epoch [137/300], Step [600/938], d_loss: 0.8908, g_loss: 1.3171, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [137/300], Step [800/938], d_loss: 0.9464, g_loss: 1.3015, D(x): 0.78, D(G(z)): 0.41\n",
      "Saving fake_images-0137.png\n",
      "Epoch [138/300], Step [200/938], d_loss: 0.9846, g_loss: 1.2575, D(x): 0.72, D(G(z)): 0.36\n",
      "Epoch [138/300], Step [400/938], d_loss: 0.8996, g_loss: 1.4192, D(x): 0.66, D(G(z)): 0.26\n",
      "Epoch [138/300], Step [600/938], d_loss: 0.9262, g_loss: 1.4819, D(x): 0.63, D(G(z)): 0.26\n",
      "Epoch [138/300], Step [800/938], d_loss: 0.8560, g_loss: 1.9242, D(x): 0.68, D(G(z)): 0.27\n",
      "Saving fake_images-0138.png\n",
      "Epoch [139/300], Step [200/938], d_loss: 0.8020, g_loss: 1.5925, D(x): 0.74, D(G(z)): 0.29\n",
      "Epoch [139/300], Step [400/938], d_loss: 0.9624, g_loss: 1.6538, D(x): 0.60, D(G(z)): 0.24\n",
      "Epoch [139/300], Step [600/938], d_loss: 1.0475, g_loss: 1.5767, D(x): 0.67, D(G(z)): 0.30\n",
      "Epoch [139/300], Step [800/938], d_loss: 1.1430, g_loss: 1.2182, D(x): 0.70, D(G(z)): 0.41\n",
      "Saving fake_images-0139.png\n",
      "Epoch [140/300], Step [200/938], d_loss: 0.9441, g_loss: 2.0365, D(x): 0.66, D(G(z)): 0.27\n",
      "Epoch [140/300], Step [400/938], d_loss: 0.9764, g_loss: 1.4603, D(x): 0.70, D(G(z)): 0.35\n",
      "Epoch [140/300], Step [600/938], d_loss: 0.9273, g_loss: 1.3494, D(x): 0.72, D(G(z)): 0.32\n",
      "Epoch [140/300], Step [800/938], d_loss: 0.7847, g_loss: 1.6872, D(x): 0.73, D(G(z)): 0.28\n",
      "Saving fake_images-0140.png\n",
      "Epoch [141/300], Step [200/938], d_loss: 0.9038, g_loss: 1.5135, D(x): 0.77, D(G(z)): 0.38\n",
      "Epoch [141/300], Step [400/938], d_loss: 1.0767, g_loss: 1.2577, D(x): 0.80, D(G(z)): 0.48\n",
      "Epoch [141/300], Step [600/938], d_loss: 1.1428, g_loss: 1.5367, D(x): 0.55, D(G(z)): 0.19\n",
      "Epoch [141/300], Step [800/938], d_loss: 0.8865, g_loss: 1.3969, D(x): 0.82, D(G(z)): 0.41\n",
      "Saving fake_images-0141.png\n",
      "Epoch [142/300], Step [200/938], d_loss: 0.8138, g_loss: 1.3634, D(x): 0.72, D(G(z)): 0.28\n",
      "Epoch [142/300], Step [400/938], d_loss: 0.9185, g_loss: 1.3478, D(x): 0.66, D(G(z)): 0.27\n",
      "Epoch [142/300], Step [600/938], d_loss: 0.8889, g_loss: 2.1527, D(x): 0.66, D(G(z)): 0.21\n",
      "Epoch [142/300], Step [800/938], d_loss: 1.0886, g_loss: 1.0580, D(x): 0.60, D(G(z)): 0.31\n",
      "Saving fake_images-0142.png\n",
      "Epoch [143/300], Step [200/938], d_loss: 0.8597, g_loss: 1.9777, D(x): 0.72, D(G(z)): 0.26\n",
      "Epoch [143/300], Step [400/938], d_loss: 0.8043, g_loss: 1.5688, D(x): 0.80, D(G(z)): 0.38\n",
      "Epoch [143/300], Step [600/938], d_loss: 1.0416, g_loss: 1.4527, D(x): 0.70, D(G(z)): 0.36\n",
      "Epoch [143/300], Step [800/938], d_loss: 0.8437, g_loss: 1.4637, D(x): 0.79, D(G(z)): 0.35\n",
      "Saving fake_images-0143.png\n",
      "Epoch [144/300], Step [200/938], d_loss: 0.8012, g_loss: 1.8089, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [144/300], Step [400/938], d_loss: 1.0631, g_loss: 1.4955, D(x): 0.71, D(G(z)): 0.39\n",
      "Epoch [144/300], Step [600/938], d_loss: 0.8995, g_loss: 1.4083, D(x): 0.69, D(G(z)): 0.30\n",
      "Epoch [144/300], Step [800/938], d_loss: 1.0494, g_loss: 1.5825, D(x): 0.60, D(G(z)): 0.24\n",
      "Saving fake_images-0144.png\n",
      "Epoch [145/300], Step [200/938], d_loss: 0.8845, g_loss: 1.5474, D(x): 0.75, D(G(z)): 0.31\n",
      "Epoch [145/300], Step [400/938], d_loss: 0.9263, g_loss: 1.3530, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [145/300], Step [600/938], d_loss: 0.9167, g_loss: 1.5348, D(x): 0.73, D(G(z)): 0.35\n",
      "Epoch [145/300], Step [800/938], d_loss: 0.8563, g_loss: 1.0835, D(x): 0.76, D(G(z)): 0.37\n",
      "Saving fake_images-0145.png\n",
      "Epoch [146/300], Step [200/938], d_loss: 0.8997, g_loss: 1.5563, D(x): 0.69, D(G(z)): 0.31\n",
      "Epoch [146/300], Step [400/938], d_loss: 0.8740, g_loss: 1.4347, D(x): 0.66, D(G(z)): 0.27\n",
      "Epoch [146/300], Step [600/938], d_loss: 1.0205, g_loss: 1.6431, D(x): 0.70, D(G(z)): 0.38\n",
      "Epoch [146/300], Step [800/938], d_loss: 0.9404, g_loss: 1.4416, D(x): 0.72, D(G(z)): 0.32\n",
      "Saving fake_images-0146.png\n",
      "Epoch [147/300], Step [200/938], d_loss: 0.9658, g_loss: 1.9716, D(x): 0.56, D(G(z)): 0.17\n",
      "Epoch [147/300], Step [400/938], d_loss: 0.9214, g_loss: 1.3632, D(x): 0.69, D(G(z)): 0.28\n",
      "Epoch [147/300], Step [600/938], d_loss: 0.9202, g_loss: 1.5655, D(x): 0.71, D(G(z)): 0.30\n",
      "Epoch [147/300], Step [800/938], d_loss: 0.9960, g_loss: 1.6399, D(x): 0.66, D(G(z)): 0.29\n",
      "Saving fake_images-0147.png\n",
      "Epoch [148/300], Step [200/938], d_loss: 0.7088, g_loss: 1.7815, D(x): 0.68, D(G(z)): 0.17\n",
      "Epoch [148/300], Step [400/938], d_loss: 1.0747, g_loss: 1.3152, D(x): 0.64, D(G(z)): 0.35\n",
      "Epoch [148/300], Step [600/938], d_loss: 0.9451, g_loss: 1.3530, D(x): 0.68, D(G(z)): 0.30\n",
      "Epoch [148/300], Step [800/938], d_loss: 0.9750, g_loss: 1.2159, D(x): 0.64, D(G(z)): 0.26\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0148.png\n",
      "Epoch [149/300], Step [200/938], d_loss: 1.0561, g_loss: 1.4496, D(x): 0.80, D(G(z)): 0.45\n",
      "Epoch [149/300], Step [400/938], d_loss: 0.8506, g_loss: 1.4423, D(x): 0.74, D(G(z)): 0.28\n",
      "Epoch [149/300], Step [600/938], d_loss: 0.9446, g_loss: 1.4591, D(x): 0.74, D(G(z)): 0.38\n",
      "Epoch [149/300], Step [800/938], d_loss: 0.8389, g_loss: 1.5735, D(x): 0.75, D(G(z)): 0.32\n",
      "Saving fake_images-0149.png\n",
      "Epoch [150/300], Step [200/938], d_loss: 0.8592, g_loss: 1.8687, D(x): 0.72, D(G(z)): 0.31\n",
      "Epoch [150/300], Step [400/938], d_loss: 0.8867, g_loss: 1.8177, D(x): 0.67, D(G(z)): 0.22\n",
      "Epoch [150/300], Step [600/938], d_loss: 0.9174, g_loss: 1.4092, D(x): 0.74, D(G(z)): 0.31\n",
      "Epoch [150/300], Step [800/938], d_loss: 0.8046, g_loss: 1.7824, D(x): 0.70, D(G(z)): 0.26\n",
      "Saving fake_images-0150.png\n",
      "Epoch [151/300], Step [200/938], d_loss: 1.0545, g_loss: 1.5534, D(x): 0.61, D(G(z)): 0.25\n",
      "Epoch [151/300], Step [400/938], d_loss: 0.7224, g_loss: 1.7631, D(x): 0.69, D(G(z)): 0.20\n",
      "Epoch [151/300], Step [600/938], d_loss: 0.8928, g_loss: 1.7078, D(x): 0.76, D(G(z)): 0.35\n",
      "Epoch [151/300], Step [800/938], d_loss: 0.7497, g_loss: 1.6356, D(x): 0.78, D(G(z)): 0.28\n",
      "Saving fake_images-0151.png\n",
      "\t*** SAVED MODEL FOR EPOCH 150 ***\n",
      "Epoch [152/300], Step [200/938], d_loss: 0.8829, g_loss: 1.5838, D(x): 0.69, D(G(z)): 0.28\n",
      "Epoch [152/300], Step [400/938], d_loss: 0.8885, g_loss: 1.5643, D(x): 0.68, D(G(z)): 0.31\n",
      "Epoch [152/300], Step [600/938], d_loss: 0.9402, g_loss: 1.8636, D(x): 0.63, D(G(z)): 0.25\n",
      "Epoch [152/300], Step [800/938], d_loss: 0.9434, g_loss: 1.2844, D(x): 0.73, D(G(z)): 0.35\n",
      "Saving fake_images-0152.png\n",
      "Epoch [153/300], Step [200/938], d_loss: 0.7621, g_loss: 1.8669, D(x): 0.75, D(G(z)): 0.28\n",
      "Epoch [153/300], Step [400/938], d_loss: 0.9571, g_loss: 1.2097, D(x): 0.66, D(G(z)): 0.32\n",
      "Epoch [153/300], Step [600/938], d_loss: 1.0207, g_loss: 1.2408, D(x): 0.74, D(G(z)): 0.36\n",
      "Epoch [153/300], Step [800/938], d_loss: 1.0114, g_loss: 1.3300, D(x): 0.76, D(G(z)): 0.40\n",
      "Saving fake_images-0153.png\n",
      "Epoch [154/300], Step [200/938], d_loss: 0.9008, g_loss: 1.6208, D(x): 0.70, D(G(z)): 0.30\n",
      "Epoch [154/300], Step [400/938], d_loss: 1.0526, g_loss: 1.3839, D(x): 0.64, D(G(z)): 0.30\n",
      "Epoch [154/300], Step [600/938], d_loss: 0.7529, g_loss: 2.0352, D(x): 0.72, D(G(z)): 0.24\n",
      "Epoch [154/300], Step [800/938], d_loss: 0.9869, g_loss: 1.1616, D(x): 0.66, D(G(z)): 0.30\n",
      "Saving fake_images-0154.png\n",
      "Epoch [155/300], Step [200/938], d_loss: 1.1741, g_loss: 1.4156, D(x): 0.57, D(G(z)): 0.29\n",
      "Epoch [155/300], Step [400/938], d_loss: 0.9849, g_loss: 1.9655, D(x): 0.68, D(G(z)): 0.36\n",
      "Epoch [155/300], Step [600/938], d_loss: 0.7802, g_loss: 1.8836, D(x): 0.68, D(G(z)): 0.18\n",
      "Epoch [155/300], Step [800/938], d_loss: 1.0046, g_loss: 1.5605, D(x): 0.69, D(G(z)): 0.31\n",
      "Saving fake_images-0155.png\n",
      "Epoch [156/300], Step [200/938], d_loss: 0.9725, g_loss: 1.4976, D(x): 0.70, D(G(z)): 0.34\n",
      "Epoch [156/300], Step [400/938], d_loss: 0.9013, g_loss: 1.3907, D(x): 0.67, D(G(z)): 0.26\n",
      "Epoch [156/300], Step [600/938], d_loss: 0.9168, g_loss: 1.2922, D(x): 0.74, D(G(z)): 0.36\n",
      "Epoch [156/300], Step [800/938], d_loss: 1.2507, g_loss: 2.1953, D(x): 0.49, D(G(z)): 0.18\n",
      "Saving fake_images-0156.png\n",
      "Epoch [157/300], Step [200/938], d_loss: 0.9722, g_loss: 1.5322, D(x): 0.65, D(G(z)): 0.28\n",
      "Epoch [157/300], Step [400/938], d_loss: 0.9421, g_loss: 1.5576, D(x): 0.69, D(G(z)): 0.29\n",
      "Epoch [157/300], Step [600/938], d_loss: 0.8580, g_loss: 1.3919, D(x): 0.72, D(G(z)): 0.32\n",
      "Epoch [157/300], Step [800/938], d_loss: 0.7997, g_loss: 2.2510, D(x): 0.70, D(G(z)): 0.24\n",
      "Saving fake_images-0157.png\n",
      "Epoch [158/300], Step [200/938], d_loss: 0.9882, g_loss: 1.8465, D(x): 0.77, D(G(z)): 0.40\n",
      "Epoch [158/300], Step [400/938], d_loss: 0.9126, g_loss: 1.4291, D(x): 0.73, D(G(z)): 0.33\n",
      "Epoch [158/300], Step [600/938], d_loss: 0.7426, g_loss: 1.5674, D(x): 0.82, D(G(z)): 0.33\n",
      "Epoch [158/300], Step [800/938], d_loss: 1.0643, g_loss: 1.3563, D(x): 0.66, D(G(z)): 0.32\n",
