simpleGAN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "simpleGAN.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNi0sXlnIM0o48H3WRhbPDv",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/foxtrotmike/CS909/blob/master/simpleGAN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zLPmvLXGB4vD"
      },
      "source": [
        "## A Barebones GAN in PyTorch for generating coin flips\r\n",
        "By\r\n",
        "Fayyaz Minhas"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "45xioqUrUBd_"
      },
      "source": [
        "Let's consider a very simple coing toss as a process that generates coin flips with a probability of 0.3 of producing heads. We can describe the underlying probability distribution for this generative process (coin toss) as $p(x)$ where $x\\in \\{H=1,T=0\\}$ is sampled from $p(x)$, i.e., $x\\backsim p(x)$. We would like to use a Generative Adversarial Network (GAN) to model this process using a number of data samples or observations from the original process for training. Specifically, we would like to have a GAN with such a generator that you (and its discriminator) wouldn't be able to tell if a series of coin tosses has been generated using the GAN or the underlying true process! In more mathematical terms, we would like to train a generative model $x=G(z;\\theta_G)$ that can generate samples $x$ using Normally distributed random input ($z\\backsim N(0,1)$) such that the probability distribution of these generated samples $p_G(x)$ is close to $p(x)$ without knowing $p(x)$ in advance or explicitly modelling $p_G(x)$. \r\n",
        "\r\n",
        "Using a GAN is an overkill for this simple task and there are much simpler and more effective ways of modelling this simple problem. However, this GAN based solution is intended to help you understand how GANs can model complex densities implicitly and can be used to generate samples that mimic the true or natural generative process. \r\n",
        "\r\n",
        "We first simulate the coin toss and generate 1024 training samples below. The histogram shows the (sample estimate of) the true density. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "id": "VF5i_1tI9WdE",
        "outputId": "8b91cfe1-bfe1-4e05-eb6a-a7cabdf0afd4"
      },
      "source": [
        "\"\"\"\r\n",
        "A toy GAN to generate coin tosses\r\n",
        "\"\"\"\r\n",
        "\r\n",
        "# Let's model the natural density and generate some data using that\r\n",
        "\r\n",
        "import torch\r\n",
        "from torch import nn\r\n",
        "\r\n",
        "import math\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "import numpy as np\r\n",
        "train_data_length = 1024\r\n",
        "def cointoss(t):\r\n",
        "  phead = 0.3\r\n",
        "  return 1.0*(t<phead)\r\n",
        "\r\n",
        "train_data = torch.zeros((train_data_length, 1))\r\n",
        "train_data[:, 0] = cointoss(torch.rand(train_data_length))\r\n",
        "bins = np.linspace(-5, 5, 100)\r\n",
        "plt.hist(train_data[:,0],bins=bins,alpha=0.5, label='True')\r\n",
        "plt.title('Histogram');plt.xlabel('$x$');plt.ylabel('$count(x)$');plt.legend()\r\n",
        "plt.show()\r\n",
        "print('Example Data',train_data.T)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Example Data tensor([[0., 0., 0.,  ..., 0., 1., 1.]])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xYybUZWQUwMq"
      },
      "source": [
        "Below, we make a generator and a discriminator and train them with a batch size of 32. The generator will produce coin toss results which should have the same probabilities as the original coin and it will learn its weight parameters based on the observations only and its interaction with the discriminator. The generator takes a Gaussian random noise prior as input, i.e., its input is a single Gaussian Random number with mean zero and variance 1.0. It will learn to map this input number to the target variable (heads or tails) by learning the sampling based on given observations. Consequently, our generator is a single input, single output neural network. \r\n",
        "\r\n",
        "The discriminator will take a random number as input and decide whether it corresponds to fake (generated by the Generator) or real (generated by the real process) coin flip by learning to produce an output of 0.0 for fake and a label of 1.0 for real samples. \r\n",
        "\r\n",