      "Saving fake_images-0158.png\n",
      "Epoch [159/300], Step [200/938], d_loss: 1.0409, g_loss: 1.5160, D(x): 0.67, D(G(z)): 0.34\n",
      "Epoch [159/300], Step [400/938], d_loss: 0.9655, g_loss: 1.8866, D(x): 0.70, D(G(z)): 0.33\n",
      "Epoch [159/300], Step [600/938], d_loss: 0.8289, g_loss: 1.1895, D(x): 0.80, D(G(z)): 0.34\n",
      "Epoch [159/300], Step [800/938], d_loss: 1.1579, g_loss: 1.9677, D(x): 0.58, D(G(z)): 0.25\n",
      "Saving fake_images-0159.png\n",
      "Epoch [160/300], Step [200/938], d_loss: 1.0065, g_loss: 1.2645, D(x): 0.68, D(G(z)): 0.34\n",
      "Epoch [160/300], Step [400/938], d_loss: 0.9568, g_loss: 1.2061, D(x): 0.63, D(G(z)): 0.24\n",
      "Epoch [160/300], Step [600/938], d_loss: 1.0051, g_loss: 1.7411, D(x): 0.73, D(G(z)): 0.37\n",
      "Epoch [160/300], Step [800/938], d_loss: 0.8525, g_loss: 1.8527, D(x): 0.71, D(G(z)): 0.30\n",
      "Saving fake_images-0160.png\n",
      "Epoch [161/300], Step [200/938], d_loss: 0.8162, g_loss: 1.5158, D(x): 0.76, D(G(z)): 0.31\n",
      "Epoch [161/300], Step [400/938], d_loss: 0.9795, g_loss: 1.4795, D(x): 0.73, D(G(z)): 0.38\n",
      "Epoch [161/300], Step [600/938], d_loss: 0.9760, g_loss: 1.7001, D(x): 0.70, D(G(z)): 0.35\n",
      "Epoch [161/300], Step [800/938], d_loss: 0.7877, g_loss: 1.7276, D(x): 0.69, D(G(z)): 0.25\n",
      "Saving fake_images-0161.png\n",
      "Epoch [162/300], Step [200/938], d_loss: 1.2762, g_loss: 1.6249, D(x): 0.78, D(G(z)): 0.46\n",
      "Epoch [162/300], Step [400/938], d_loss: 0.8448, g_loss: 1.4031, D(x): 0.76, D(G(z)): 0.33\n",
      "Epoch [162/300], Step [600/938], d_loss: 0.8516, g_loss: 1.7398, D(x): 0.76, D(G(z)): 0.35\n",
      "Epoch [162/300], Step [800/938], d_loss: 0.9917, g_loss: 1.2733, D(x): 0.75, D(G(z)): 0.42\n",
      "Saving fake_images-0162.png\n",
      "Epoch [163/300], Step [200/938], d_loss: 0.9286, g_loss: 1.7703, D(x): 0.66, D(G(z)): 0.27\n",
      "Epoch [163/300], Step [400/938], d_loss: 0.7620, g_loss: 1.7871, D(x): 0.77, D(G(z)): 0.29\n",
      "Epoch [163/300], Step [600/938], d_loss: 0.8589, g_loss: 1.7361, D(x): 0.63, D(G(z)): 0.24\n",
      "Epoch [163/300], Step [800/938], d_loss: 0.9872, g_loss: 1.6838, D(x): 0.72, D(G(z)): 0.35\n",
      "Saving fake_images-0163.png\n",
      "Epoch [164/300], Step [200/938], d_loss: 1.1105, g_loss: 1.7981, D(x): 0.71, D(G(z)): 0.37\n",
      "Epoch [164/300], Step [400/938], d_loss: 0.7493, g_loss: 1.7257, D(x): 0.70, D(G(z)): 0.21\n",
      "Epoch [164/300], Step [600/938], d_loss: 0.9131, g_loss: 1.3763, D(x): 0.75, D(G(z)): 0.36\n",
      "Epoch [164/300], Step [800/938], d_loss: 0.8637, g_loss: 1.7334, D(x): 0.69, D(G(z)): 0.29\n",
      "Saving fake_images-0164.png\n",
      "Epoch [165/300], Step [200/938], d_loss: 0.8652, g_loss: 1.5394, D(x): 0.77, D(G(z)): 0.37\n",
      "Epoch [165/300], Step [400/938], d_loss: 1.0400, g_loss: 1.6270, D(x): 0.71, D(G(z)): 0.38\n",
      "Epoch [165/300], Step [600/938], d_loss: 0.9106, g_loss: 1.2431, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [165/300], Step [800/938], d_loss: 0.7913, g_loss: 1.6350, D(x): 0.73, D(G(z)): 0.29\n",
      "Saving fake_images-0165.png\n",
      "Epoch [166/300], Step [200/938], d_loss: 1.0517, g_loss: 1.6811, D(x): 0.61, D(G(z)): 0.24\n",
      "Epoch [166/300], Step [400/938], d_loss: 0.9242, g_loss: 1.5940, D(x): 0.75, D(G(z)): 0.37\n",
      "Epoch [166/300], Step [600/938], d_loss: 0.8902, g_loss: 1.7569, D(x): 0.71, D(G(z)): 0.31\n",
      "Epoch [166/300], Step [800/938], d_loss: 0.7585, g_loss: 1.1820, D(x): 0.75, D(G(z)): 0.30\n",
      "Saving fake_images-0166.png\n",
      "Epoch [167/300], Step [200/938], d_loss: 0.9231, g_loss: 1.4196, D(x): 0.71, D(G(z)): 0.32\n",
      "Epoch [167/300], Step [400/938], d_loss: 1.0081, g_loss: 1.5545, D(x): 0.77, D(G(z)): 0.38\n",
      "Epoch [167/300], Step [600/938], d_loss: 0.9041, g_loss: 1.5155, D(x): 0.60, D(G(z)): 0.19\n",
      "Epoch [167/300], Step [800/938], d_loss: 0.7747, g_loss: 1.6774, D(x): 0.70, D(G(z)): 0.23\n",
      "Saving fake_images-0167.png\n",
      "Epoch [168/300], Step [200/938], d_loss: 0.9694, g_loss: 1.7443, D(x): 0.60, D(G(z)): 0.21\n",
      "Epoch [168/300], Step [400/938], d_loss: 1.0567, g_loss: 1.2607, D(x): 0.75, D(G(z)): 0.40\n",
      "Epoch [168/300], Step [600/938], d_loss: 0.7621, g_loss: 1.4817, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [168/300], Step [800/938], d_loss: 0.7411, g_loss: 1.5196, D(x): 0.79, D(G(z)): 0.31\n",
      "Saving fake_images-0168.png\n",
      "Epoch [169/300], Step [200/938], d_loss: 0.7494, g_loss: 1.4554, D(x): 0.77, D(G(z)): 0.31\n",
      "Epoch [169/300], Step [400/938], d_loss: 0.8190, g_loss: 1.4358, D(x): 0.78, D(G(z)): 0.34\n",
      "Epoch [169/300], Step [600/938], d_loss: 0.9683, g_loss: 1.6265, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [169/300], Step [800/938], d_loss: 0.9401, g_loss: 1.7641, D(x): 0.63, D(G(z)): 0.26\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0169.png\n",
      "Epoch [170/300], Step [200/938], d_loss: 1.0794, g_loss: 1.6908, D(x): 0.62, D(G(z)): 0.29\n",
      "Epoch [170/300], Step [400/938], d_loss: 1.0900, g_loss: 2.1019, D(x): 0.54, D(G(z)): 0.16\n",
      "Epoch [170/300], Step [600/938], d_loss: 0.8836, g_loss: 1.5566, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [170/300], Step [800/938], d_loss: 0.8846, g_loss: 1.7343, D(x): 0.75, D(G(z)): 0.34\n",
      "Saving fake_images-0170.png\n",
      "Epoch [171/300], Step [200/938], d_loss: 0.8145, g_loss: 1.7382, D(x): 0.76, D(G(z)): 0.30\n",
      "Epoch [171/300], Step [400/938], d_loss: 0.7909, g_loss: 1.7799, D(x): 0.74, D(G(z)): 0.31\n",
      "Epoch [171/300], Step [600/938], d_loss: 0.7742, g_loss: 1.8783, D(x): 0.70, D(G(z)): 0.25\n",
      "Epoch [171/300], Step [800/938], d_loss: 0.9306, g_loss: 1.4958, D(x): 0.67, D(G(z)): 0.29\n",
      "Saving fake_images-0171.png\n",
      "Epoch [172/300], Step [200/938], d_loss: 0.9182, g_loss: 1.5642, D(x): 0.68, D(G(z)): 0.31\n",
      "Epoch [172/300], Step [400/938], d_loss: 0.9038, g_loss: 1.5955, D(x): 0.63, D(G(z)): 0.23\n",
      "Epoch [172/300], Step [600/938], d_loss: 1.0354, g_loss: 1.3835, D(x): 0.73, D(G(z)): 0.36\n",
      "Epoch [172/300], Step [800/938], d_loss: 0.9503, g_loss: 1.4394, D(x): 0.68, D(G(z)): 0.28\n",
      "Saving fake_images-0172.png\n",
      "Epoch [173/300], Step [200/938], d_loss: 0.9986, g_loss: 1.6007, D(x): 0.61, D(G(z)): 0.26\n",
      "Epoch [173/300], Step [400/938], d_loss: 1.0998, g_loss: 1.4432, D(x): 0.61, D(G(z)): 0.28\n",
      "Epoch [173/300], Step [600/938], d_loss: 1.1015, g_loss: 1.7487, D(x): 0.63, D(G(z)): 0.31\n",
      "Epoch [173/300], Step [800/938], d_loss: 0.9866, g_loss: 1.3597, D(x): 0.74, D(G(z)): 0.38\n",
      "Saving fake_images-0173.png\n",
      "Epoch [174/300], Step [200/938], d_loss: 0.7495, g_loss: 1.6329, D(x): 0.73, D(G(z)): 0.26\n",
      "Epoch [174/300], Step [400/938], d_loss: 0.9049, g_loss: 2.0485, D(x): 0.72, D(G(z)): 0.31\n",
      "Epoch [174/300], Step [600/938], d_loss: 0.6836, g_loss: 1.5392, D(x): 0.71, D(G(z)): 0.19\n",
      "Epoch [174/300], Step [800/938], d_loss: 0.9673, g_loss: 1.4521, D(x): 0.66, D(G(z)): 0.28\n",
      "Saving fake_images-0174.png\n",
      "Epoch [175/300], Step [200/938], d_loss: 0.9685, g_loss: 1.6939, D(x): 0.69, D(G(z)): 0.34\n",
      "Epoch [175/300], Step [400/938], d_loss: 0.8764, g_loss: 1.6733, D(x): 0.64, D(G(z)): 0.24\n",
      "Epoch [175/300], Step [600/938], d_loss: 0.9450, g_loss: 1.5096, D(x): 0.80, D(G(z)): 0.40\n",
      "Epoch [175/300], Step [800/938], d_loss: 0.8725, g_loss: 1.6969, D(x): 0.64, D(G(z)): 0.21\n",
      "Saving fake_images-0175.png\n",
      "Epoch [176/300], Step [200/938], d_loss: 0.8800, g_loss: 1.5415, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [176/300], Step [400/938], d_loss: 0.9143, g_loss: 1.6793, D(x): 0.72, D(G(z)): 0.32\n",
      "Epoch [176/300], Step [600/938], d_loss: 0.9446, g_loss: 1.4346, D(x): 0.69, D(G(z)): 0.32\n",
      "Epoch [176/300], Step [800/938], d_loss: 0.8669, g_loss: 1.7935, D(x): 0.74, D(G(z)): 0.36\n",
      "Saving fake_images-0176.png\n",
      "Epoch [177/300], Step [200/938], d_loss: 0.7464, g_loss: 1.1857, D(x): 0.76, D(G(z)): 0.30\n",
      "Epoch [177/300], Step [400/938], d_loss: 0.8981, g_loss: 1.4614, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [177/300], Step [600/938], d_loss: 0.9637, g_loss: 1.6234, D(x): 0.68, D(G(z)): 0.33\n",
      "Epoch [177/300], Step [800/938], d_loss: 0.8739, g_loss: 1.7743, D(x): 0.74, D(G(z)): 0.33\n",
      "Saving fake_images-0177.png\n",
      "Epoch [178/300], Step [200/938], d_loss: 0.8196, g_loss: 1.6691, D(x): 0.75, D(G(z)): 0.28\n",
      "Epoch [178/300], Step [400/938], d_loss: 0.7784, g_loss: 2.2650, D(x): 0.65, D(G(z)): 0.17\n",
      "Epoch [178/300], Step [600/938], d_loss: 0.8197, g_loss: 1.3389, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [178/300], Step [800/938], d_loss: 0.9203, g_loss: 1.7938, D(x): 0.66, D(G(z)): 0.26\n",
      "Saving fake_images-0178.png\n",
      "Epoch [179/300], Step [200/938], d_loss: 0.9733, g_loss: 1.8833, D(x): 0.64, D(G(z)): 0.25\n",
      "Epoch [179/300], Step [400/938], d_loss: 0.7518, g_loss: 1.4066, D(x): 0.77, D(G(z)): 0.31\n",
      "Epoch [179/300], Step [600/938], d_loss: 0.8637, g_loss: 1.2849, D(x): 0.76, D(G(z)): 0.33\n",
      "Epoch [179/300], Step [800/938], d_loss: 1.0164, g_loss: 1.6262, D(x): 0.63, D(G(z)): 0.27\n",
      "Saving fake_images-0179.png\n",
      "Epoch [180/300], Step [200/938], d_loss: 0.9587, g_loss: 1.5201, D(x): 0.72, D(G(z)): 0.36\n",
      "Epoch [180/300], Step [400/938], d_loss: 0.9057, g_loss: 1.4938, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [180/300], Step [600/938], d_loss: 0.7159, g_loss: 1.6603, D(x): 0.78, D(G(z)): 0.28\n",
      "Epoch [180/300], Step [800/938], d_loss: 0.9346, g_loss: 1.4714, D(x): 0.75, D(G(z)): 0.40\n",
      "Saving fake_images-0180.png\n",
      "Epoch [181/300], Step [200/938], d_loss: 0.9775, g_loss: 1.4809, D(x): 0.64, D(G(z)): 0.29\n",
      "Epoch [181/300], Step [400/938], d_loss: 0.7977, g_loss: 1.8373, D(x): 0.75, D(G(z)): 0.33\n",
      "Epoch [181/300], Step [600/938], d_loss: 0.7446, g_loss: 1.2336, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [181/300], Step [800/938], d_loss: 0.9634, g_loss: 1.4818, D(x): 0.65, D(G(z)): 0.30\n",
      "Saving fake_images-0181.png\n",
      "Epoch [182/300], Step [200/938], d_loss: 0.9648, g_loss: 1.4197, D(x): 0.64, D(G(z)): 0.28\n",
      "Epoch [182/300], Step [400/938], d_loss: 0.8619, g_loss: 1.5323, D(x): 0.71, D(G(z)): 0.28\n",
      "Epoch [182/300], Step [600/938], d_loss: 0.9406, g_loss: 1.4291, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [182/300], Step [800/938], d_loss: 0.6997, g_loss: 2.1121, D(x): 0.79, D(G(z)): 0.27\n",
      "Saving fake_images-0182.png\n",
      "Epoch [183/300], Step [200/938], d_loss: 0.8919, g_loss: 1.9159, D(x): 0.68, D(G(z)): 0.25\n",
      "Epoch [183/300], Step [400/938], d_loss: 0.7689, g_loss: 2.0642, D(x): 0.68, D(G(z)): 0.22\n",
      "Epoch [183/300], Step [600/938], d_loss: 0.9060, g_loss: 1.7499, D(x): 0.73, D(G(z)): 0.32\n",
      "Epoch [183/300], Step [800/938], d_loss: 0.9505, g_loss: 1.3971, D(x): 0.66, D(G(z)): 0.29\n",
      "Saving fake_images-0183.png\n",
      "Epoch [184/300], Step [200/938], d_loss: 0.8393, g_loss: 1.4716, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [184/300], Step [400/938], d_loss: 0.8883, g_loss: 1.3423, D(x): 0.76, D(G(z)): 0.36\n",
      "Epoch [184/300], Step [600/938], d_loss: 0.9031, g_loss: 1.3682, D(x): 0.66, D(G(z)): 0.22\n",
      "Epoch [184/300], Step [800/938], d_loss: 0.9010, g_loss: 1.6213, D(x): 0.68, D(G(z)): 0.25\n",
      "Saving fake_images-0184.png\n",
      "Epoch [185/300], Step [200/938], d_loss: 0.8986, g_loss: 1.6606, D(x): 0.72, D(G(z)): 0.32\n",
      "Epoch [185/300], Step [400/938], d_loss: 0.9100, g_loss: 1.8286, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [185/300], Step [600/938], d_loss: 0.7235, g_loss: 1.2724, D(x): 0.78, D(G(z)): 0.29\n",
      "Epoch [185/300], Step [800/938], d_loss: 0.8763, g_loss: 1.6061, D(x): 0.66, D(G(z)): 0.21\n",
      "Saving fake_images-0185.png\n",
      "Epoch [186/300], Step [200/938], d_loss: 1.1291, g_loss: 1.4057, D(x): 0.65, D(G(z)): 0.33\n",
      "Epoch [186/300], Step [400/938], d_loss: 0.8871, g_loss: 1.6474, D(x): 0.68, D(G(z)): 0.28\n",
      "Epoch [186/300], Step [600/938], d_loss: 1.0078, g_loss: 1.7195, D(x): 0.64, D(G(z)): 0.30\n",
      "Epoch [186/300], Step [800/938], d_loss: 1.1079, g_loss: 1.7703, D(x): 0.62, D(G(z)): 0.30\n",
      "Saving fake_images-0186.png\n",
      "Epoch [187/300], Step [200/938], d_loss: 0.9854, g_loss: 1.4122, D(x): 0.71, D(G(z)): 0.37\n",
      "Epoch [187/300], Step [400/938], d_loss: 0.8796, g_loss: 1.6326, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [187/300], Step [600/938], d_loss: 0.9506, g_loss: 1.4758, D(x): 0.72, D(G(z)): 0.35\n",
      "Epoch [187/300], Step [800/938], d_loss: 0.9346, g_loss: 1.1165, D(x): 0.77, D(G(z)): 0.39\n",
      "Saving fake_images-0187.png\n",
      "Epoch [188/300], Step [200/938], d_loss: 0.8595, g_loss: 1.5087, D(x): 0.74, D(G(z)): 0.33\n",
      "Epoch [188/300], Step [400/938], d_loss: 0.9179, g_loss: 1.8856, D(x): 0.64, D(G(z)): 0.21\n",
      "Epoch [188/300], Step [600/938], d_loss: 0.9179, g_loss: 1.5478, D(x): 0.78, D(G(z)): 0.41\n",
      "Epoch [188/300], Step [800/938], d_loss: 0.8807, g_loss: 2.0828, D(x): 0.67, D(G(z)): 0.21\n",
      "Saving fake_images-0188.png\n",
      "Epoch [189/300], Step [200/938], d_loss: 0.9065, g_loss: 1.4046, D(x): 0.65, D(G(z)): 0.23\n",
      "Epoch [189/300], Step [400/938], d_loss: 0.9397, g_loss: 1.1823, D(x): 0.72, D(G(z)): 0.34\n",
      "Epoch [189/300], Step [600/938], d_loss: 0.8900, g_loss: 1.6719, D(x): 0.65, D(G(z)): 0.25\n",
      "Epoch [189/300], Step [800/938], d_loss: 0.9238, g_loss: 1.4365, D(x): 0.70, D(G(z)): 0.32\n",
      "Saving fake_images-0189.png\n",
      "Epoch [190/300], Step [200/938], d_loss: 1.1293, g_loss: 1.3613, D(x): 0.69, D(G(z)): 0.39\n",
      "Epoch [190/300], Step [400/938], d_loss: 0.8452, g_loss: 1.2816, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [190/300], Step [600/938], d_loss: 0.9954, g_loss: 1.6051, D(x): 0.73, D(G(z)): 0.39\n",
      "Epoch [190/300], Step [800/938], d_loss: 0.6984, g_loss: 1.7628, D(x): 0.77, D(G(z)): 0.27\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0190.png\n",
      "Epoch [191/300], Step [200/938], d_loss: 0.8370, g_loss: 1.8205, D(x): 0.73, D(G(z)): 0.29\n",
      "Epoch [191/300], Step [400/938], d_loss: 0.7622, g_loss: 1.6780, D(x): 0.80, D(G(z)): 0.35\n",
      "Epoch [191/300], Step [600/938], d_loss: 0.8896, g_loss: 1.6926, D(x): 0.79, D(G(z)): 0.36\n",
      "Epoch [191/300], Step [800/938], d_loss: 0.9984, g_loss: 1.5398, D(x): 0.72, D(G(z)): 0.36\n",
      "Saving fake_images-0191.png\n",
      "Epoch [192/300], Step [200/938], d_loss: 0.9385, g_loss: 1.6627, D(x): 0.69, D(G(z)): 0.30\n",
      "Epoch [192/300], Step [400/938], d_loss: 0.8761, g_loss: 1.6353, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [192/300], Step [600/938], d_loss: 1.0718, g_loss: 1.8735, D(x): 0.67, D(G(z)): 0.32\n",
      "Epoch [192/300], Step [800/938], d_loss: 1.0782, g_loss: 1.1330, D(x): 0.76, D(G(z)): 0.41\n",
      "Saving fake_images-0192.png\n",
      "Epoch [193/300], Step [200/938], d_loss: 0.7901, g_loss: 1.3637, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [193/300], Step [400/938], d_loss: 0.6515, g_loss: 1.5247, D(x): 0.82, D(G(z)): 0.30\n",
      "Epoch [193/300], Step [600/938], d_loss: 0.9313, g_loss: 1.4896, D(x): 0.75, D(G(z)): 0.35\n",
      "Epoch [193/300], Step [800/938], d_loss: 0.9824, g_loss: 2.0200, D(x): 0.64, D(G(z)): 0.27\n",
      "Saving fake_images-0193.png\n",
      "Epoch [194/300], Step [200/938], d_loss: 0.7728, g_loss: 1.6878, D(x): 0.71, D(G(z)): 0.25\n",
      "Epoch [194/300], Step [400/938], d_loss: 0.8965, g_loss: 1.6102, D(x): 0.69, D(G(z)): 0.29\n",
      "Epoch [194/300], Step [600/938], d_loss: 0.8790, g_loss: 1.6244, D(x): 0.71, D(G(z)): 0.30\n",
      "Epoch [194/300], Step [800/938], d_loss: 1.0646, g_loss: 1.5505, D(x): 0.74, D(G(z)): 0.41\n",
      "Saving fake_images-0194.png\n",
      "Epoch [195/300], Step [200/938], d_loss: 0.7813, g_loss: 1.5860, D(x): 0.74, D(G(z)): 0.31\n",
      "Epoch [195/300], Step [400/938], d_loss: 0.7024, g_loss: 1.5388, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [195/300], Step [600/938], d_loss: 1.0527, g_loss: 2.0256, D(x): 0.61, D(G(z)): 0.26\n",
      "Epoch [195/300], Step [800/938], d_loss: 0.8715, g_loss: 1.5444, D(x): 0.71, D(G(z)): 0.32\n",
      "Saving fake_images-0195.png\n",
      "Epoch [196/300], Step [200/938], d_loss: 0.8264, g_loss: 1.7877, D(x): 0.74, D(G(z)): 0.29\n",
      "Epoch [196/300], Step [400/938], d_loss: 0.6899, g_loss: 2.4616, D(x): 0.75, D(G(z)): 0.20\n",
      "Epoch [196/300], Step [600/938], d_loss: 1.0394, g_loss: 2.1629, D(x): 0.67, D(G(z)): 0.28\n",
      "Epoch [196/300], Step [800/938], d_loss: 0.9043, g_loss: 1.5379, D(x): 0.73, D(G(z)): 0.34\n",
      "Saving fake_images-0196.png\n",
      "Epoch [197/300], Step [200/938], d_loss: 0.7198, g_loss: 1.4774, D(x): 0.72, D(G(z)): 0.23\n",
      "Epoch [197/300], Step [400/938], d_loss: 0.9120, g_loss: 1.8282, D(x): 0.67, D(G(z)): 0.26\n",
      "Epoch [197/300], Step [600/938], d_loss: 0.9555, g_loss: 1.3549, D(x): 0.79, D(G(z)): 0.41\n",
      "Epoch [197/300], Step [800/938], d_loss: 0.9650, g_loss: 1.8174, D(x): 0.74, D(G(z)): 0.37\n",
      "Saving fake_images-0197.png\n",
      "Epoch [198/300], Step [200/938], d_loss: 0.9652, g_loss: 1.4073, D(x): 0.76, D(G(z)): 0.39\n",
      "Epoch [198/300], Step [400/938], d_loss: 0.9490, g_loss: 1.6054, D(x): 0.73, D(G(z)): 0.32\n",
      "Epoch [198/300], Step [600/938], d_loss: 0.8420, g_loss: 1.5707, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [198/300], Step [800/938], d_loss: 0.7701, g_loss: 1.7069, D(x): 0.69, D(G(z)): 0.23\n",
      "Saving fake_images-0198.png\n",
      "Epoch [199/300], Step [200/938], d_loss: 0.8614, g_loss: 1.4073, D(x): 0.68, D(G(z)): 0.26\n",
      "Epoch [199/300], Step [400/938], d_loss: 1.0142, g_loss: 1.7504, D(x): 0.59, D(G(z)): 0.25\n",
      "Epoch [199/300], Step [600/938], d_loss: 0.6089, g_loss: 2.0717, D(x): 0.83, D(G(z)): 0.27\n",
      "Epoch [199/300], Step [800/938], d_loss: 0.8254, g_loss: 1.5269, D(x): 0.78, D(G(z)): 0.33\n",
      "Saving fake_images-0199.png\n",
      "Epoch [200/300], Step [200/938], d_loss: 0.9843, g_loss: 1.8571, D(x): 0.63, D(G(z)): 0.24\n",
      "Epoch [200/300], Step [400/938], d_loss: 0.9446, g_loss: 2.0423, D(x): 0.73, D(G(z)): 0.35\n",
      "Epoch [200/300], Step [600/938], d_loss: 1.2254, g_loss: 1.1627, D(x): 0.54, D(G(z)): 0.28\n",
      "Epoch [200/300], Step [800/938], d_loss: 0.7068, g_loss: 1.3849, D(x): 0.77, D(G(z)): 0.28\n",
      "Saving fake_images-0200.png\n",
      "Epoch [201/300], Step [200/938], d_loss: 0.8563, g_loss: 1.5724, D(x): 0.74, D(G(z)): 0.33\n",
      "Epoch [201/300], Step [400/938], d_loss: 0.8110, g_loss: 1.6380, D(x): 0.69, D(G(z)): 0.24\n",
      "Epoch [201/300], Step [600/938], d_loss: 0.9582, g_loss: 1.7197, D(x): 0.75, D(G(z)): 0.36\n",
      "Epoch [201/300], Step [800/938], d_loss: 0.7595, g_loss: 1.6548, D(x): 0.77, D(G(z)): 0.29\n",