        "We use binary cross entropy as the loss function in this case. At each training step the discriminator learns to classify real and fake data samples whereas the generator tries to produce samples that match the natural process and can fool the discriminator. For visualization, we plot the current estimate of the histogram after eevry few epochs and compare it to the true density. However, it is important to note that GANs do not explicitly model the density $p(x)$ but rather generate samples $x$ through an implicit modeling of the underlying probability density. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "o1EXdwldZl8R",
        "outputId": "e2c84cef-268d-452b-afe0-217080ac0881"
      },
      "source": [
        "train_labels = torch.zeros(train_data_length)\r\n",
        "train_set = [\r\n",
        "    (train_data[i], train_labels[i]) for i in range(train_data_length)]\r\n",
        "batch_size = 32\r\n",
        "train_loader = torch.utils.data.DataLoader(\r\n",
        "    train_set, batch_size=batch_size, shuffle=True)\r\n",
        "\r\n",
        "# Discriminator classifies a given input into real (1) or fake (0)\r\n",
        "class Discriminator(nn.Module):\r\n",
        "    def __init__(self):\r\n",
        "        super().__init__()\r\n",
        "        self.model = nn.Sequential(\r\n",
        "            nn.Linear(1, 32),\r\n",
        "            nn.ReLU(),            \r\n",
        "            nn.Linear(32, 1),\r\n",
        "            nn.Sigmoid())\r\n",
        "\r\n",
        "    def forward(self, x):\r\n",
        "        output = self.model(x)\r\n",
        "        return output\r\n",
        "\r\n",
        "discriminator = Discriminator()\r\n",
        "\r\n",
        "\r\n",
        "#Generator generates a sample using latent random noise factors as input\r\n",
        "class Generator(nn.Module):\r\n",
        "    def __init__(self):\r\n",
        "        super().__init__()\r\n",
        "        self.model = nn.Sequential(\r\n",
        "            nn.Linear(1, 64),\r\n",
        "            nn.ReLU(),\r\n",
        "            nn.Linear(64, 1))\r\n",
        "\r\n",
        "    def forward(self, x):\r\n",
        "        output = self.model(x)\r\n",
        "        return output\r\n",
        "\r\n",
        "generator = Generator()\r\n",
        "lr = 0.001\r\n",
        "num_epochs = 500\r\n",
        "loss_function = nn.BCELoss()\r\n",
        "\r\n",
        "optimizer_discriminator = torch.optim.Adam(discriminator.parameters(), lr=lr)\r\n",
        "optimizer_generator = torch.optim.Adam(generator.parameters(), lr=lr)\r\n",
        "showPlot=True\r\n",
        "for epoch in range(num_epochs):\r\n",
        "  \r\n",
        "    if epoch%50==0 and showPlot:\r\n",
        "        latent_space_samples = torch.randn(1000, 1)\r\n",
        "        generated_samples = generator(latent_space_samples)\r\n",
        "        latent_space_samples = torch.randn(1000, 1).detach()\r\n",
        "        generated_samples = generated_samples.detach()\r\n",
        "        plt.hist(train_data[:,0],bins=bins,alpha=0.5, label='True')\r\n",
        "        plt.hist(generated_samples[:,0],bins=bins,alpha=0.5, label='Estimated')\r\n",
        "        plt.title('Histogram');plt.xlabel('$x$');plt.ylabel('$count(x)$');plt.legend()\r\n",
        "        plt.show()\r\n",
        "\r\n",
        "\r\n",
        "    for n, (real_samples, _) in enumerate(train_loader):\r\n",
        "        # Data for descriminator training\r\n",
        "        real_samples_labels = torch.ones((batch_size, 1))\r\n",
        "        latent_space_samples = torch.randn((batch_size, 1))\r\n",
        "        generated_samples = generator(latent_space_samples)\r\n",
        "        generated_samples_labels = torch.zeros((batch_size, 1))\r\n",
        "        all_samples = torch.cat((real_samples, generated_samples))\r\n",
        "        all_samples_labels = torch.cat(\r\n",
        "            (real_samples_labels, generated_samples_labels))\r\n",
        "\r\n",
        "        # Discriminator training\r\n",
        "        discriminator.zero_grad()\r\n",
        "        output_discriminator = discriminator(all_samples)\r\n",
        "        loss_discriminator = loss_function(\r\n",
        "            output_discriminator, all_samples_labels)\r\n",
        "        loss_discriminator.backward()\r\n",
        "        optimizer_discriminator.step()\r\n",
        "\r\n",
        "        # Data for generator training\r\n",