      "Saving fake_images-0201.png\n",
      "Epoch [202/300], Step [200/938], d_loss: 0.7932, g_loss: 1.5828, D(x): 0.71, D(G(z)): 0.27\n",
      "Epoch [202/300], Step [400/938], d_loss: 0.8438, g_loss: 1.5288, D(x): 0.70, D(G(z)): 0.28\n",
      "Epoch [202/300], Step [600/938], d_loss: 0.9598, g_loss: 1.0298, D(x): 0.74, D(G(z)): 0.35\n",
      "Epoch [202/300], Step [800/938], d_loss: 0.7457, g_loss: 1.7786, D(x): 0.79, D(G(z)): 0.31\n",
      "Saving fake_images-0202.png\n",
      "Epoch [203/300], Step [200/938], d_loss: 1.0793, g_loss: 1.5726, D(x): 0.68, D(G(z)): 0.35\n",
      "Epoch [203/300], Step [400/938], d_loss: 0.8335, g_loss: 1.2513, D(x): 0.70, D(G(z)): 0.28\n",
      "Epoch [203/300], Step [600/938], d_loss: 0.9599, g_loss: 1.4537, D(x): 0.66, D(G(z)): 0.30\n",
      "Epoch [203/300], Step [800/938], d_loss: 0.7259, g_loss: 1.9765, D(x): 0.72, D(G(z)): 0.22\n",
      "Saving fake_images-0203.png\n",
      "Epoch [204/300], Step [200/938], d_loss: 0.9393, g_loss: 1.5974, D(x): 0.69, D(G(z)): 0.29\n",
      "Epoch [204/300], Step [400/938], d_loss: 0.9968, g_loss: 1.8225, D(x): 0.67, D(G(z)): 0.28\n",
      "Epoch [204/300], Step [600/938], d_loss: 1.1261, g_loss: 1.6539, D(x): 0.68, D(G(z)): 0.34\n",
      "Epoch [204/300], Step [800/938], d_loss: 0.9294, g_loss: 1.1250, D(x): 0.80, D(G(z)): 0.36\n",
      "Saving fake_images-0204.png\n",
      "Epoch [205/300], Step [200/938], d_loss: 0.8600, g_loss: 1.5764, D(x): 0.73, D(G(z)): 0.32\n",
      "Epoch [205/300], Step [400/938], d_loss: 0.7943, g_loss: 1.5928, D(x): 0.71, D(G(z)): 0.26\n",
      "Epoch [205/300], Step [600/938], d_loss: 0.7599, g_loss: 1.6451, D(x): 0.76, D(G(z)): 0.31\n",
      "Epoch [205/300], Step [800/938], d_loss: 1.0533, g_loss: 2.0256, D(x): 0.62, D(G(z)): 0.25\n",
      "Saving fake_images-0205.png\n",
      "Epoch [206/300], Step [200/938], d_loss: 0.9697, g_loss: 1.4584, D(x): 0.64, D(G(z)): 0.27\n",
      "Epoch [206/300], Step [400/938], d_loss: 0.7943, g_loss: 2.0753, D(x): 0.75, D(G(z)): 0.25\n",
      "Epoch [206/300], Step [600/938], d_loss: 1.0040, g_loss: 1.2713, D(x): 0.64, D(G(z)): 0.27\n",
      "Epoch [206/300], Step [800/938], d_loss: 0.8989, g_loss: 1.6439, D(x): 0.70, D(G(z)): 0.30\n",
      "Saving fake_images-0206.png\n",
      "Epoch [207/300], Step [200/938], d_loss: 0.8607, g_loss: 1.3474, D(x): 0.75, D(G(z)): 0.34\n",
      "Epoch [207/300], Step [400/938], d_loss: 0.9358, g_loss: 1.8414, D(x): 0.70, D(G(z)): 0.30\n",
      "Epoch [207/300], Step [600/938], d_loss: 0.8488, g_loss: 1.5653, D(x): 0.69, D(G(z)): 0.28\n",
      "Epoch [207/300], Step [800/938], d_loss: 0.9411, g_loss: 1.5160, D(x): 0.83, D(G(z)): 0.41\n",
      "Saving fake_images-0207.png\n",
      "Epoch [208/300], Step [200/938], d_loss: 1.0784, g_loss: 1.9888, D(x): 0.71, D(G(z)): 0.37\n",
      "Epoch [208/300], Step [400/938], d_loss: 0.7520, g_loss: 1.7788, D(x): 0.76, D(G(z)): 0.28\n",
      "Epoch [208/300], Step [600/938], d_loss: 0.8633, g_loss: 1.6373, D(x): 0.72, D(G(z)): 0.32\n",
      "Epoch [208/300], Step [800/938], d_loss: 0.9589, g_loss: 1.8477, D(x): 0.66, D(G(z)): 0.28\n",
      "Saving fake_images-0208.png\n",
      "Epoch [209/300], Step [200/938], d_loss: 0.8460, g_loss: 1.7469, D(x): 0.72, D(G(z)): 0.28\n",
      "Epoch [209/300], Step [400/938], d_loss: 0.9323, g_loss: 1.4148, D(x): 0.67, D(G(z)): 0.29\n",
      "Epoch [209/300], Step [600/938], d_loss: 0.8682, g_loss: 1.3838, D(x): 0.69, D(G(z)): 0.28\n",
      "Epoch [209/300], Step [800/938], d_loss: 0.6529, g_loss: 1.6916, D(x): 0.78, D(G(z)): 0.26\n",
      "Saving fake_images-0209.png\n",
      "Epoch [210/300], Step [200/938], d_loss: 0.8837, g_loss: 1.4639, D(x): 0.67, D(G(z)): 0.26\n",
      "Epoch [210/300], Step [400/938], d_loss: 0.8477, g_loss: 2.1377, D(x): 0.71, D(G(z)): 0.24\n",
      "Epoch [210/300], Step [600/938], d_loss: 1.0245, g_loss: 1.7691, D(x): 0.64, D(G(z)): 0.26\n",
      "Epoch [210/300], Step [800/938], d_loss: 1.0021, g_loss: 0.9501, D(x): 0.81, D(G(z)): 0.47\n",
      "Saving fake_images-0210.png\n",
      "Epoch [211/300], Step [200/938], d_loss: 0.8273, g_loss: 2.0339, D(x): 0.73, D(G(z)): 0.27\n",
      "Epoch [211/300], Step [400/938], d_loss: 0.8172, g_loss: 1.5135, D(x): 0.75, D(G(z)): 0.28\n",
      "Epoch [211/300], Step [600/938], d_loss: 0.9441, g_loss: 1.8284, D(x): 0.70, D(G(z)): 0.32\n",
      "Epoch [211/300], Step [800/938], d_loss: 0.8011, g_loss: 1.3643, D(x): 0.83, D(G(z)): 0.35\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0211.png\n",
      "Epoch [212/300], Step [200/938], d_loss: 0.8354, g_loss: 1.9616, D(x): 0.66, D(G(z)): 0.22\n",
      "Epoch [212/300], Step [400/938], d_loss: 0.9608, g_loss: 1.4400, D(x): 0.78, D(G(z)): 0.41\n",
      "Epoch [212/300], Step [600/938], d_loss: 0.8562, g_loss: 1.8653, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [212/300], Step [800/938], d_loss: 0.7292, g_loss: 2.0768, D(x): 0.80, D(G(z)): 0.29\n",
      "Saving fake_images-0212.png\n",
      "Epoch [213/300], Step [200/938], d_loss: 0.9020, g_loss: 1.3852, D(x): 0.77, D(G(z)): 0.35\n",
      "Epoch [213/300], Step [400/938], d_loss: 0.8970, g_loss: 1.9393, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [213/300], Step [600/938], d_loss: 0.8478, g_loss: 1.6448, D(x): 0.66, D(G(z)): 0.21\n",
      "Epoch [213/300], Step [800/938], d_loss: 1.0252, g_loss: 1.1679, D(x): 0.59, D(G(z)): 0.23\n",
      "Saving fake_images-0213.png\n",
      "Epoch [214/300], Step [200/938], d_loss: 0.8269, g_loss: 1.4355, D(x): 0.80, D(G(z)): 0.37\n",
      "Epoch [214/300], Step [400/938], d_loss: 0.8706, g_loss: 1.6993, D(x): 0.72, D(G(z)): 0.30\n",
      "Epoch [214/300], Step [600/938], d_loss: 0.8707, g_loss: 1.8450, D(x): 0.69, D(G(z)): 0.26\n",
      "Epoch [214/300], Step [800/938], d_loss: 0.8150, g_loss: 1.6078, D(x): 0.77, D(G(z)): 0.33\n",
      "Saving fake_images-0214.png\n",
      "Epoch [215/300], Step [200/938], d_loss: 0.8499, g_loss: 1.6140, D(x): 0.76, D(G(z)): 0.28\n",
      "Epoch [215/300], Step [400/938], d_loss: 0.8755, g_loss: 1.6033, D(x): 0.78, D(G(z)): 0.31\n",
      "Epoch [215/300], Step [600/938], d_loss: 0.9608, g_loss: 1.5332, D(x): 0.69, D(G(z)): 0.31\n",
      "Epoch [215/300], Step [800/938], d_loss: 1.0404, g_loss: 1.8043, D(x): 0.62, D(G(z)): 0.27\n",
      "Saving fake_images-0215.png\n",
      "Epoch [216/300], Step [200/938], d_loss: 1.1219, g_loss: 2.0179, D(x): 0.75, D(G(z)): 0.43\n",
      "Epoch [216/300], Step [400/938], d_loss: 0.7477, g_loss: 1.8533, D(x): 0.78, D(G(z)): 0.28\n",
      "Epoch [216/300], Step [600/938], d_loss: 0.8872, g_loss: 1.5838, D(x): 0.63, D(G(z)): 0.22\n",
      "Epoch [216/300], Step [800/938], d_loss: 0.9418, g_loss: 1.4923, D(x): 0.69, D(G(z)): 0.30\n",
      "Saving fake_images-0216.png\n",
      "Epoch [217/300], Step [200/938], d_loss: 0.8338, g_loss: 1.7616, D(x): 0.67, D(G(z)): 0.22\n",
      "Epoch [217/300], Step [400/938], d_loss: 0.7908, g_loss: 1.9364, D(x): 0.75, D(G(z)): 0.28\n",
      "Epoch [217/300], Step [600/938], d_loss: 0.8229, g_loss: 1.7263, D(x): 0.69, D(G(z)): 0.26\n",
      "Epoch [217/300], Step [800/938], d_loss: 0.8561, g_loss: 1.5129, D(x): 0.75, D(G(z)): 0.30\n",
      "Saving fake_images-0217.png\n",
      "Epoch [218/300], Step [200/938], d_loss: 0.8755, g_loss: 1.7191, D(x): 0.83, D(G(z)): 0.38\n",
      "Epoch [218/300], Step [400/938], d_loss: 0.8590, g_loss: 1.5015, D(x): 0.66, D(G(z)): 0.20\n",
      "Epoch [218/300], Step [600/938], d_loss: 0.7949, g_loss: 1.8010, D(x): 0.70, D(G(z)): 0.23\n",
      "Epoch [218/300], Step [800/938], d_loss: 0.8722, g_loss: 1.7215, D(x): 0.65, D(G(z)): 0.22\n",
      "Saving fake_images-0218.png\n",
      "Epoch [219/300], Step [200/938], d_loss: 0.8550, g_loss: 1.5387, D(x): 0.69, D(G(z)): 0.23\n",
      "Epoch [219/300], Step [400/938], d_loss: 0.8642, g_loss: 1.6997, D(x): 0.70, D(G(z)): 0.30\n",
      "Epoch [219/300], Step [600/938], d_loss: 0.9346, g_loss: 1.5352, D(x): 0.71, D(G(z)): 0.32\n",
      "Epoch [219/300], Step [800/938], d_loss: 1.0032, g_loss: 1.7481, D(x): 0.75, D(G(z)): 0.40\n",
      "Saving fake_images-0219.png\n",
      "Epoch [220/300], Step [200/938], d_loss: 0.9913, g_loss: 1.7790, D(x): 0.62, D(G(z)): 0.24\n",
      "Epoch [220/300], Step [400/938], d_loss: 0.9308, g_loss: 1.4254, D(x): 0.77, D(G(z)): 0.37\n",
      "Epoch [220/300], Step [600/938], d_loss: 0.6872, g_loss: 1.3849, D(x): 0.75, D(G(z)): 0.24\n",
      "Epoch [220/300], Step [800/938], d_loss: 0.8748, g_loss: 1.8249, D(x): 0.72, D(G(z)): 0.32\n",
      "Saving fake_images-0220.png\n",
      "Epoch [221/300], Step [200/938], d_loss: 0.9822, g_loss: 1.6605, D(x): 0.64, D(G(z)): 0.27\n",
      "Epoch [221/300], Step [400/938], d_loss: 0.7309, g_loss: 1.8254, D(x): 0.77, D(G(z)): 0.28\n",
      "Epoch [221/300], Step [600/938], d_loss: 0.7444, g_loss: 1.6402, D(x): 0.70, D(G(z)): 0.23\n",
      "Epoch [221/300], Step [800/938], d_loss: 1.1551, g_loss: 1.2834, D(x): 0.77, D(G(z)): 0.47\n",
      "Saving fake_images-0221.png\n",
      "Epoch [222/300], Step [200/938], d_loss: 0.7993, g_loss: 1.5782, D(x): 0.72, D(G(z)): 0.29\n",
      "Epoch [222/300], Step [400/938], d_loss: 0.8360, g_loss: 1.1551, D(x): 0.75, D(G(z)): 0.34\n",
      "Epoch [222/300], Step [600/938], d_loss: 0.7318, g_loss: 1.5379, D(x): 0.74, D(G(z)): 0.29\n",
      "Epoch [222/300], Step [800/938], d_loss: 0.8286, g_loss: 1.7747, D(x): 0.70, D(G(z)): 0.27\n",
      "Saving fake_images-0222.png\n",
      "Epoch [223/300], Step [200/938], d_loss: 0.9104, g_loss: 1.7739, D(x): 0.76, D(G(z)): 0.33\n",
      "Epoch [223/300], Step [400/938], d_loss: 0.9412, g_loss: 1.5453, D(x): 0.68, D(G(z)): 0.28\n",
      "Epoch [223/300], Step [600/938], d_loss: 0.8422, g_loss: 1.9049, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [223/300], Step [800/938], d_loss: 0.8155, g_loss: 1.2938, D(x): 0.71, D(G(z)): 0.24\n",
      "Saving fake_images-0223.png\n",
      "Epoch [224/300], Step [200/938], d_loss: 0.7885, g_loss: 2.0726, D(x): 0.76, D(G(z)): 0.28\n",
      "Epoch [224/300], Step [400/938], d_loss: 0.9649, g_loss: 1.6288, D(x): 0.68, D(G(z)): 0.29\n",
      "Epoch [224/300], Step [600/938], d_loss: 1.0474, g_loss: 1.4078, D(x): 0.75, D(G(z)): 0.36\n",
      "Epoch [224/300], Step [800/938], d_loss: 0.9029, g_loss: 2.1972, D(x): 0.67, D(G(z)): 0.27\n",
      "Saving fake_images-0224.png\n",
      "Epoch [225/300], Step [200/938], d_loss: 0.8903, g_loss: 1.6425, D(x): 0.72, D(G(z)): 0.30\n",
      "Epoch [225/300], Step [400/938], d_loss: 0.9139, g_loss: 1.5760, D(x): 0.68, D(G(z)): 0.29\n",
      "Epoch [225/300], Step [600/938], d_loss: 0.8325, g_loss: 1.7171, D(x): 0.70, D(G(z)): 0.26\n",
      "Epoch [225/300], Step [800/938], d_loss: 0.7558, g_loss: 1.8095, D(x): 0.76, D(G(z)): 0.30\n",
      "Saving fake_images-0225.png\n",
      "Epoch [226/300], Step [200/938], d_loss: 1.0515, g_loss: 1.4285, D(x): 0.69, D(G(z)): 0.35\n",
      "Epoch [226/300], Step [400/938], d_loss: 0.7451, g_loss: 1.7387, D(x): 0.74, D(G(z)): 0.24\n",
      "Epoch [226/300], Step [600/938], d_loss: 0.9880, g_loss: 1.4199, D(x): 0.76, D(G(z)): 0.38\n",
      "Epoch [226/300], Step [800/938], d_loss: 0.7480, g_loss: 2.2251, D(x): 0.72, D(G(z)): 0.22\n",
      "Saving fake_images-0226.png\n",
      "Epoch [227/300], Step [200/938], d_loss: 1.0439, g_loss: 1.7587, D(x): 0.69, D(G(z)): 0.35\n",
      "Epoch [227/300], Step [400/938], d_loss: 0.8757, g_loss: 1.3872, D(x): 0.76, D(G(z)): 0.35\n",
      "Epoch [227/300], Step [600/938], d_loss: 0.9677, g_loss: 1.4928, D(x): 0.71, D(G(z)): 0.33\n",
      "Epoch [227/300], Step [800/938], d_loss: 1.0258, g_loss: 1.8015, D(x): 0.70, D(G(z)): 0.36\n",
      "Saving fake_images-0227.png\n",
      "Epoch [228/300], Step [200/938], d_loss: 0.8046, g_loss: 1.8262, D(x): 0.68, D(G(z)): 0.24\n",
      "Epoch [228/300], Step [400/938], d_loss: 0.7194, g_loss: 1.5946, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [228/300], Step [600/938], d_loss: 0.9940, g_loss: 1.1700, D(x): 0.76, D(G(z)): 0.40\n",
      "Epoch [228/300], Step [800/938], d_loss: 0.7367, g_loss: 1.4780, D(x): 0.71, D(G(z)): 0.23\n",
      "Saving fake_images-0228.png\n",
      "Epoch [229/300], Step [200/938], d_loss: 0.7566, g_loss: 1.8442, D(x): 0.74, D(G(z)): 0.25\n",
      "Epoch [229/300], Step [400/938], d_loss: 0.9534, g_loss: 1.6261, D(x): 0.67, D(G(z)): 0.25\n",
      "Epoch [229/300], Step [600/938], d_loss: 0.9616, g_loss: 1.1733, D(x): 0.75, D(G(z)): 0.35\n",
      "Epoch [229/300], Step [800/938], d_loss: 1.0260, g_loss: 1.6605, D(x): 0.63, D(G(z)): 0.27\n",
      "Saving fake_images-0229.png\n",
      "Epoch [230/300], Step [200/938], d_loss: 0.8087, g_loss: 1.4875, D(x): 0.78, D(G(z)): 0.33\n",
      "Epoch [230/300], Step [400/938], d_loss: 0.9684, g_loss: 1.4902, D(x): 0.73, D(G(z)): 0.35\n",
      "Epoch [230/300], Step [600/938], d_loss: 0.7709, g_loss: 1.9131, D(x): 0.68, D(G(z)): 0.24\n",
      "Epoch [230/300], Step [800/938], d_loss: 0.8217, g_loss: 1.7654, D(x): 0.77, D(G(z)): 0.30\n",
      "Saving fake_images-0230.png\n",
      "Epoch [231/300], Step [200/938], d_loss: 0.7291, g_loss: 1.3615, D(x): 0.71, D(G(z)): 0.17\n",
      "Epoch [231/300], Step [400/938], d_loss: 0.8754, g_loss: 1.4975, D(x): 0.72, D(G(z)): 0.31\n",
      "Epoch [231/300], Step [600/938], d_loss: 0.9674, g_loss: 1.6618, D(x): 0.68, D(G(z)): 0.33\n",
      "Epoch [231/300], Step [800/938], d_loss: 0.9687, g_loss: 1.7317, D(x): 0.67, D(G(z)): 0.30\n",
      "Saving fake_images-0231.png\n",
      "Epoch [232/300], Step [200/938], d_loss: 0.9336, g_loss: 1.4005, D(x): 0.70, D(G(z)): 0.30\n",
      "Epoch [232/300], Step [400/938], d_loss: 0.9966, g_loss: 1.6578, D(x): 0.77, D(G(z)): 0.38\n",
      "Epoch [232/300], Step [600/938], d_loss: 0.9032, g_loss: 1.6893, D(x): 0.75, D(G(z)): 0.34\n",
      "Epoch [232/300], Step [800/938], d_loss: 0.8581, g_loss: 1.6181, D(x): 0.71, D(G(z)): 0.25\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0232.png\n",
      "Epoch [233/300], Step [200/938], d_loss: 0.8994, g_loss: 1.8417, D(x): 0.74, D(G(z)): 0.33\n",
      "Epoch [233/300], Step [400/938], d_loss: 0.8523, g_loss: 1.4843, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [233/300], Step [600/938], d_loss: 0.8324, g_loss: 1.4476, D(x): 0.75, D(G(z)): 0.32\n",
      "Epoch [233/300], Step [800/938], d_loss: 0.8005, g_loss: 1.7241, D(x): 0.73, D(G(z)): 0.30\n",
      "Saving fake_images-0233.png\n",
      "Epoch [234/300], Step [200/938], d_loss: 0.8423, g_loss: 1.4908, D(x): 0.75, D(G(z)): 0.31\n",
      "Epoch [234/300], Step [400/938], d_loss: 1.1546, g_loss: 1.6143, D(x): 0.70, D(G(z)): 0.32\n",
      "Epoch [234/300], Step [600/938], d_loss: 1.0359, g_loss: 1.6932, D(x): 0.69, D(G(z)): 0.31\n",
      "Epoch [234/300], Step [800/938], d_loss: 0.8415, g_loss: 1.6289, D(x): 0.67, D(G(z)): 0.19\n",
      "Saving fake_images-0234.png\n",
      "Epoch [235/300], Step [200/938], d_loss: 1.0775, g_loss: 1.2762, D(x): 0.66, D(G(z)): 0.34\n",
      "Epoch [235/300], Step [400/938], d_loss: 1.0851, g_loss: 1.4060, D(x): 0.71, D(G(z)): 0.32\n",
      "Epoch [235/300], Step [600/938], d_loss: 1.1536, g_loss: 1.6817, D(x): 0.62, D(G(z)): 0.32\n",
      "Epoch [235/300], Step [800/938], d_loss: 0.7969, g_loss: 2.2244, D(x): 0.65, D(G(z)): 0.19\n",
      "Saving fake_images-0235.png\n",
      "Epoch [236/300], Step [200/938], d_loss: 0.8504, g_loss: 1.6443, D(x): 0.71, D(G(z)): 0.30\n",
      "Epoch [236/300], Step [400/938], d_loss: 0.9784, g_loss: 1.4349, D(x): 0.60, D(G(z)): 0.22\n",
      "Epoch [236/300], Step [600/938], d_loss: 1.0030, g_loss: 1.7052, D(x): 0.76, D(G(z)): 0.35\n",
      "Epoch [236/300], Step [800/938], d_loss: 0.8475, g_loss: 1.7451, D(x): 0.68, D(G(z)): 0.24\n",
      "Saving fake_images-0236.png\n",
      "Epoch [237/300], Step [200/938], d_loss: 0.9372, g_loss: 1.4582, D(x): 0.73, D(G(z)): 0.33\n",
      "Epoch [237/300], Step [400/938], d_loss: 0.9069, g_loss: 1.7027, D(x): 0.74, D(G(z)): 0.34\n",
      "Epoch [237/300], Step [600/938], d_loss: 1.1751, g_loss: 1.4309, D(x): 0.63, D(G(z)): 0.35\n",
      "Epoch [237/300], Step [800/938], d_loss: 0.7268, g_loss: 1.4865, D(x): 0.76, D(G(z)): 0.27\n",
      "Saving fake_images-0237.png\n",
      "Epoch [238/300], Step [200/938], d_loss: 0.8119, g_loss: 1.6509, D(x): 0.74, D(G(z)): 0.28\n",
      "Epoch [238/300], Step [400/938], d_loss: 1.0431, g_loss: 1.2771, D(x): 0.77, D(G(z)): 0.42\n",
      "Epoch [238/300], Step [600/938], d_loss: 0.8947, g_loss: 1.6014, D(x): 0.75, D(G(z)): 0.34\n",
      "Epoch [238/300], Step [800/938], d_loss: 0.9057, g_loss: 2.2224, D(x): 0.62, D(G(z)): 0.18\n",
      "Saving fake_images-0238.png\n",
      "Epoch [239/300], Step [200/938], d_loss: 0.9989, g_loss: 1.3527, D(x): 0.81, D(G(z)): 0.40\n",
      "Epoch [239/300], Step [400/938], d_loss: 0.9979, g_loss: 1.3272, D(x): 0.69, D(G(z)): 0.31\n",
      "Epoch [239/300], Step [600/938], d_loss: 1.0155, g_loss: 1.7927, D(x): 0.61, D(G(z)): 0.23\n",
      "Epoch [239/300], Step [800/938], d_loss: 0.8030, g_loss: 1.6261, D(x): 0.71, D(G(z)): 0.25\n",
      "Saving fake_images-0239.png\n",
      "Epoch [240/300], Step [200/938], d_loss: 1.0079, g_loss: 1.7967, D(x): 0.69, D(G(z)): 0.31\n",
      "Epoch [240/300], Step [400/938], d_loss: 0.8138, g_loss: 1.7844, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [240/300], Step [600/938], d_loss: 0.6891, g_loss: 1.5320, D(x): 0.76, D(G(z)): 0.22\n",
      "Epoch [240/300], Step [800/938], d_loss: 0.8625, g_loss: 1.6033, D(x): 0.74, D(G(z)): 0.31\n",
      "Saving fake_images-0240.png\n",
      "Epoch [241/300], Step [200/938], d_loss: 0.8957, g_loss: 1.7050, D(x): 0.75, D(G(z)): 0.32\n",
      "Epoch [241/300], Step [400/938], d_loss: 0.8205, g_loss: 1.6522, D(x): 0.75, D(G(z)): 0.34\n",
      "Epoch [241/300], Step [600/938], d_loss: 1.0267, g_loss: 1.3359, D(x): 0.81, D(G(z)): 0.43\n",
      "Epoch [241/300], Step [800/938], d_loss: 0.7418, g_loss: 1.6104, D(x): 0.78, D(G(z)): 0.30\n",
      "Saving fake_images-0241.png\n",
      "Epoch [242/300], Step [200/938], d_loss: 0.8800, g_loss: 1.8873, D(x): 0.68, D(G(z)): 0.23\n",
      "Epoch [242/300], Step [400/938], d_loss: 0.9052, g_loss: 1.7207, D(x): 0.61, D(G(z)): 0.21\n",
      "Epoch [242/300], Step [600/938], d_loss: 0.8503, g_loss: 1.5002, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [242/300], Step [800/938], d_loss: 0.9245, g_loss: 1.3727, D(x): 0.70, D(G(z)): 0.32\n",
      "Saving fake_images-0242.png\n",
      "Epoch [243/300], Step [200/938], d_loss: 0.9442, g_loss: 1.6432, D(x): 0.83, D(G(z)): 0.41\n",
      "Epoch [243/300], Step [400/938], d_loss: 0.8935, g_loss: 1.7518, D(x): 0.73, D(G(z)): 0.30\n",
      "Epoch [243/300], Step [600/938], d_loss: 0.8422, g_loss: 1.4774, D(x): 0.77, D(G(z)): 0.33\n",
      "Epoch [243/300], Step [800/938], d_loss: 0.8598, g_loss: 1.8607, D(x): 0.73, D(G(z)): 0.29\n",
      "Saving fake_images-0243.png\n",
      "Epoch [244/300], Step [200/938], d_loss: 0.9226, g_loss: 1.4535, D(x): 0.67, D(G(z)): 0.28\n",
      "Epoch [244/300], Step [400/938], d_loss: 0.8880, g_loss: 1.7823, D(x): 0.69, D(G(z)): 0.28\n",
      "Epoch [244/300], Step [600/938], d_loss: 0.8527, g_loss: 1.6503, D(x): 0.71, D(G(z)): 0.27\n",
      "Epoch [244/300], Step [800/938], d_loss: 0.8041, g_loss: 1.8746, D(x): 0.74, D(G(z)): 0.25\n",
      "Saving fake_images-0244.png\n",