        "        latent_space_samples = torch.randn((batch_size, 1))\r\n",
        "\r\n",
        "        # Generator training\r\n",
        "        generator.zero_grad()\r\n",
        "        generated_samples = generator(latent_space_samples)\r\n",
        "        output_discriminator_generated = discriminator(generated_samples)\r\n",
        "        loss_generator = loss_function(\r\n",
        "            output_discriminator_generated, real_samples_labels)\r\n",
        "        loss_generator.backward()\r\n",
        "        optimizer_generator.step()\r\n",
        "        \r\n",
        "\r\n",
        "        # Output value of loss function\r\n",
        "        if epoch % 50 == 0 and n == batch_size - 1:\r\n",
        "            print(f\"Epoch: {epoch} Loss D.: {loss_discriminator}\")\r\n",
        "            print(f\"Epoch: {epoch} Loss G.: {loss_generator}\")\r\n",
        "\r\n",
        "           \r\n",
        "\r\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAbyElEQVR4nO3dfZRV1Z3m8e8joBUBQ8JbM4AWMdAGQZGUEWOWbSyjYCOYFWMwpgXDCp3RyRgTTcB0J5m0aU07S6OTaJqJTkOLUQfDS+gkg0OTcYIxBrA0KhrwhaEILyVaSLUQrfY3f9xdeCkPUBfqnnupej5r1brn7PNyfxfWqqf2Pufuo4jAzMysvaMqXYCZmVUnB4SZmWVyQJiZWSYHhJmZZXJAmJlZJgeEmZllckCYJZKekXROpeswqxYOCOs2JL0s6bx2bTMk/RogIk6OiF8d5By1kkJSzzKWalYVHBBmVcTBY9XEAWGWFPcwJH1E0mpJr0vaJunWtNsj6bVZUoukMyUdJelvJG2UtF3SfEnvLTrvFWnbDkl/2+59vi1poaR7Jb0OzEjv/RtJzZK2SPqBpKOLzheSrpK0XtIuSX8n6URJj6Z6Hyze3+xQOSDMst0O3B4RxwEnAg+m9rPTa7+I6BMRvwFmpJ+PAx8A+gA/AJA0GrgTuBwYArwXGNruvaYCC4F+wALg34FrgQHAmUA9cFW7Yy4APgxMAL4GzAU+BwwHxgCXHcZnNwMcENb9LE5/mTdLaqbwyzvLW8AHJQ2IiJaIeOwA57wcuDUiXoyIFmAOMC0NF10C/Cwifh0RbwLfBNpPgPabiFgcEW9HxO6IWBMRj0VEa0S8DPwj8BftjvmHiHg9Ip4BngaWp/ffCfwCOK3j/yRm2RwQ1t1cHBH92n5491/mbWYCo4DnJP1O0uQDnPM/ABuL1jcCPYHBadumtg0R8Qawo93xm4pXJI2StEzS1jTs9PcUehPFthUt785Y73OAes06xAFhliEi1kfEZcAg4HvAQkm9efdf/wB/BE4oWj8eaKXwS3sLMKxtg6T3AP3bv1279buA54CRaYjrBkCH/mnMDo0DwiyDpM9JGhgRbwPNqfltoCm9fqBo958A10oaIakPhb/4H4iIVgrXFi6S9NF04fjbHPyXfV/gdaBF0knAf+ysz2VWCgeEWbaJwDOSWihcsJ6Wrg+8AXwXWJWuY0wA7gH+mcIdTi8Be4AvAaRrBF8C7qfQm2gBtgN/OsB7Xwd8FtgF/Hfggc7/eGYHJz8wyCw/qYfRTGH46KVK12N2IO5BmJWZpIskHZuuYfxX4PfAy5WtyuzgHBBm5TeVwoXsPwIjKQxXuetuVc9DTGZmlsk9CDMzy9RlJgYbMGBA1NbWVroMM7Mjypo1a16JiIFZ27pMQNTW1rJ69epKl2FmdkSRtHF/2zzEZGZmmRwQZmaWyQFhZmaZusw1CDPrOt566y0aGxvZs2dPpUvpMmpqahg2bBi9evXq8DEOCDOrOo2NjfTt25fa2lokT2R7uCKCHTt20NjYyIgRIzp8nIeYzKzq7Nmzh/79+zscOokk+vfvX3KPzAFhZlXJ4dC5DuXf0wFhZmaZcrkGIenP2XdO+w9QeDbv/NReS2F2y0sj4jUVou524ELgDWBGRKzNo1Yzqz63PfyHTj3ftZ8Ytd9tO3bsoL6+HoCtW7fSo0cPBg4sfNH48ccf5+ijj+7UWqpZLgEREc8D4wAk9QA2A4uA2cCKiLhZ0uy0/nVgEoVZL0cCZ1B4BOMZedRq1pmKf7Ed6JeSVY/+/fvT0NAAwLe//W369OnDddddt3d7a2srPXt2j/t7KvEp64EXImKjpKnAOal9HvArCgExFZifpkR+TFI/SUMiYksF6jWzbm7GjBnU1NTwxBNPcNZZZ3HcccftExxjxoxh2bJl1NbWcu+993LHHXfw5ptvcsYZZ3DnnXfSo0ePCn+CQ1OJaxDTKDzDF2Bw0S/9rcDgtDwU2FR0TGNq24ekWZJWS1rd1NRUrnrNzGhsbOTRRx/l1ltv3e8+69at44EHHmDVqlU0NDTQo0cPFixYkGOVnSvXHkR6aPsUYE77bRERkkp6OEVEzAXmAtTV1fnBFmZWNp/+9KcP2hNYsWIFa9as4fTTTwdg9+7dDBo0KI/yyiLvIaZJwNqI2JbWt7UNHUkaQuFh7lC4RjG86Lhhqc3MrCJ69+69d7lnz568/fbbe9fbvl8QEUyfPp2bbrop9/rKIe8hpst4Z3gJYCkwPS1PB5YUtV+hggnATl9/MLNqUVtby9q1hRsr165dy0svvQRAfX09CxcuZPv2wt+6r776Khs37nc27aqXWw8iPbD9E8BfFzXfDDwoaSawEbg0tf+cwi2uGyjc5nplXnWaWfWptjvAPvWpTzF//nxOPvlkzjjjDEaNKtQ3evRobrzxRs4//3zefvttevXqxQ9/+ENOOOGECld8aLrMM6nr6urCDwyyauPbXA/NunXr+NCHPlTpMrqcrH9XSWsioi5rf3+T2szMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLFP3mJLQzI5sKzv5m8kff9dsP/vo0aMHY8eO3bs+bdo0Zs+enbnv4sWLGTVqFKNHjwbgm9/8JmeffTbnnXfeYZXY3NzMfffdx1VXXVXScVkz0B4qB4SZWTvvec979k75fTCLFy9m8uTJewPiO9/5TqfU0NzczJ133llyQHQmDzGZmXXQ7NmzGT16NKeccgrXXXcdjz76KEuXLuX6669n3LhxvPDCC8yYMYOFCxcChSk55syZw7hx46irq2Pt2rVccMEFnHjiifzoRz8CoKWlhfr6esaPH8/YsWNZsmTJ3vd64YUXGDduHNdffz0At9xyC6effjqnnHIK3/rWt/bW9d3vfpdRo0bxsY99jOeff77TPq97EGZm7ezevZtx48btXZ8zZw7nnXceixYt4rnnnkMSzc3N9OvXjylTpjB58mQuueSSzHMdf/zxNDQ0cO211zJjxgxWrVrFnj17GDNmDF/84hepqalh0aJFHHfccbzyyitMmDCBKVOmcPPNN/P000/v7cksX76c9evX8/jjjxMRTJkyhUceeYTevXtz//3309DQQGtrK+PHj+fDH/5wp/w7OCDMzNrJGmJqbW2lpqaGmTNnMnnyZCZPntyhc02ZMgWAsWPH0tLSQt++fenbty/HHHMMzc3N9O7dmxtuuIFHHnmEo446is2bN7Nt27Z3nWf58uUsX76c0047DSj0PNavX8+uXbv45Cc/ybHHHrvP+3UGDzGZmXVAz549efzxx7nkkktYtmwZEydO7NBxxxxzDABHHXXU3uW29dbWVhYsWEBTUxNr1qyhoaGBwYMH750+vFhEMGfOHBoaGmhoaGDDhg3MnDmzcz7cfjggzMw6oKWlhZ07d3LhhRdy22238eSTTwLQt29fdu3adcjn3blzJ4MGDaJXr16sXLly7/Tg7c97wQUXcM8999DS0gLA5s2b2b59O2effTaLFy9m9+7d7Nq1i5/97GeH8Sn35SEmM6t+B7kttbO1vwYxceJErrnmGqZOncqePXuIiL2PHp02bRpf+MIXuOOOO/ZenC7F5ZdfzkUXXcTYsWOpq6vjpJNOAqB///6cddZZjBkzhkmTJnHLLbewbt06zjzzTAD69OnDvffey/jx4/nMZz7DqaeeyqBBg/Y+za4zeLpvszLydN+HxtN9l4en+zYzs07hgDAzs0wOCDOrSl1l+LtaHMq/pwPCzKpOTU0NO3bscEh0kohgx44d1NTUlHSc72Iys6ozbNgwGhsbaWpqqnQpXUZNTQ3Dhg0r6ZjcAkJSP+DHwBgggM8DzwMPALXAy8ClEfGaJAG3AxcCbwAzImJtXrWaWWX16tWLESNGVLqMbi/PIabbgV9GxEnAqcA6YDawIiJGAivSOsAkYGT6mQXclWOdZmZGTgEh6b3A2cDdABHxZkQ0A1OBeWm3ecDFaXkqMD8KHgP6SRqSR61mZlaQVw9iBNAE/A9JT0j6saTewOCI2JL22QoMTstDgU1FxzemNjMzy0leAdETGA/cFRGnAf/GO8NJAEThdoWSblmQNEvSakmrfTHLzKxz5RUQjUBjRPw2rS+kEBjb2oaO0uv2tH0zMLzo+GGpbR8RMTci6iKibuDAgWUr3sysO8olICJiK7BJ0p+npnrgWWApMD21TQeWpOWlwBUqmADsLBqKMjOzHOT5PYgvAQskHQ28CFxJIaAelDQT2Ahcmvb9OYVbXDdQuM31yhzrNDMzcgyIiGgAsmYMrM/YN4Cry16UmZntl6faMDOzTA4IMzPL5IAwM7NMDggzM8vkgDAzs0wOCDMzy+SAMDOzTA4IMzPL5IAwM7NMDggzM8vkgDAzs0wOCDMzy+SAMDOzTA4IMzPL5IAwM7NMDggzM8vkgDAzs0wOCDMzy+SAMDOzTA4IMzPL5IAwM7NMuQWEpJcl/V5Sg6TVqe39kh6WtD69vi+1S9IdkjZIekrS+LzqNDOzgrx7EB+PiHERUZfWZwMrImIksCKtA0wCRqafWcBdOddpZtbtVXqIaSowLy3PAy4uap8fBY8B/SQNqUSBZmbdVZ4BEcBySWskzUptgyNiS1reCgxOy0OBTUXHNqa2fUiaJWm1pNVNTU3lqtvMrFvqmeN7fSwiNksaBDws6bnijRERkqKUE0bEXGAuQF1dXUnHmpnZgeXWg4iIzel1O7AI+AiwrW3oKL1uT7tvBoYXHT4stZmZWU5yCQhJvSX1bVsGzgeeBpYC09Nu04ElaXkpcEW6m2kCsLNoKMrMzHKQ1xDTYGCRpLb3vC8ifinpd8CDkmYCG4FL0/4/By4ENgBvAFfmVKeZmSW5BEREvAicmtG+A6jPaA/g6hxKMzOz/aj0ba5mZlalHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmUoOiDQza49yFGNmZtXjoAEh6ShJn5X0L5K2A88BWyQ9K+kWSR8sf5lmZpa3jvQgVgInAnOAP4uI4RExCPgY8BjwPUmfK2ONZmZWAR2Z7vu8iHirfWNEvAo8BDwkqVenV2ZmZhV10B5EWzhIul3piT/728fMzLqOUi5S7wKWpkeGIukCSavKU5aZmVVah58oFxF/I+mzwK8kvQm0ALPLVpmZmVVUhwNCUj3wBeDfgCHA5yPi+XIVZmZmlVXKENM3gL+NiHOAS4AHJJ1blqrMzKziOhwQEXFuRPw6Lf8emATcWMqbSeoh6QlJy9L6CEm/lbRB0gOSjk7tx6T1DWl7bSnvY2Zmh68jX5Tb351LW4D6A+2T4RpgXdH694DbIuKDwGvAzNQ+E3gttd+W9jMzsxx16Itykr4k6fjixvTX/pmS5gHTD3YSScOAvwR+nNYFnAssTLvMAy5Oy1PTOml7fQkhZGZmnaAjF6knAp8HfiJpBNAM1AA9gOXA9yPiiQ6c5/vA14C+ab0/0BwRrWm9ERialocCmwAiolXSzrT/K8UnlDQLmAVw/PH75JeZmR2mgwZEROwB7pT0Cwq/xAcAuyOiuaNvImkysD0i1kg651CLzahtLjAXoK6uLjrrvGZmVtpdTA9FxFsRsaUtHCRN6OCxZwFTJL0M3E9haOl2oJ+ktpAaBmxOy5uB4ek9egLvBXaUUKuZmR2mjlykvlTSzUBfSR+SVHzM3I68SUTMiYhhEVELTAP+NSIupzAR4CVpt+nAkrS8lHeua1yS9ncPwcwsRx3pQawCngXeB9wKbJC0Nt2quvsw3//rwFckbaBwjeHu1H430D+1fwV/Y9vMLHcduQaxGZgv6YWIWAUgqT9QS+HZECWJiF8Bv0rLLwIfydhnD/DpUs9tZmadp5S5mFYVLe/A1wTM3rHypneWPz6ncnWYdaJS5mI6F7icwm2uTwNPAU9HxJ/KVJuZmVVQhwMCuAf4MtALOIXCl9pOBvzIUTOzLqiUgNgYEYvT8v8sRzFmZlY9SvkexCOSrvWUF2Zm3UMpPYjRwFjg65LWAA1AQ0S4N2Fm1gWVchfTpwAkvYd3wmICHm4yM+uSSrmLaTXwJPD79PMvEfFPZarLzMwqrJRrEFMo9BaOBv4aeDnNrWRmZl1QKUNMfwT+CPwSQNKHeGceJTMz62I63IOQdELxekSsA0Z1ekVmZlYVSrmL6SfpqXIvUbgGsRMYU5aqzMys4koZYvpo+g7EiRTuYHo/cFG5CjMzs8oq5S6m9wPXAoMoTP89PyJeK1dhZmZWWaXcxXQ/sAv4GXAs8GtJ75qq28zMuoZSrkEMjIh/SMvLJD0A3Efhy3JmZtbFlNKDeFXS2LaV9LCfYzu/JDMzqwal9CCuAh6S9H8p3MV0MvBCWaoyM7OKK6UH0Rf4KLCSwoXqBuCychRlZmaVV0oPYj4wPiIeBJA0ADgPWFaOwszs8Nz28B/2Ll/7CX+n1UpXSg9iT0TsaVuJiFeA73TkQEk1kh6X9KSkZyT9l9Q+QtJvJW2Q9ICko1P7MWl9Q9peW0KdZmbWCUoJiBclTWrXdnQHj/0TcG5EnAqMAyZKmgB8D7gtIj4IvAbMTPvPBF5L7bel/czMLEelBMSXgJsk3SfpP0v6Rzp4kToKWtJqr/QTwLnAwtQ+j8JzrgGmpnXS9no/yc7MLF8dDoiI2AJ8GHgIGAg8BXy2o8dL6iGpAdgOPEwhXJojojXt0ggMTctDgU3pfVspzPvUP+OcsyStlrS6qampo6WYmVkHlHKRmoj4dwoB8VCpb5SOHSepH7AIOKnUc2Sccy4wF6Curi4O93xmZvaOUoaYOkVENFO4VfZMoJ+ktpAaBmxOy5uB4QBp+3uBHTmXambWreUSEJIGpp5D2zOtPwGsoxAUbQ8dmg4sSctL0zpp+79GhHsIZmY5KmmI6TAMAeZJ6kEhlB6MiGWSngXul3Qj8ARwd9r/buCfJW0AXgWm5VSnmZkluQRERDwFnJbR/iLwrhlh0/ctPp1DaWZmth+5X4MwM7MjgwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwy5RIQkoZLWinpWUnPSLomtb9f0sOS1qfX96V2SbpD0gZJT0kan0edZmb2jrx6EK3AVyNiNDABuFrSaGA2sCIiRgIr0jrAJGBk+pkF3JVTnWZmluQSEBGxJSLWpuVdwDpgKDAVmJd2mwdcnJanAvOj4DGgn6QhedRqZmYFuV+DkFQLnAb8FhgcEVvSpq3A4LQ8FNhUdFhjamt/rlmSVkta3dTUVLaazcy6o1wDQlIf4CHgyxHxevG2iAggSjlfRMyNiLqIqBs4cGAnVmpmZrkFhKReFMJhQUT8NDVvaxs6Sq/bU/tmYHjR4cNSm5mZ5SSvu5gE3A2si4hbizYtBaan5enAkqL2K9LdTBOAnUVDUWZmloOeOb3PWcBfAb+X1JDabgBuBh6UNBPYCFyatv0cuBDYALwBXJlTnWZmluQSEBHxa0D72VyfsX8AV5e1KDMzOyB/k9rMzDI5IMzMLFNe1yDMup6VN1W6ArOycg/CzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMuUSEJLukbRd0tNFbe+X9LCk9en1faldku6QtEHSU5LG51GjmZntK68exD8BE9u1zQZWRMRIYEVaB5gEjEw/s4C7cqrRzMyK5BIQEfEI8Gq75qnAvLQ8D7i4qH1+FDwG9JM0JI86zczsHZW8BjE4Irak5a3A4LQ8FNhUtF9jajMzsxxVxUXqiAggSj1O0ixJqyWtbmpqKkNlZmbdVyUDYlvb0FF63Z7aNwPDi/YbltreJSLmRkRdRNQNHDiwrMWamXU3lQyIpcD0tDwdWFLUfkW6m2kCsLNoKMrMzHLSM483kfQT4BxggKRG4FvAzcCDkmYCG4FL0+4/By4ENgBvAFfmUaOZme0rl4CIiMv2s6k+Y98Ari5vRWaHaOVNla7ALDdVcZHazMyqjwPCzMwy5TLEZHZEaT+M9PE5lanDrMIcEGZw4GsLvu5g3ZSHmMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDL5Nlfrvnz7qtkBuQdhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmXwXk3UvvnPJrMPcgzAzs0zuQVjX5J6C2WFzD8LMzDI5IMzMLFPVDjFJmgjcDvQAfhwRN1e4JKtGHkoyK5uq7EFI6gH8EJgEjAYukzS6slWZmXUv1dqD+AiwISJeBJB0PzAVeLaiVVl1cK/BLBfVGhBDgU1F643AGe13kjQLmJVWWyQ9n0NtnW0A8Eqli8hZF//MN7RvGAC88pUKVNKmAu/dxf+PMx2pn/mE/W2o1oDokIiYC8ytdB2HQ9LqiKirdB156m6fubt9XvBn7iqq8hoEsBkYXrQ+LLWZmVlOqjUgfgeMlDRC0tHANGBphWsyM+tWqnKIKSJaJf0n4H9RuM31noh4psJllcsRPUR2iLrbZ+5unxf8mbsERUSlazAzsypUrUNMZmZWYQ4IMzPL5ICoIpK+KikkDah0LeUk6RZJz0l6StIiSf0qXVO5SJoo6XlJGyTNrnQ95SZpuKSVkp6V9IykaypdU14k9ZD0hKRlla6lszggqoSk4cD5wP+rdC05eBgYExGnAH8A5lS4nrLoplPGtAJfjYjRwATg6m7wmdtcA6yrdBGdyQFRPW4DvgZ0+bsGImJ5RLSm1ccofM+lK9o7ZUxEvAm0TRnTZUXElohYm5Z3UfiFObSyVZWfpGHAXwI/rnQtnckBUQUkTQU2R8STla6lAj4P/KLSRZRJ1pQxXf6XZRtJtcBpwG8rW0kuvk/hD7y3K11IZ6rK70F0RZL+N/BnGZu+QWHynvPzrai8DvR5I2JJ2ucbFIYkFuRZm5WfpD7AQ8CXI+L1StdTTpImA9sjYo2kcypdT2dyQOQkIs7Lapc0FhgBPCkJCsMtayV9JCK25lhip9rf520jaQYwGaiPrvtlnG45ZYykXhTCYUFE/LTS9eTgLGCKpAuBGuA4SfdGxOcqXNdh8xflqoykl4G6iDgSZ4XskPQwqFuBv4iIpkrXUy6SelK4CF9PIRh+B3y2C88KgAp/5cwDXo2IL1e6nrylHsR1ETG50rV0Bl+DsEr4AdAXeFhSg6QfVbqgckgX4tumjFkHPNiVwyE5C/gr4Nz0f9uQ/rK2I5B7EGZmlsk9CDMzy+SAMDOzTA4IMzPL5IAwM7NMDggzM8vkgDAzs0wOCDMzy+SAMCuT9FyET6TlGyX9t0rXZFYKz8VkVj7fAr4jaRCFWU2nVLges5L4m9RmZSTp/wB9gHPS8xHMjhgeYjIrkzRT7xDgTYeDHYkcEGZlIGkIhedcTAVa0gy2ZkcUB4RZJ5N0LPBTCs9mXgf8HYXrEWZHFF+DMDOzTO5BmJlZJgeEmZllckCYmVkmB4SZmWVyQJiZWSYHhJmZZXJAmJlZpv8PAikCatB1nFQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 0 Loss D.: 0.6912013292312622\n",
            "Epoch: 0 Loss G.: 0.6844253540039062\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 50 Loss D.: 0.7365711331367493\n",
            "Epoch: 50 Loss G.: 0.6040360331535339\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 100 Loss D.: 0.6204550266265869\n",
            "Epoch: 100 Loss G.: 0.8009586930274963\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 150 Loss D.: 0.622590184211731\n",
            "Epoch: 150 Loss G.: 0.856653094291687\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEWCAYAAABxMXBSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAZ60lEQVR4nO3dfZRddX3v8feHJBAegmhIKCVgUoRqTBBiFBAXtYbKQ2NilygotqSlcnu1FkOxEtsKy2rF0kLlKnq5yi1ckIcL8mBa23ARLxVETGBUMFgCSEnkIQYTkysppHzvH2cnjCHA7GRmzkzm/Vpr1tn7t/fZ57uTteYzv99vn71TVUiS1Fc7dLsASdLwYnBIkloxOCRJrRgckqRWDA5JUisGhySpFYND6oMk9yZ5S7frkIYCg0MCkvw4yVGbtc1L8i2AqnptVX3zJY4xOUklGT2ApUpdZ3BIw4SBpKHC4JD6oHePJMkbkyxO8vMkjyc5r9nt1uZ1dZJ1SQ5PskOSv0jycJInklya5GW9jvt7zbZVSf5ys885O8k1SS5L8nNgXvPZ306yOsmjST6XZMdex6skH0hyf5K1Sf4qyf5Jbm/qvbr3/tLWMDik9j4LfLaqdgf2B65u2o9sXveoqt2q6tvAvObnN4FfA3YDPgeQZCpwIXASsDfwMmCfzT5rLnANsAdwOfCfwHxgT+BwYBbwgc3eczTweuAw4M+Ai4D3AfsC04D3bMO5SwaH1Mv1zV/yq5OspvNLfUueAV6VZM+qWldVd7zIMU8CzquqB6tqHbAAOLEZdjoe+FpVfauqngY+Dmx+87hvV9X1VfVsVT1VVUuq6o6q2lBVPwb+O/Abm73nb6rq51V1L3APsKj5/DXA14FD+v5PIj2fwSE95x1VtcfGH57/l/xGpwAHAvcl+W6S2S9yzF8FHu61/jAwGtir2fbIxg1V9Qtg1Wbvf6T3SpIDkyxM8lgzfPXXdHofvT3ea/mpLazv9iL1Si/J4JBaqqr7q+o9wETgM8A1SXbl+b0FgJ8Ar+y1vh+wgc4v80eBSRs3JNkZGL/5x222/gXgPuCAZqjsY0C2/myk9gwOqaUk70syoaqeBVY3zc8CK5vXX+u1+xXA/CRTkuxGp4dwVVVtoDN38fYkb2omrM/mpUNgHPBzYF2SVwP/tb/OS+org0Nq7xjg3iTr6EyUn9jMP/wC+BRwWzNPchhwMfC/6Fxx9RCwHvgQQDMH8SHgSjq9j3XAE8B/vMhnnwG8F1gL/A/gqv4/PenFxQc5SUND0yNZTWcY6qFu1yO9EHscUhcleXuSXZo5kr8FfgD8uLtVSS/O4JC6ay6dCfSfAAfQGfZyGEBDmkNVkqRW7HFIklrZ7m+atueee9bkyZO7XYYkDStLliz5aVVN2NK27T44Jk+ezOLFi7tdhiQNK0kefqFtDlVJkloxOCRJrRgckqRWtvs5Dknbj2eeeYbly5ezfv36bpey3Rg7diyTJk1izJgxfX6PwSFp2Fi+fDnjxo1j8uTJJN4UeFtVFatWrWL58uVMmTKlz+9zqErSsLF+/XrGjx9vaPSTJIwfP751D25QgiPJxc3zlu/p1faKJDc1z0a+KcnLm/YkuSDJsiTfTzKj13tObva/P8nJg1G7pKHF0OhfW/PvOVg9jn+gcyvq3s4Ebq6qA4Cbm3WAY+ncs+cA4FQ6D64hySuAs4BDgTcCZ20MG0nS4BmUOY6qujXJ5M2a5wJvaZYvAb4JfLRpv7S50dsdSfZIsnez701V9SRAkpvohNEVA1y+pCHq/Jv+rV+PN/+3DnzR7atWrWLWrFkAPPbYY4waNYoJEzpfrr7zzjvZcccd+7Weoaqbk+N7VdWjzfJjdJ7BDLAPv/yc5eVN2wu1P0+SU+n0Vthvv/36sWSpn9zy6eeWf3NB9+pQK+PHj6enpweAs88+m912240zzjhj0/YNGzYwevT2f83RkDjDqqok/Xab3qq6CLgIYObMmd7+V9KAmTdvHmPHjuXuu+/miCOOYPfdd/+lQJk2bRoLFy5k8uTJXHbZZVxwwQU8/fTTHHrooVx44YWMGjWqy2fQXjevqnq8GYKieX2iaV8B7Ntrv0lN2wu1S1JXLV++nNtvv53zzjvvBfdZunQpV111Fbfddhs9PT2MGjWKyy+/fBCr7D/d7HHcCJwMnNO83tCr/Y+TXElnInxNVT2a5F+Av+41If42wD6+pK5717ve9ZI9h5tvvpklS5bwhje8AYCnnnqKiRMnDkZ5/W5QgiPJFXQmt/dMspzO1VHnAFcnOQV4GHh3s/s/AccBy4BfAL8PUFVPJvkr4LvNfp/YOFEuSd206667bloePXo0zz777Kb1jd+RqCpOPvlkPv3pTz/v/cPNYF1V9Z4X2DRrC/sW8MEXOM7FwMX9WJok9avJkyezcOFCAO666y4eeughAGbNmsXcuXOZP38+EydO5Mknn2Tt2rW88pWv7Ga5W2VITI5L0tZ4qctnu+Gd73wnl156Ka997Ws59NBDOfDATo1Tp07lk5/8JG9729t49tlnGTNmDJ///OcNDkkaKc4+++wttu+8884sWrRoi9tOOOEETjjhhAGsanB4rypJUisGhySpFYNDktSKwSFJasXgkCS1YnBIklrxclxJw9ct/fwt7D7cqXjUqFFMnz590/qJJ57ImWeeucV9r7/+eg488ECmTp0KwMc//nGOPPJIjjrqqG0qc/Xq1XzlK1/hAx/4QKv3bemOvlvD4JCkFnbeeedNt1Z/Kddffz2zZ8/eFByf+MQn+qWG1atXc+GFF7YOjv7iUJUk9YMzzzyTqVOnctBBB3HGGWdw++23c+ONN/KRj3yEgw8+mAceeIB58+ZxzTXXAJ1bkyxYsICDDz6YmTNnctddd3H00Uez//7788UvfhGAdevWMWvWLGbMmMH06dO54YYbNn3WAw88wMEHH8xHPvIRAM4991ze8IY3cNBBB3HWWWdtqutTn/oUBx54IG9+85v50Y9+1C/nao9Dklp46qmnOPjggzetL1iwgKOOOorrrruO++67jySsXr2aPfbYgzlz5jB79myOP/74LR5rv/32o6enh/nz5zNv3jxuu+021q9fz7Rp0/ijP/ojxo4dy3XXXcfuu+/OT3/6Uw477DDmzJnDOeecwz333LOp57No0SLuv/9+7rzzTqqKOXPmcOutt7Lrrrty5ZVX0tPTw4YNG5gxYwavf/3rt/nfwOCQpBa2NFS1YcMGxo4dyymnnMLs2bOZPXt2n441Z84cAKZPn866desYN24c48aNY6eddmL16tXsuuuufOxjH+PWW29lhx12YMWKFTz++OPPO86iRYtYtGgRhxxyCNDpqdx///2sXbuW3/md32GXXXb5pc/bVg5VSdI2Gj16NHfeeSfHH388Cxcu5JhjjunT+3baaScAdthhh03LG9c3bNjA5ZdfzsqVK1myZAk9PT3stddem27T3ltVsWDBAnp6eujp6WHZsmWccsop/XNyW2BwSNI2WrduHWvWrOG4447j/PPP53vf+x4A48aNY+3atVt93DVr1jBx4kTGjBnDLbfcwsMPP7zF4x599NFcfPHFrFu3DoAVK1bwxBNPcOSRR3L99dfz1FNPsXbtWr72ta9tw1k+x6EqScNXHy6f7W+bz3Ecc8wxnHbaacydO5f169dTVZseIXviiSfy/ve/nwsuuGDTpHgbJ510Em9/+9uZPn06M2fO5NWvfjUA48eP54gjjmDatGkce+yxnHvuuSxdupTDDz8cgN12243LLruMGTNmcMIJJ/C6172OiRMnbnr64LZK57lJ26+ZM2fW4sWLu12G9Mt6f/+gC7/8hqulS5fymte8pttlbHe29O+aZElVzdzS/g5VSZJaMTgkSa0YHJKGle19eH2wbc2/p8EhadgYO3Ysq1atMjz6SVWxatUqxo4d2+p9XlUladiYNGkSy5cvZ+XKld0uZbsxduxYJk2a1Oo9BoekYWPMmDFMmTKl22WMeA5VSZJaMTgkSa0YHJKkVgwOSVIrBockqRWDQ5LUisEhSWrF4JAktWJwSJJa6XpwJJmf5N4k9yS5IsnYJFOSfCfJsiRXJdmx2XenZn1Zs31yd6uXpJGnq8GRZB/gT4CZVTUNGAWcCHwGOL+qXgX8DNj48NxTgJ817ec3+0mSBlHXexx07pe1c5LRwC7Ao8BbgY3PWbwEeEezPLdZp9k+K0kGsVZJGvG6GhxVtQL4W+Df6QTGGmAJsLqqNjS7LQf2aZb3AR5p3ruh2X/85sdNcmqSxUkWexdNSepf3R6qejmdXsQU4FeBXYFjtvW4VXVRVc2sqpkTJkzY1sNJknrp9lDVUcBDVbWyqp4BvgocAezRDF0BTAJWNMsrgH0Bmu0vA1YNbsmSNLJ1Ozj+HTgsyS7NXMUs4IfALcDxzT4nAzc0yzc26zTbv1E+CkySBlW35zi+Q2eS+y7gB009FwEfBU5PsozOHMaXm7d8GRjftJ8OnDnoRUvSCNf1JwBW1VnAWZs1Pwi8cQv7rgfeNRh1SZK2rNtDVZKkYcbgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktSKwSFJasXgkCS1YnBIkloxOCRJrRgckqRWun53XGkk+vaDzz1/7PDf7GIh0lawxyFJasXgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktSKwSFJasXgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktSKwSFJaqV1cCTZNcmogShGkjT0vWRwJNkhyXuT/GOSJ4D7gEeT/DDJuUletS0FJNkjyTVJ7kuyNMnhSV6R5KYk9zevL2/2TZILkixL8v0kM7blsyVJ7fWlx3ELsD+wAPiVqtq3qiYCbwbuAD6T5H3bUMNngX+uqlcDrwOWAmcCN1fVAcDNzTrAscABzc+pwBe24XMlSVthdB/2Oaqqntm8saqeBK4Frk0yZms+PMnLgCOBec0xnwaeTjIXeEuz2yXAN4GPAnOBS6uqgDua3sreVfXo1ny+JKm9l+xxbAyNJJ9NkhfbZytMAVYC/zPJ3Um+lGRXYK9eYfAYsFezvA/wSK/3L2/aJEmDpM3k+FrgxuYXO0mOTnLbNn7+aGAG8IWqOgT4fzw3LAVA07uoNgdNcmqSxUkWr1y5chtLlCT11ufgqKq/AK4AvtkExuls9kt+KywHllfVd5r1a+gEyeNJ9gZoXp9otq8A9u31/klN2+a1XlRVM6tq5oQJE7axRElSb30OjiSzgPfT6RXsCfxJVf3rtnx4VT0GPJLk15umWcAPgRuBk5u2k4EbmuUbgd9rrq46DFjj/IYkDa6+TI5v9OfAX1bVt5JMB65KcnpVfWMba/gQcHmSHYEHgd+nE2hXJzkFeBh4d7PvPwHHAcuAXzT7SpIGUZ+Do6re2mv5B0mOpXNV1Zu2pYCq6gFmbmHTrC3sW8AHt+XzJEnbpi9fAHyhK6kepfnl/kL7SJK2P336AmCSDyXZr3djM7R0eJJLeG4+QpK0nevLUNUxwB8AVySZAqwGxgKjgEXA31fV3QNXoiRpKHnJ4Kiq9cCFSb5O5/LZPYGnqmr1QBcnSRp62nwB8NqqeqaqHt0YGs0lsZKkEaQvk+PvTnIOMC7Ja5L0fs9FA1eaJGko6sscx2105jT+EDgP+PUkq4GfAE8NYG2SpCGoL3McK4BLkzxQVbcBJBkPTKbzbA5J0gjS5guAt/VaXgWsGpCKJElDWp+DI8lbgZPoXI57D/B94J6q+o8Bqk2SNAS1uVfVxcCHgTHAQcA7gNcC2/ToWEnS8NImOB6uquub5f89EMVIkoa+Nt/juDXJfO9LJUkjW5sex1RgOvDRJEuAHqCnqux9SNII0uaqqncCJNmZ50LkMBy2kqQRpc1VVYuB7wE/aH7+sar+YYDqkiQNUW3mOObQ6V3sCPwX4MdJfjwQRUmShq42Q1U/oXObkX8GSPIa4PgBqkuSNET1uceR5JW916tqKXBgv1ckSRrS2lxVdUXzFMCH6MxxrAGmDUhVkqQhq81Q1Zua73DsT+eKqlcAbx+owiRJQ1Obq6peAcwHJgI/BC6tqp8NVGGSpKGpzVVVVwJrga8BuwDfSvLGAalKkjRktZnjmFBVf9MsL0xyFfAVOl8ClCSNEG16HE8mmb5xpaoepNPzkCSNIG16HB8Ark3yr3Suqnot8MCAVCVJGrLa9DjGAW8CbqEzQd4DvGcgipIkDV1tehyXAjOq6mqAJHsCRwELB6IwSdLQ1KbHsb6q1m9cqaqfAp/o/5IkSUNZm+B4MMmxm7Xt2J/FSJKGvjZDVR8Cvp7kd4E7cHJckkakPvc4qupR4PXAtcAE4PvAeweoLknSENWmx0FV/Sed4Lh2YMqRJA11beY4BkySUUnuTrKwWZ+S5DtJliW5KsmOTftOzfqyZvvkbtYtSSPRkAgO4DRgaa/1zwDnV9WrgJ8BpzTtpwA/a9rPb/aTJA2irgdHkknAbwNfatYDvBW4ptnlEuAdzfLcZp1m+6xmf0nSIOl6cAB/D/wZ8GyzPh5YXVUbmvXlwD7N8j7AIwDN9jXN/r8kyalJFidZvHLlyoGsXZJGnK4GR5LZwBNVtaQ/j1tVF1XVzKqaOWHChP48tCSNeK2uqhoARwBzkhwHjAV2Bz4L7JFkdNOrmASsaPZfAewLLE8yGngZsGrwy5akkaurPY6qWlBVk6pqMnAi8I2qOonOjRSPb3Y7GbihWb6xWafZ/o2qqkEsWZJGvKEwx7ElHwVOT7KMzhzGl5v2LwPjm/bTgTO7VJ8kjVjdHqrapKq+CXyzWX4QeN5jaZubLL5rUAuTJP2SodrjkCQNUQaHJKkVg0OS1IrBIUlqxeCQJLVicEiSWjE4JEmtGBySpFYMDklSKwaHJKkVg0OS1IrBIUlqZcjc5FDS4Dj/pn/btDz/tw7sYiUaruxxSJJaMTgkSa0YHJKkVgwOSVIrBockqRWDQ5LUisEhSWrF4JAktWJwSJJaMTgkSa0YHJKkVgwOSVIrBockqRWDQ5LUisEhSWrF4JAktWJwSJJaMTgkSa0YHJKkVroaHEn2TXJLkh8muTfJaU37K5LclOT+5vXlTXuSXJBkWZLvJ5nRzfolaSTqdo9jA/CnVTUVOAz4YJKpwJnAzVV1AHBzsw5wLHBA83Mq8IXBL1mSRrauBkdVPVpVdzXLa4GlwD7AXOCSZrdLgHc0y3OBS6vjDmCPJHsPctmSNKJ1u8exSZLJwCHAd4C9qurRZtNjwF7N8j7AI73etrxp2/xYpyZZnGTxypUrB6xmSRqJhkRwJNkNuBb4cFX9vPe2qiqg2hyvqi6qqplVNXPChAn9WKkkqevBkWQMndC4vKq+2jQ/vnEIqnl9omlfAezb6+2TmjZJ0iDp9lVVAb4MLK2q83ptuhE4uVk+GbihV/vvNVdXHQas6TWkJUkaBKO7/PlHAL8L/CBJT9P2MeAc4OokpwAPA+9utv0TcBywDPgF8PuDW64kqavBUVXfAvICm2dtYf8CPjigRUmSXlTX5zgkScOLwSFJasXgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktSKwSFJasXgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktSKwSFJasXgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktSKwSFJasXgkCS1YnBIkloxOCRJrRgckqRWDA5JUisGhySpFYNDktTKsAyOJMck+VGSZUnO7HY9kjSSDLvgSDIK+DxwLDAVeE+Sqd2tSpJGjmEXHMAbgWVV9WBVPQ1cCcztck2SNGKM7nYBW2Ef4JFe68uBQ3vvkORU4NRmdV2SHw1Sbf1pT+Cn3S5ikI3Mc/7Dv+vaOZ/enY8daf/Pw/V8X/lCG4ZjcLykqroIuKjbdWyLJIurama36xhMnvPIMNLOeXs83+E4VLUC2LfX+qSmTZI0CIZjcHwXOCDJlCQ7AicCN3a5JkkaMYbdUFVVbUjyx8C/AKOAi6vq3i6XNRCG9VDbVvKcR4aRds7b3fmmqrpdgyRpGBmOQ1WSpC4yOCRJrRgcw0CSP01SSfbsdi0DLcm5Se5L8v0k1yXZo9s1DYSRdtucJPsmuSXJD5Pcm+S0btc0WJKMSnJ3koXdrqW/GBxDXJJ9gbcB/97tWgbJTcC0qjoI+DdgQZfr6Xcj9LY5G4A/raqpwGHAB0fAOW90GrC020X0J4Nj6Dsf+DNgRFzFUFWLqmpDs3oHne/pbG9G3G1zqurRqrqrWV5L5xfpPt2tauAlmQT8NvClbtfSnwyOISzJXGBFVX2v27V0yR8AX+92EQNgS7fN2e5/iW6UZDJwCPCd7lYyKP6ezh9+z3a7kP407L7Hsb1J8n+AX9nCpj8HPkZnmGq78mLnXFU3NPv8OZ3hjcsHszYNrCS7AdcCH66qn3e7noGUZDbwRFUtSfKWbtfTnwyOLquqo7bUnmQ6MAX4XhLoDNncleSNVfXYIJbY717onDdKMg+YDcyq7fOLRiPytjlJxtAJjcur6qvdrmcQHAHMSXIcMBbYPcllVfW+Lte1zfwC4DCR5MfAzKoajnfZ7LMkxwDnAb9RVSu7Xc9ASDKazsT/LDqB8V3gvdvpHRAASOevn0uAJ6vqw92uZ7A1PY4zqmp2t2vpD85xaKj5HDAOuClJT5Ivdrug/tZM/m+8bc5S4OrtOTQaRwC/C7y1+X/taf4S1zBkj0OS1Io9DklSKwaHJKkVg0OS1IrBIUlqxeCQJLVicEiSWjE4JEmtGBxSFzTPpvitZvmTSf5bt2uS+sp7VUndcRbwiSQT6dwpdk6X65H6zG+OS12S5P8CuwFvaZ5RIQ0LDlVJXdDc/Xhv4GlDQ8ONwSENsiR703nOyFxgXXNHYGnYMDikQZRkF+CrdJ6/vRT4KzrzHdKw4RyHJKkVexySpFYMDklSKwaHJKkVg0OS1IrBIUlqxeCQJLVicEiSWvn/2l8ozEnoX6kAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 200 Loss D.: 0.6585134267807007\n",
            "Epoch: 200 Loss G.: 0.8182554841041565\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 250 Loss D.: 0.6343465447425842\n",
            "Epoch: 250 Loss G.: 0.8828651905059814\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 300 Loss D.: 0.6180165410041809\n",
            "Epoch: 300 Loss G.: 0.8914716243743896\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 350 Loss D.: 0.67656010389328\n",
            "Epoch: 350 Loss G.: 0.7469807267189026\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 400 Loss D.: 0.6735788583755493\n",
            "Epoch: 400 Loss G.: 0.7817835211753845\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAbt0lEQVR4nO3df5QV5Z3n8fdHQDsChgSBYWm0iYExCIqkjRhzHGNrBAfBnBiDMSMYNkxWN2ucaAJmJsk4ZmPGrEY30QwbnYERow6GH2GSLC5j1g3GGMDWqGjAH4Qm/GjRRnqEaI/f/eM+jZe2kL7Qt+6l+/M6p8+teurH/Rac05+up6qeUkRgZmbW0WGVLsDMzKqTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMEklPSTqz0nWYVQsHhPUYkl6UdHaHthmSfgkQESdExC/2s486SSGpdxlLNasKDgizKuLgsWrigDBLis8wJH1I0ipJr0raKummtNpD6bNFUquk0yQdJumvJW2QtE3SfEnvLtrvpWnZdkl/0+F7viFpoaS7JL0KzEjf/StJLZI2S/qepMOL9heSLpe0TtJOSX8n6ThJD6d67yte3+xAOSDMst0C3BIRRwHHAfel9jPS54CI6BcRvwJmpJ+PAu8D+gHfA5A0GrgNuAQYCrwbGNbhu6YCC4EBwALgP4CrgKOB04AG4PIO25wLfBCYAHwZmAt8BhgOjAEuPohjNwMcENbzLE5/mbdIaqHwyzvLG8D7JR0dEa0R8cg77PMS4KaIeD4iWoE5wLTUXXQh8JOI+GVEvA58Deg4ANqvImJxRLwZEbsiYnVEPBIRbRHxIvAPwJ912ObvI+LViHgKeBJYnr5/B/Az4OTO/5OYZXNAWE9zQUQMaP/h7X+Zt5sJjAKekfQbSZPfYZ//CdhQNL8B6A0MScs2ti+IiNeA7R2231g8I2mUpGWStqRup/9O4Wyi2Nai6V0Z8/3eoV6zTnFAmGWIiHURcTEwGPg2sFBSX97+1z/AH4Bji+aPAdoo/NLeDNS2L5D0LmBgx6/rMH878AwwMnVxXQvowI/G7MA4IMwySPqMpEER8SbQkprfBJrT5/uKVv8RcJWkEZL6UfiL/96IaKNwbeF8SR9OF46/wf5/2fcHXgVaJR0P/JeuOi6zUjggzLJNBJ6S1ErhgvW0dH3gNeCbwMp0HWMCcCfwzxTucHoB2A18ASBdI/gCcA+Fs4lWYBvwx3f47quBTwM7gf8F3Nv1h2e2f/ILg8zyk84wWih0H71Q6XrM3onPIMzKTNL5ko5M1zC+A/wWeLGyVZntnwPCrPymUriQ/QdgJIXuKp+6W9VzF5OZmWXyGYSZmWXqNgODHX300VFXV1fpMszMDimrV69+KSIGZS3rNgFRV1fHqlWrKl2GmdkhRdKGfS1zF5OZmWVyQJiZWSYHhJmZZeo21yDMrPt44403aGpqYvfu3ZUupduoqamhtraWPn36dHobB4SZVZ2mpib69+9PXV0dkgeyPVgRwfbt22lqamLEiBGd3s5dTGZWdXbv3s3AgQMdDl1EEgMHDiz5jMwBYWZVyeHQtQ7k39MBYWZmmXK5BiHpT9l7TPv3UXg37/zUXkdhdMuLIuIVFaLuFuA84DVgRkSsyaNWM6s+Nz/wuy7d31XnjNrnsu3bt9PQ0ADAli1b6NWrF4MGFR40fvTRRzn88MO7tJZqlktARMSzwDgASb2ATcAiYDawIiJukDQ7zX8FmERh1MuRwKkUXsF4ah61mnWl4l9s7/RLyarHwIEDaWxsBOAb3/gG/fr14+qrr96zvK2tjd69e8b9PZU4ygbguYjYIGkqcGZqnwf8gkJATAXmpyGRH5E0QNLQiNhcgXrNrIebMWMGNTU1PPbYY