      "Epoch [245/300], Step [200/938], d_loss: 0.9984, g_loss: 1.8560, D(x): 0.65, D(G(z)): 0.28\n",
      "Epoch [245/300], Step [400/938], d_loss: 0.8310, g_loss: 1.2749, D(x): 0.78, D(G(z)): 0.34\n",
      "Epoch [245/300], Step [600/938], d_loss: 0.8445, g_loss: 1.5972, D(x): 0.70, D(G(z)): 0.23\n",
      "Epoch [245/300], Step [800/938], d_loss: 0.8916, g_loss: 1.5654, D(x): 0.68, D(G(z)): 0.27\n",
      "Saving fake_images-0245.png\n",
      "Epoch [246/300], Step [200/938], d_loss: 0.8927, g_loss: 1.5346, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [246/300], Step [400/938], d_loss: 0.8418, g_loss: 1.7854, D(x): 0.68, D(G(z)): 0.23\n",
      "Epoch [246/300], Step [600/938], d_loss: 1.0264, g_loss: 1.9247, D(x): 0.60, D(G(z)): 0.23\n",
      "Epoch [246/300], Step [800/938], d_loss: 0.9052, g_loss: 1.8543, D(x): 0.68, D(G(z)): 0.30\n",
      "Saving fake_images-0246.png\n",
      "Epoch [247/300], Step [200/938], d_loss: 0.6613, g_loss: 1.6363, D(x): 0.78, D(G(z)): 0.25\n",
      "Epoch [247/300], Step [400/938], d_loss: 0.9074, g_loss: 1.6029, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [247/300], Step [600/938], d_loss: 0.8163, g_loss: 2.1758, D(x): 0.75, D(G(z)): 0.29\n",
      "Epoch [247/300], Step [800/938], d_loss: 0.9725, g_loss: 1.6333, D(x): 0.70, D(G(z)): 0.29\n",
      "Saving fake_images-0247.png\n",
      "Epoch [248/300], Step [200/938], d_loss: 0.9795, g_loss: 2.1364, D(x): 0.64, D(G(z)): 0.28\n",
      "Epoch [248/300], Step [400/938], d_loss: 0.8173, g_loss: 2.0840, D(x): 0.70, D(G(z)): 0.24\n",
      "Epoch [248/300], Step [600/938], d_loss: 0.8945, g_loss: 1.6436, D(x): 0.63, D(G(z)): 0.22\n",
      "Epoch [248/300], Step [800/938], d_loss: 0.8785, g_loss: 1.5569, D(x): 0.73, D(G(z)): 0.31\n",
      "Saving fake_images-0248.png\n",
      "Epoch [249/300], Step [200/938], d_loss: 0.8712, g_loss: 1.3924, D(x): 0.77, D(G(z)): 0.36\n",
      "Epoch [249/300], Step [400/938], d_loss: 0.8313, g_loss: 1.7868, D(x): 0.70, D(G(z)): 0.22\n",
      "Epoch [249/300], Step [600/938], d_loss: 1.0029, g_loss: 1.7024, D(x): 0.64, D(G(z)): 0.28\n",
      "Epoch [249/300], Step [800/938], d_loss: 0.8735, g_loss: 1.2839, D(x): 0.83, D(G(z)): 0.37\n",
      "Saving fake_images-0249.png\n",
      "Epoch [250/300], Step [200/938], d_loss: 0.8800, g_loss: 2.2208, D(x): 0.65, D(G(z)): 0.21\n",
      "Epoch [250/300], Step [400/938], d_loss: 0.9218, g_loss: 1.6452, D(x): 0.72, D(G(z)): 0.31\n",
      "Epoch [250/300], Step [600/938], d_loss: 0.6714, g_loss: 1.9107, D(x): 0.72, D(G(z)): 0.19\n",
      "Epoch [250/300], Step [800/938], d_loss: 0.9993, g_loss: 1.5721, D(x): 0.79, D(G(z)): 0.39\n",
      "Saving fake_images-0250.png\n",
      "Epoch [251/300], Step [200/938], d_loss: 0.7122, g_loss: 2.0143, D(x): 0.70, D(G(z)): 0.22\n",
      "Epoch [251/300], Step [400/938], d_loss: 0.9069, g_loss: 1.4329, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [251/300], Step [600/938], d_loss: 0.8928, g_loss: 1.6834, D(x): 0.76, D(G(z)): 0.33\n",
      "Epoch [251/300], Step [800/938], d_loss: 1.1446, g_loss: 1.9023, D(x): 0.68, D(G(z)): 0.32\n",
      "Saving fake_images-0251.png\n",
      "Epoch [252/300], Step [200/938], d_loss: 0.8503, g_loss: 1.8247, D(x): 0.68, D(G(z)): 0.22\n",
      "Epoch [252/300], Step [400/938], d_loss: 0.8374, g_loss: 1.6649, D(x): 0.67, D(G(z)): 0.23\n",
      "Epoch [252/300], Step [600/938], d_loss: 0.8941, g_loss: 1.6064, D(x): 0.72, D(G(z)): 0.30\n",
      "Epoch [252/300], Step [800/938], d_loss: 1.1153, g_loss: 1.7170, D(x): 0.65, D(G(z)): 0.28\n",
      "Saving fake_images-0252.png\n",
      "Epoch [253/300], Step [200/938], d_loss: 0.9140, g_loss: 1.6266, D(x): 0.68, D(G(z)): 0.28\n",
      "Epoch [253/300], Step [400/938], d_loss: 0.7096, g_loss: 1.4714, D(x): 0.76, D(G(z)): 0.24\n",
      "Epoch [253/300], Step [600/938], d_loss: 0.7826, g_loss: 1.5413, D(x): 0.69, D(G(z)): 0.24\n",
      "Epoch [253/300], Step [800/938], d_loss: 0.8638, g_loss: 1.5232, D(x): 0.69, D(G(z)): 0.25\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0253.png\n",
      "Epoch [254/300], Step [200/938], d_loss: 0.8744, g_loss: 1.2604, D(x): 0.79, D(G(z)): 0.39\n",
      "Epoch [254/300], Step [400/938], d_loss: 0.9539, g_loss: 1.6799, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [254/300], Step [600/938], d_loss: 0.8911, g_loss: 1.4928, D(x): 0.66, D(G(z)): 0.26\n",
      "Epoch [254/300], Step [800/938], d_loss: 0.6986, g_loss: 1.7376, D(x): 0.77, D(G(z)): 0.23\n",
      "Saving fake_images-0254.png\n",
      "Epoch [255/300], Step [200/938], d_loss: 0.8670, g_loss: 1.7212, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [255/300], Step [400/938], d_loss: 0.9641, g_loss: 1.5423, D(x): 0.66, D(G(z)): 0.26\n",
      "Epoch [255/300], Step [600/938], d_loss: 0.8888, g_loss: 1.3685, D(x): 0.74, D(G(z)): 0.30\n",
      "Epoch [255/300], Step [800/938], d_loss: 0.8528, g_loss: 1.8132, D(x): 0.73, D(G(z)): 0.29\n",
      "Saving fake_images-0255.png\n",
      "Epoch [256/300], Step [200/938], d_loss: 0.7748, g_loss: 1.4464, D(x): 0.75, D(G(z)): 0.31\n",
      "Epoch [256/300], Step [400/938], d_loss: 0.6681, g_loss: 1.7845, D(x): 0.73, D(G(z)): 0.20\n",
      "Epoch [256/300], Step [600/938], d_loss: 0.8287, g_loss: 1.3166, D(x): 0.80, D(G(z)): 0.34\n",
      "Epoch [256/300], Step [800/938], d_loss: 0.9711, g_loss: 1.5233, D(x): 0.79, D(G(z)): 0.38\n",
      "Saving fake_images-0256.png\n",
      "Epoch [257/300], Step [200/938], d_loss: 0.9096, g_loss: 1.3830, D(x): 0.74, D(G(z)): 0.33\n",
      "Epoch [257/300], Step [400/938], d_loss: 0.8448, g_loss: 1.6366, D(x): 0.70, D(G(z)): 0.26\n",
      "Epoch [257/300], Step [600/938], d_loss: 0.8483, g_loss: 1.7471, D(x): 0.71, D(G(z)): 0.27\n",
      "Epoch [257/300], Step [800/938], d_loss: 0.8192, g_loss: 1.5270, D(x): 0.71, D(G(z)): 0.28\n",
      "Saving fake_images-0257.png\n",
      "Epoch [258/300], Step [200/938], d_loss: 1.1561, g_loss: 1.9265, D(x): 0.63, D(G(z)): 0.29\n",
      "Epoch [258/300], Step [400/938], d_loss: 0.8076, g_loss: 1.6815, D(x): 0.76, D(G(z)): 0.30\n",
      "Epoch [258/300], Step [600/938], d_loss: 0.8919, g_loss: 1.5777, D(x): 0.71, D(G(z)): 0.31\n",
      "Epoch [258/300], Step [800/938], d_loss: 0.9914, g_loss: 1.4465, D(x): 0.72, D(G(z)): 0.33\n",
      "Saving fake_images-0258.png\n",
      "Epoch [259/300], Step [200/938], d_loss: 0.7535, g_loss: 1.5522, D(x): 0.74, D(G(z)): 0.26\n",
      "Epoch [259/300], Step [400/938], d_loss: 0.9999, g_loss: 1.8744, D(x): 0.59, D(G(z)): 0.21\n",
      "Epoch [259/300], Step [600/938], d_loss: 0.8824, g_loss: 1.7362, D(x): 0.74, D(G(z)): 0.32\n",
      "Epoch [259/300], Step [800/938], d_loss: 1.0967, g_loss: 1.8136, D(x): 0.67, D(G(z)): 0.32\n",
      "Saving fake_images-0259.png\n",
      "Epoch [260/300], Step [200/938], d_loss: 0.7796, g_loss: 1.7570, D(x): 0.77, D(G(z)): 0.29\n",
      "Epoch [260/300], Step [400/938], d_loss: 0.8340, g_loss: 1.5961, D(x): 0.75, D(G(z)): 0.30\n",
      "Epoch [260/300], Step [600/938], d_loss: 1.0510, g_loss: 1.7830, D(x): 0.65, D(G(z)): 0.26\n",
      "Epoch [260/300], Step [800/938], d_loss: 1.1140, g_loss: 1.7154, D(x): 0.64, D(G(z)): 0.28\n",
      "Saving fake_images-0260.png\n",
      "Epoch [261/300], Step [200/938], d_loss: 0.9063, g_loss: 1.2644, D(x): 0.82, D(G(z)): 0.38\n",
      "Epoch [261/300], Step [400/938], d_loss: 0.8123, g_loss: 1.7074, D(x): 0.73, D(G(z)): 0.26\n",
      "Epoch [261/300], Step [600/938], d_loss: 0.7619, g_loss: 1.6782, D(x): 0.77, D(G(z)): 0.29\n",
      "Epoch [261/300], Step [800/938], d_loss: 1.1463, g_loss: 1.7689, D(x): 0.56, D(G(z)): 0.25\n",
      "Saving fake_images-0261.png\n",
      "Epoch [262/300], Step [200/938], d_loss: 0.9604, g_loss: 1.3370, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [262/300], Step [400/938], d_loss: 0.7916, g_loss: 1.3518, D(x): 0.67, D(G(z)): 0.20\n",
      "Epoch [262/300], Step [600/938], d_loss: 0.9853, g_loss: 1.8439, D(x): 0.73, D(G(z)): 0.32\n",
      "Epoch [262/300], Step [800/938], d_loss: 0.7621, g_loss: 1.5308, D(x): 0.76, D(G(z)): 0.29\n",
      "Saving fake_images-0262.png\n",
      "Epoch [263/300], Step [200/938], d_loss: 0.7431, g_loss: 1.9022, D(x): 0.75, D(G(z)): 0.26\n",
      "Epoch [263/300], Step [400/938], d_loss: 1.0499, g_loss: 2.0910, D(x): 0.64, D(G(z)): 0.26\n",
      "Epoch [263/300], Step [600/938], d_loss: 0.9165, g_loss: 1.5346, D(x): 0.69, D(G(z)): 0.29\n",
      "Epoch [263/300], Step [800/938], d_loss: 0.8690, g_loss: 1.6705, D(x): 0.74, D(G(z)): 0.33\n",
      "Saving fake_images-0263.png\n",
      "Epoch [264/300], Step [200/938], d_loss: 0.9542, g_loss: 1.5730, D(x): 0.68, D(G(z)): 0.27\n",
      "Epoch [264/300], Step [400/938], d_loss: 0.8960, g_loss: 2.0072, D(x): 0.65, D(G(z)): 0.21\n",
      "Epoch [264/300], Step [600/938], d_loss: 0.8715, g_loss: 1.6388, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [264/300], Step [800/938], d_loss: 0.7881, g_loss: 1.8715, D(x): 0.81, D(G(z)): 0.34\n",
      "Saving fake_images-0264.png\n",
      "Epoch [265/300], Step [200/938], d_loss: 0.9343, g_loss: 1.3496, D(x): 0.73, D(G(z)): 0.33\n",
      "Epoch [265/300], Step [400/938], d_loss: 0.7750, g_loss: 1.9055, D(x): 0.74, D(G(z)): 0.23\n",
      "Epoch [265/300], Step [600/938], d_loss: 0.8278, g_loss: 1.7495, D(x): 0.74, D(G(z)): 0.27\n",
      "Epoch [265/300], Step [800/938], d_loss: 0.9097, g_loss: 1.5820, D(x): 0.70, D(G(z)): 0.27\n",
      "Saving fake_images-0265.png\n",
      "Epoch [266/300], Step [200/938], d_loss: 0.8958, g_loss: 1.4990, D(x): 0.72, D(G(z)): 0.30\n",
      "Epoch [266/300], Step [400/938], d_loss: 0.8235, g_loss: 2.0597, D(x): 0.68, D(G(z)): 0.18\n",
      "Epoch [266/300], Step [600/938], d_loss: 0.8227, g_loss: 1.7118, D(x): 0.68, D(G(z)): 0.20\n",
      "Epoch [266/300], Step [800/938], d_loss: 0.9314, g_loss: 1.5930, D(x): 0.68, D(G(z)): 0.25\n",
      "Saving fake_images-0266.png\n",
      "Epoch [267/300], Step [200/938], d_loss: 0.7556, g_loss: 1.4573, D(x): 0.72, D(G(z)): 0.27\n",
      "Epoch [267/300], Step [400/938], d_loss: 0.8042, g_loss: 1.7373, D(x): 0.77, D(G(z)): 0.31\n",
      "Epoch [267/300], Step [600/938], d_loss: 0.8325, g_loss: 2.0127, D(x): 0.69, D(G(z)): 0.25\n",
      "Epoch [267/300], Step [800/938], d_loss: 0.8557, g_loss: 2.0338, D(x): 0.69, D(G(z)): 0.27\n",
      "Saving fake_images-0267.png\n",
      "Epoch [268/300], Step [200/938], d_loss: 0.8830, g_loss: 1.7855, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [268/300], Step [400/938], d_loss: 0.8326, g_loss: 1.7318, D(x): 0.80, D(G(z)): 0.28\n",
      "Epoch [268/300], Step [600/938], d_loss: 1.1200, g_loss: 1.7328, D(x): 0.73, D(G(z)): 0.42\n",
      "Epoch [268/300], Step [800/938], d_loss: 0.8932, g_loss: 1.6575, D(x): 0.67, D(G(z)): 0.22\n",
      "Saving fake_images-0268.png\n",
      "Epoch [269/300], Step [200/938], d_loss: 0.8063, g_loss: 1.5594, D(x): 0.73, D(G(z)): 0.24\n",
      "Epoch [269/300], Step [400/938], d_loss: 0.7897, g_loss: 1.4509, D(x): 0.78, D(G(z)): 0.32\n",
      "Epoch [269/300], Step [600/938], d_loss: 0.7389, g_loss: 2.1337, D(x): 0.74, D(G(z)): 0.24\n",
      "Epoch [269/300], Step [800/938], d_loss: 0.9963, g_loss: 1.9425, D(x): 0.68, D(G(z)): 0.30\n",
      "Saving fake_images-0269.png\n",
      "Epoch [270/300], Step [200/938], d_loss: 0.6795, g_loss: 1.7597, D(x): 0.74, D(G(z)): 0.22\n",
      "Epoch [270/300], Step [400/938], d_loss: 0.9660, g_loss: 1.4668, D(x): 0.74, D(G(z)): 0.34\n",
      "Epoch [270/300], Step [600/938], d_loss: 0.9143, g_loss: 1.6558, D(x): 0.68, D(G(z)): 0.26\n",
      "Epoch [270/300], Step [800/938], d_loss: 0.7448, g_loss: 1.7364, D(x): 0.82, D(G(z)): 0.31\n",
      "Saving fake_images-0270.png\n",
      "Epoch [271/300], Step [200/938], d_loss: 0.9269, g_loss: 1.3458, D(x): 0.79, D(G(z)): 0.38\n",
      "Epoch [271/300], Step [400/938], d_loss: 0.8629, g_loss: 1.7789, D(x): 0.68, D(G(z)): 0.24\n",
      "Epoch [271/300], Step [600/938], d_loss: 0.8719, g_loss: 1.7457, D(x): 0.67, D(G(z)): 0.24\n",
      "Epoch [271/300], Step [800/938], d_loss: 0.7844, g_loss: 1.9945, D(x): 0.76, D(G(z)): 0.26\n",
      "Saving fake_images-0271.png\n",
      "Epoch [272/300], Step [200/938], d_loss: 0.7987, g_loss: 1.7628, D(x): 0.77, D(G(z)): 0.27\n",
      "Epoch [272/300], Step [400/938], d_loss: 1.1138, g_loss: 1.6001, D(x): 0.64, D(G(z)): 0.29\n",
      "Epoch [272/300], Step [600/938], d_loss: 0.8684, g_loss: 1.6306, D(x): 0.71, D(G(z)): 0.26\n",
      "Epoch [272/300], Step [800/938], d_loss: 0.8566, g_loss: 1.8436, D(x): 0.68, D(G(z)): 0.25\n",
      "Saving fake_images-0272.png\n",
      "Epoch [273/300], Step [200/938], d_loss: 0.9581, g_loss: 2.0051, D(x): 0.77, D(G(z)): 0.36\n",
      "Epoch [273/300], Step [400/938], d_loss: 0.8200, g_loss: 1.7788, D(x): 0.80, D(G(z)): 0.33\n",
      "Epoch [273/300], Step [600/938], d_loss: 0.8679, g_loss: 1.6985, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [273/300], Step [800/938], d_loss: 0.9050, g_loss: 1.4502, D(x): 0.70, D(G(z)): 0.28\n",
      "Saving fake_images-0273.png\n",
      "Epoch [274/300], Step [200/938], d_loss: 0.8041, g_loss: 1.5317, D(x): 0.74, D(G(z)): 0.29\n",
      "Epoch [274/300], Step [400/938], d_loss: 0.8961, g_loss: 1.3865, D(x): 0.82, D(G(z)): 0.39\n",
      "Epoch [274/300], Step [600/938], d_loss: 0.6988, g_loss: 1.9455, D(x): 0.79, D(G(z)): 0.28\n",
      "Epoch [274/300], Step [800/938], d_loss: 1.0045, g_loss: 1.6654, D(x): 0.64, D(G(z)): 0.25\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0274.png\n",
      "Epoch [275/300], Step [200/938], d_loss: 0.9118, g_loss: 1.4056, D(x): 0.70, D(G(z)): 0.27\n",
      "Epoch [275/300], Step [400/938], d_loss: 0.7824, g_loss: 1.3613, D(x): 0.75, D(G(z)): 0.27\n",
      "Epoch [275/300], Step [600/938], d_loss: 0.7531, g_loss: 1.7711, D(x): 0.73, D(G(z)): 0.25\n",
      "Epoch [275/300], Step [800/938], d_loss: 0.7886, g_loss: 1.8198, D(x): 0.78, D(G(z)): 0.31\n",
      "Saving fake_images-0275.png\n",
      "Epoch [276/300], Step [200/938], d_loss: 0.8878, g_loss: 1.7569, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [276/300], Step [400/938], d_loss: 1.0337, g_loss: 1.8797, D(x): 0.68, D(G(z)): 0.31\n",
      "Epoch [276/300], Step [600/938], d_loss: 0.7145, g_loss: 1.6917, D(x): 0.78, D(G(z)): 0.27\n",
      "Epoch [276/300], Step [800/938], d_loss: 0.9195, g_loss: 1.7651, D(x): 0.82, D(G(z)): 0.38\n",
      "Saving fake_images-0276.png\n",
      "Epoch [277/300], Step [200/938], d_loss: 0.7421, g_loss: 1.6973, D(x): 0.78, D(G(z)): 0.29\n",
      "Epoch [277/300], Step [400/938], d_loss: 1.0014, g_loss: 1.4410, D(x): 0.67, D(G(z)): 0.29\n",
      "Epoch [277/300], Step [600/938], d_loss: 0.7504, g_loss: 1.4617, D(x): 0.74, D(G(z)): 0.26\n",
      "Epoch [277/300], Step [800/938], d_loss: 0.8756, g_loss: 1.7158, D(x): 0.70, D(G(z)): 0.25\n",
      "Saving fake_images-0277.png\n",
      "Epoch [278/300], Step [200/938], d_loss: 0.7626, g_loss: 1.8364, D(x): 0.69, D(G(z)): 0.19\n",
      "Epoch [278/300], Step [400/938], d_loss: 0.7591, g_loss: 1.7811, D(x): 0.77, D(G(z)): 0.28\n",
      "Epoch [278/300], Step [600/938], d_loss: 1.0064, g_loss: 1.5658, D(x): 0.68, D(G(z)): 0.31\n",
      "Epoch [278/300], Step [800/938], d_loss: 0.9703, g_loss: 1.6126, D(x): 0.67, D(G(z)): 0.29\n",
      "Saving fake_images-0278.png\n",
      "Epoch [279/300], Step [200/938], d_loss: 0.7356, g_loss: 1.5737, D(x): 0.78, D(G(z)): 0.30\n",
      "Epoch [279/300], Step [400/938], d_loss: 0.8389, g_loss: 2.0117, D(x): 0.68, D(G(z)): 0.25\n",
      "Epoch [279/300], Step [600/938], d_loss: 0.8956, g_loss: 1.9264, D(x): 0.78, D(G(z)): 0.37\n",
      "Epoch [279/300], Step [800/938], d_loss: 0.9370, g_loss: 1.7731, D(x): 0.73, D(G(z)): 0.30\n",
      "Saving fake_images-0279.png\n",
      "Epoch [280/300], Step [200/938], d_loss: 0.8695, g_loss: 1.3965, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [280/300], Step [400/938], d_loss: 0.7529, g_loss: 1.9621, D(x): 0.74, D(G(z)): 0.25\n",
      "Epoch [280/300], Step [600/938], d_loss: 1.1166, g_loss: 1.5069, D(x): 0.61, D(G(z)): 0.28\n",
      "Epoch [280/300], Step [800/938], d_loss: 0.7065, g_loss: 1.5532, D(x): 0.80, D(G(z)): 0.28\n",
      "Saving fake_images-0280.png\n",
      "Epoch [281/300], Step [200/938], d_loss: 0.9556, g_loss: 1.6786, D(x): 0.70, D(G(z)): 0.31\n",
      "Epoch [281/300], Step [400/938], d_loss: 0.7024, g_loss: 2.2326, D(x): 0.70, D(G(z)): 0.17\n",
      "Epoch [281/300], Step [600/938], d_loss: 0.8541, g_loss: 1.7841, D(x): 0.75, D(G(z)): 0.30\n",
      "Epoch [281/300], Step [800/938], d_loss: 0.7920, g_loss: 1.3229, D(x): 0.83, D(G(z)): 0.38\n",
      "Saving fake_images-0281.png\n",
      "Epoch [282/300], Step [200/938], d_loss: 0.9808, g_loss: 1.2498, D(x): 0.63, D(G(z)): 0.27\n",
      "Epoch [282/300], Step [400/938], d_loss: 0.6153, g_loss: 1.6600, D(x): 0.78, D(G(z)): 0.22\n",
      "Epoch [282/300], Step [600/938], d_loss: 0.8595, g_loss: 2.0223, D(x): 0.74, D(G(z)): 0.26\n",
      "Epoch [282/300], Step [800/938], d_loss: 0.8588, g_loss: 1.9695, D(x): 0.74, D(G(z)): 0.28\n",
      "Saving fake_images-0282.png\n",
      "Epoch [283/300], Step [200/938], d_loss: 0.8221, g_loss: 1.9614, D(x): 0.71, D(G(z)): 0.21\n",
      "Epoch [283/300], Step [400/938], d_loss: 0.7082, g_loss: 1.6359, D(x): 0.73, D(G(z)): 0.25\n",
      "Epoch [283/300], Step [600/938], d_loss: 0.7083, g_loss: 1.7119, D(x): 0.74, D(G(z)): 0.23\n",
      "Epoch [283/300], Step [800/938], d_loss: 0.9867, g_loss: 1.4373, D(x): 0.68, D(G(z)): 0.32\n",
      "Saving fake_images-0283.png\n",
      "Epoch [284/300], Step [200/938], d_loss: 1.1076, g_loss: 1.5279, D(x): 0.63, D(G(z)): 0.27\n",
      "Epoch [284/300], Step [400/938], d_loss: 0.8959, g_loss: 1.8441, D(x): 0.72, D(G(z)): 0.32\n",
      "Epoch [284/300], Step [600/938], d_loss: 0.9541, g_loss: 1.7045, D(x): 0.78, D(G(z)): 0.39\n",
      "Epoch [284/300], Step [800/938], d_loss: 0.7539, g_loss: 1.7190, D(x): 0.74, D(G(z)): 0.27\n",
      "Saving fake_images-0284.png\n",
      "Epoch [285/300], Step [200/938], d_loss: 0.7160, g_loss: 1.6878, D(x): 0.81, D(G(z)): 0.32\n",
      "Epoch [285/300], Step [400/938], d_loss: 1.0694, g_loss: 1.7129, D(x): 0.72, D(G(z)): 0.35\n",
      "Epoch [285/300], Step [600/938], d_loss: 0.9029, g_loss: 1.6635, D(x): 0.73, D(G(z)): 0.31\n",
      "Epoch [285/300], Step [800/938], d_loss: 0.8507, g_loss: 1.5080, D(x): 0.75, D(G(z)): 0.35\n",
      "Saving fake_images-0285.png\n",
      "Epoch [286/300], Step [200/938], d_loss: 0.9113, g_loss: 1.9848, D(x): 0.66, D(G(z)): 0.25\n",
      "Epoch [286/300], Step [400/938], d_loss: 1.0024, g_loss: 1.5771, D(x): 0.73, D(G(z)): 0.35\n",
      "Epoch [286/300], Step [600/938], d_loss: 0.8985, g_loss: 1.4426, D(x): 0.71, D(G(z)): 0.31\n",
      "Epoch [286/300], Step [800/938], d_loss: 0.9627, g_loss: 1.7317, D(x): 0.66, D(G(z)): 0.23\n",
      "Saving fake_images-0286.png\n",
      "Epoch [287/300], Step [200/938], d_loss: 0.8274, g_loss: 1.4625, D(x): 0.76, D(G(z)): 0.30\n",
      "Epoch [287/300], Step [400/938], d_loss: 0.7250, g_loss: 1.9385, D(x): 0.73, D(G(z)): 0.20\n",
      "Epoch [287/300], Step [600/938], d_loss: 0.8922, g_loss: 1.3488, D(x): 0.78, D(G(z)): 0.34\n",
      "Epoch [287/300], Step [800/938], d_loss: 1.0179, g_loss: 1.7167, D(x): 0.63, D(G(z)): 0.28\n",
      "Saving fake_images-0287.png\n",
      "Epoch [288/300], Step [200/938], d_loss: 0.8406, g_loss: 1.9642, D(x): 0.68, D(G(z)): 0.23\n",
      "Epoch [288/300], Step [400/938], d_loss: 1.0513, g_loss: 2.1492, D(x): 0.62, D(G(z)): 0.17\n",
      "Epoch [288/300], Step [600/938], d_loss: 0.7565, g_loss: 1.9739, D(x): 0.68, D(G(z)): 0.19\n",
      "Epoch [288/300], Step [800/938], d_loss: 0.8518, g_loss: 1.7902, D(x): 0.75, D(G(z)): 0.30\n",
      "Saving fake_images-0288.png\n",
      "Epoch [289/300], Step [200/938], d_loss: 0.8543, g_loss: 1.9716, D(x): 0.75, D(G(z)): 0.31\n",
      "Epoch [289/300], Step [400/938], d_loss: 1.0679, g_loss: 1.4718, D(x): 0.57, D(G(z)): 0.23\n",
      "Epoch [289/300], Step [600/938], d_loss: 0.7769, g_loss: 1.7660, D(x): 0.68, D(G(z)): 0.19\n",
      "Epoch [289/300], Step [800/938], d_loss: 0.8143, g_loss: 1.5127, D(x): 0.73, D(G(z)): 0.28\n",
      "Saving fake_images-0289.png\n",
      "Epoch [290/300], Step [200/938], d_loss: 0.8154, g_loss: 1.5534, D(x): 0.74, D(G(z)): 0.28\n",
      "Epoch [290/300], Step [400/938], d_loss: 0.7868, g_loss: 1.5919, D(x): 0.74, D(G(z)): 0.26\n",
      "Epoch [290/300], Step [600/938], d_loss: 0.7548, g_loss: 1.9871, D(x): 0.72, D(G(z)): 0.20\n",
      "Epoch [290/300], Step [800/938], d_loss: 0.8950, g_loss: 1.5665, D(x): 0.74, D(G(z)): 0.31\n",
      "Saving fake_images-0290.png\n",
      "Epoch [291/300], Step [200/938], d_loss: 0.7793, g_loss: 1.5396, D(x): 0.84, D(G(z)): 0.32\n",
      "Epoch [291/300], Step [400/938], d_loss: 0.8170, g_loss: 2.0017, D(x): 0.69, D(G(z)): 0.23\n",
      "Epoch [291/300], Step [600/938], d_loss: 0.9008, g_loss: 1.6801, D(x): 0.70, D(G(z)): 0.29\n",
      "Epoch [291/300], Step [800/938], d_loss: 0.8028, g_loss: 1.7600, D(x): 0.76, D(G(z)): 0.31\n",
      "Saving fake_images-0291.png\n",
      "Epoch [292/300], Step [200/938], d_loss: 0.8705, g_loss: 2.2612, D(x): 0.70, D(G(z)): 0.23\n",
      "Epoch [292/300], Step [400/938], d_loss: 0.9737, g_loss: 1.9423, D(x): 0.66, D(G(z)): 0.25\n",
      "Epoch [292/300], Step [600/938], d_loss: 0.9005, g_loss: 1.5896, D(x): 0.75, D(G(z)): 0.33\n",
      "Epoch [292/300], Step [800/938], d_loss: 0.8759, g_loss: 1.3744, D(x): 0.79, D(G(z)): 0.36\n",
      "Saving fake_images-0292.png\n",
      "Epoch [293/300], Step [200/938], d_loss: 0.7030, g_loss: 1.7151, D(x): 0.72, D(G(z)): 0.22\n",
      "Epoch [293/300], Step [400/938], d_loss: 0.8903, g_loss: 1.8473, D(x): 0.72, D(G(z)): 0.30\n",
      "Epoch [293/300], Step [600/938], d_loss: 1.0229, g_loss: 1.7750, D(x): 0.69, D(G(z)): 0.32\n",
      "Epoch [293/300], Step [800/938], d_loss: 0.7393, g_loss: 2.5314, D(x): 0.71, D(G(z)): 0.21\n",
      "Saving fake_images-0293.png\n",
      "Epoch [294/300], Step [200/938], d_loss: 1.0045, g_loss: 1.6851, D(x): 0.61, D(G(z)): 0.23\n",
      "Epoch [294/300], Step [400/938], d_loss: 0.8633, g_loss: 1.5905, D(x): 0.77, D(G(z)): 0.29\n",
      "Epoch [294/300], Step [600/938], d_loss: 0.8973, g_loss: 2.1350, D(x): 0.64, D(G(z)): 0.16\n",
      "Epoch [294/300], Step [800/938], d_loss: 0.8908, g_loss: 1.9554, D(x): 0.62, D(G(z)): 0.19\n",
      "Saving fake_images-0294.png\n",
      "Epoch [295/300], Step [200/938], d_loss: 0.8497, g_loss: 1.3426, D(x): 0.76, D(G(z)): 0.32\n",
      "Epoch [295/300], Step [400/938], d_loss: 0.7384, g_loss: 1.3665, D(x): 0.79, D(G(z)): 0.30\n",
      "Epoch [295/300], Step [600/938], d_loss: 0.6442, g_loss: 1.6552, D(x): 0.77, D(G(z)): 0.23\n",
      "Epoch [295/300], Step [800/938], d_loss: 0.9251, g_loss: 2.1810, D(x): 0.66, D(G(z)): 0.24\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving fake_images-0295.png\n",
      "Epoch [296/300], Step [200/938], d_loss: 0.9103, g_loss: 2.0421, D(x): 0.72, D(G(z)): 0.28\n",
      "Epoch [296/300], Step [400/938], d_loss: 0.8319, g_loss: 1.1940, D(x): 0.71, D(G(z)): 0.25\n",
      "Epoch [296/300], Step [600/938], d_loss: 0.7688, g_loss: 1.4362, D(x): 0.71, D(G(z)): 0.24\n",
      "Epoch [296/300], Step [800/938], d_loss: 0.7173, g_loss: 1.8730, D(x): 0.75, D(G(z)): 0.26\n",
      "Saving fake_images-0296.png\n",
      "Epoch [297/300], Step [200/938], d_loss: 0.8203, g_loss: 2.1629, D(x): 0.71, D(G(z)): 0.26\n",
      "Epoch [297/300], Step [400/938], d_loss: 0.8679, g_loss: 1.5623, D(x): 0.74, D(G(z)): 0.29\n",
      "Epoch [297/300], Step [600/938], d_loss: 0.8941, g_loss: 2.1377, D(x): 0.72, D(G(z)): 0.31\n",
      "Epoch [297/300], Step [800/938], d_loss: 0.8580, g_loss: 2.1378, D(x): 0.75, D(G(z)): 0.30\n",
      "Saving fake_images-0297.png\n",
      "Epoch [298/300], Step [200/938], d_loss: 0.8426, g_loss: 1.7080, D(x): 0.71, D(G(z)): 0.24\n",
      "Epoch [298/300], Step [400/938], d_loss: 0.8889, g_loss: 1.7376, D(x): 0.64, D(G(z)): 0.19\n",
      "Epoch [298/300], Step [600/938], d_loss: 0.8332, g_loss: 1.5214, D(x): 0.75, D(G(z)): 0.30\n",
      "Epoch [298/300], Step [800/938], d_loss: 0.7586, g_loss: 1.7320, D(x): 0.80, D(G(z)): 0.32\n",
      "Saving fake_images-0298.png\n",
      "Epoch [299/300], Step [200/938], d_loss: 0.6358, g_loss: 1.5596, D(x): 0.80, D(G(z)): 0.27\n",
      "Epoch [299/300], Step [400/938], d_loss: 0.8847, g_loss: 1.6709, D(x): 0.73, D(G(z)): 0.30\n",
      "Epoch [299/300], Step [600/938], d_loss: 0.9706, g_loss: 1.7396, D(x): 0.73, D(G(z)): 0.37\n",
      "Epoch [299/300], Step [800/938], d_loss: 0.7581, g_loss: 2.0606, D(x): 0.76, D(G(z)): 0.24\n",
      "Saving fake_images-0299.png\n",
      "Epoch [300/300], Step [200/938], d_loss: 0.7578, g_loss: 1.6155, D(x): 0.73, D(G(z)): 0.26\n",
      "Epoch [300/300], Step [400/938], d_loss: 0.8461, g_loss: 1.4387, D(x): 0.81, D(G(z)): 0.34\n",
      "Epoch [300/300], Step [600/938], d_loss: 0.9573, g_loss: 1.6214, D(x): 0.75, D(G(z)): 0.34\n",
      "Epoch [300/300], Step [800/938], d_loss: 0.8867, g_loss: 2.0488, D(x): 0.70, D(G(z)): 0.26\n",
      "Saving fake_images-0300.png\n",
      "CPU times: user 56min 22s, sys: 8.13 s, total: 56min 30s\n",
      "Wall time: 55min 57s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "epochs = 300\n",
    "total_steps = len(train_dl)\n",
    "d_losses = []\n",
    "g_losses = []\n",
    "pred_reals = []\n",
    "pred_fakes = []\n",
    "\n",
    "save_on_epoch = [10, 50, 100, 150, 300]\n",
    "\n",
    "for epoch in range(epochs):\n",
    "    for i,  (imgs, _) in enumerate(train_dl):\n",
    "        imgs = imgs.view(len(imgs), -1)\n",
    "        \n",
    "        d_loss, pred_real_imgs, pred_fake_imgs = train_discriminator(imgs)\n",
    "        g_loss, fake_imgs = train_generator()\n",
    "        \n",
    "        # Inspect the losses\n",
    "        if (i+1) % 200 == 0:\n",
    "            d_losses.append(d_loss.item())\n",
    "            g_losses.append(g_loss.item())\n",
    "            pred_reals.append(pred_real_imgs.mean().item())\n",
    "            pred_fakes.append(pred_fake_imgs.mean().item())\n",
    "            print('Epoch [{}/{}], Step [{}/{}], d_loss: {:.4f}, g_loss: {:.4f}, D(x): {:.2f}, D(G(z)): {:.2f}' \n",
    "                  .format(epoch+1, epochs, i+1, total_steps, d_loss.item(), g_loss.item(), \n",
    "                          pred_real_imgs.mean().item(), pred_fake_imgs.mean().item()))\n",
    "            \n",
    "    save_fake_images(epoch+1)\n",
    "    if (epoch+1) in save_on_epoch:\n",
    "        torch.save(D.state_dict(), f'./mnist_gans/model/D_gans_epoch{epoch}.pth')\n",
    "        torch.save(G.state_dict(), f'./mnist_gans/model/G_gans_epoch{epoch}.pth')\n",
    "        print(f'\\t*** SAVED MODEL FOR EPOCH {epoch} ***')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb89f0b5",
   "metadata": {},
   "source": [
    "## Generated Images after epoch: [10, 50, 150, 300]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a1da3051",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./mnist_gans/samples/fake_images-0010.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "954e36d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./mnist_gans/samples/fake_images-0050.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "bd769e1e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./mnist_gans/samples/fake_images-0150.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "16a2ab44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image('./mnist_gans/samples/fake_images-0300.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "84453488",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "26c9ea39",
   "metadata": {},
   "source": [
    "## Make generated images as video"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "921e443c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<a href='gans_mnist.mp4' target='_blank'>gans_mnist.mp4</a><br>"
      ],
      "text/plain": [
       "/home/sophot/Documents/gans/gans_mnist.mp4"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import cv2\n",
    "from IPython.display import FileLink\n",
    "\n",
    "vid_fname = 'gans_mnist.mp4'\n",
    "\n",
    "files = [os.path.join(sample_directory, f) for f in os.listdir(sample_directory) if 'fake_images' in f]\n",
    "files.sort()\n",
    "\n",
    "img_array = []\n",
    "for f in files:\n",
    "    img = cv2.imread(f)\n",
    "    height, width, layers = img.shape\n",
    "    img_array.append(img)\n",
    "\n",
    "size = (width,height)\n",
    "out = cv2.VideoWriter(vid_fname, cv2.VideoWriter_fourcc(*'mp4v'), 20 , size)\n",
    "\n",
    "for img_arr in img_array:\n",
    "    out.write(img_arr)\n",
    "out.release()\n",
    "\n",
    "FileLink('gans_mnist.mp4')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "37308dfb",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "me",
   "language": "python",
   "name": "me"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}