5x++ukcddRRewXHmDFjWLZsGXV1ddx1113ceuutvP7665x66qncdttt9OrVq8JHcGAqcQ1iGoV3+AIMKfqlvwUYkqaHARuLtmlKbXuRNEvSKkmrmpuby1WvmRlNTU08/PDD3HTTTftcZ+3atdx7772sXLmSxsZGevXqxYIFC3KssmvlegaRXto+BZjTcVlEhKSSXk4REXOBuQD19fV+sYWZlc0nP/nJ/Z4JrFixgtWrV3PKKacAsGvXLgYPHpxHeWWRdxfTJGBNRGxN81vbu44kDaXwMncoXKMYXrRdbWozM6uIvn377pnu3bs3b7755p759ucLIoLp06fzrW99K/f6yiHvLqaLeat7CWApMD1NTweWFLVfqoIJwA5ffzCzalFXV8eaNYUbK9esWcMLL7wAQENDAwsXLmTbtsLfui+//DIbNuxzNO2ql9sZRHph+znAXxY13wDcJ2kmsAG4KLX/lMItrusp3OZ6WV51mln1qbY7wD7xiU8wf/58TjjhBE499VRGjSrUN3r0aK6//no+9rGP8eabb9KnTx++//3vc+yxx1a44gPTbd5JXV9fH35hkFUb3+Z6YNauXcsHPvCBSpfR7WT9u0paHRH1Wev7SWozM8vkgDAzs0wOCDMzy+SAMDOzTA4IMzPL1DNGnDKrkAm/n1s0952K1WF2IBwQZlb9HuziJ5M/+rbRfvbSq1cvxo4du2d+2rRpzJ49O3PdxYsXM2rUKEaPHg3A1772Nc444wzOPvvsgyqxpaWFu+++m8svv7yk7bJGoD1QDggzsw7e9a537Rnye38WL17M5MmT9wTEdddd1yU1tLS0cNttt5UcEF3J1yDMzDpp9uzZjB49mhNPPJGrr76ahx9+mKVLl3LNNdcwbtw4nnvuOWbMmMHChQuBwpAcc+bMYdy4cdTX17NmzRrOPfdcjjvuOH7wgx8A0NraSkNDA+PHj2fs2LEsWbJkz3c999xzjBs3jmuuuQaAG2+8kVNOOYUTTzyRr3/963vq+uY3v8moUaP4yEc+wrPPPttlx+szCDOzDnbt2sW4ceP2zM+ZM4ezzz6bRYsW8cwzzyCJlpYWBgwYwJQpU5g8eTIXXnhh5r6OOeYYGhsbueqqq5gxYwYrV65k9+7djBkzhs9//vPU1NSwaNEijjrqKF566SUmTJjAlClTuOGGG3jyySf3nMksX76cdevW8eijjxIRTJkyhYceeoi+fftyzz330NjYSFtbG+PHj+eDH/xgl/w7OCDMzDrI6mJqa2ujpqaGmTNnMnnyZCZPntypfU2ZMgWAsWPH0traSv/+/enfvz9HHHEELS0t9O3bl2uvvZaHHnqIww47jE2bNrF169a37Wf58uUsX76ck08+GSiceaxbt46dO3fy8Y9/nCOPPHKv7+sK7mIyM+uE3r178+ijj3LhhReybNkyJk6c2KntjjjiCAAOO+ywPdPt821tbSxYsIDm5mZWr15NY2MjQ4YM2TN8eLGIYM6cOTQ2NtLY2Mj69euZOXNm1xzcPjggzMw6obW1lR07dnDeeedx88038/jjjwPQv39/du7cecD73bFjB4MHD6ZPnz48+OCDe4YH77jfc889lzvvvJPW1lYANm3axLZt2zjjjDNYvHgxu3btYufOnfzkJz85iKPcm7uYzKz67ee21K7W8RrExIkTufLKK5k6dSq7d+8mIva8enTatGl87nOf49Zbb91zcboUl1xyCeeffz5jx46lvr6e448/HoCBAwdy+umnM2bMGCZNmsSNN97I2rVrOe200wDo168fd911F+PHj+dTn/oUJ510EoMHD97zNruu4OG+zcroV3e8dS/6aTP9oFxnebjv8vBw32Zm1iUcEGZmlskBYWZVqbt0f1eLA/n3dECYWdWpqalh+/btDokuEhFs376dmpqakrbzXUxmVnVqa2tpamqiubm50qV0GzU1NdTW1pa0TW4BIWkA8ENgDBDAZ4FngXuBOuBF4KKIeEWSgFuA84DXgBkRsSavWs2ssvr06cOIESMqXUaPl2cX0y3AzyPieOAkYC0wG1gRESOBFWkeYBIwMv3MAm7PsU4zMyOngJD0buAM4A6AiHg9IlqAqcC8tNo84II0PRWYHwWPAAMkDc2jVjMzK8jrDGIE0Az8o6THJP1QUl9gSERsTutsAYak6WHAxqLtm1KbmZnlJK+A6A2MB26PiJOBf+et7iQAonC7Qkm3LEiaJWmVpFW+mGVm1rXyCogmoCkifp3mF1IIjK3tXUfpc1tavgkYXrR9bWrbS0TMjYj6iKgfNGhQ2Yo3M+uJcgmIiNgCbJT0p6mpAXgaWApMT23TgSVpeilwqQomADuKuqLMzCwHeT4H8QVggaTDgeeByygE1H2SZgIbgIvSuj+lcIvregq3uV6WY51mZkaOARERjUDWiIENGesGcEXZizIzs33yUBtmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlim3gJD0oqTfSmqUtCq1vVfSA5LWpc/3pHZJulXSeklPSBqfV51mZlaQ9xnERyNiXETUp/nZwIqIGAmsSPMAk4CR6WcWcHvOdZqZ9XiV7mKaCsxL0/OAC4ra50fBI8AASUMrUaCZWU+VZ0AEsFzSakmzUtuQiNicprcAQ9L0MGBj0bZNqW0vkmZJWiVpVXNzc7nqNjPrkXrn+F0fiYhNkgYDD0h6pnhhRISkKGWHETEXmAtQX19f0rZmZvbOcjuDiIhN6XMbsAj4ELC1vesofW5Lq28ChhdtXpvazMwsJ7kEhKS+kvq3TwMfA54ElgLT02rTgSVpeilwabqbaQKwo6gryszMcpBXF9MQYJGk9u+8OyJ+Luk3wH2SZgIbgIvS+j8FzgPWA68Bl+VUp5mZJbkEREQ8D5yU0b4daMhoD+CKHEozM7N9qPRtrmZmVqUcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZSg6INDJrr3IUY2Zm1WO/ASHpMEmflvSvkrYBzwCbJT0t6UZJ7y9/mWZmlrfOnEE8CBwHzAH+JCKGR8Rg4CPAI8C3JX2mjDWamVkFdGa477Mj4o2OjRHxMnA/cL+kPl1emZmZVdR+zyDaw0HSLUpv/NnXOmZm1n2UcpF6J7A0vTIUSedKWlmesszMrNI6/Ua5iPhrSZ8GfiHpdaAVmF22yszMrKI6HRCSGoDPAf8ODAU+GxHPlqswMzOrrFK6mL4K/E1EnAlcCNwr6ayyVGVmZhXX6YCIiLMi4pdp+rfAJOD6Ur5MUi9Jj0laluZHSPq1pPWS7pV0eGo/Is2vT8vrSvkeMzM7eJ15UG5fdy5tBhreaZ0MVwJri+a/DdwcEe8HXgFmpvaZwCup/ea0npmZ5ahTD8pJ+oKkY4ob01/7p0maB0zf304k1QJ/DvwwzQs4C1iYVpkHXJCmp6Z50vKGEkLIzMy6QGcuUk8EPgv8SNIIoAWoAXoBy4HvRsRjndjPd4EvA/3T/ECgJSLa0nwTMCxNDwM2AkREm6Qdaf2XincoaRYwC+CYY/bKLzMzO0j7DYiI2A3cJulnFH6JHw3sioiWzn6JpMnAtohYLenMAy02o7a5wFyA+vr66Kr9mplZaXcx3R8Rb0TE5vZwkDShk9ueDkyR9CJwD4WupVuAAZLaQ6oW2JSmNwHD03f0Bt4NbC+hVjMzO0iduUh9kaQbgP6SPiCpeJu5nfmSiJgTEbURUQdMA/4tIi6hMBDghWm16cCSNL2Ut65rXJjW9xmCmVmOOnMGsRJ4GngPcBOwXtKadKvqroP8/q8AfyVpPYVrDHek9juAgan9r/AT22ZmuevMNYhNwHxJz0XESgBJA4E6Cu+GKElE/AL4RZp+HvhQxjq7gU+Wum8zM+s6pYzFtLJoeju+JmBm1q2VMhbTWcAlFG5zfRJ4AngyIv5YptrMzKyCOh0QwJ3AF4E+wIkUHmo7AfArR83MuqFSAmJDRCxO0/9SjmLMzKx6lPIcxEOSrvKQF2ZmPUMpZxCjgbHAVyStBhqBxojw2YSZWTdUyl1MnwCQ9C7eCosJuLvJzKxbKuUuplXA48Bv08+/RsQ/lakuMzOrsFKuQUyhcLZwOPCXwItpbCUzM+uGSuli+gPwB+DnAJI+wFvjKJmZWTfT6TMISccWz0fEWmBUl1dkZmZVoZS7mH6U3ir3AoVrEDuAMWWpyszMKq6ULqYPp2cgjqNwB9N7gfPLVZiZmVVWKXcxvRe4ChhMYfjv+RHxSrkKMzOzyirlLqZ7gJ3AT4AjgV9KettQ3WZm1j2Ucg1iUET8fZpeJule4G4KD8uZmVk3U8oZxMuSxrbPpJf9HNn1JZmZWTUo5QzicuB+Sf+Pwl1MJwDPlaUqMzOruFLOIPoDHwYepHChuhG4uBxFmZlZ5ZVyBjEfGB8R9wFIOho4G1hWjsLM7ODc/MDv9kxfdY6fabXSlXIGsTsidrfPRMRLwHWd2VBSjaRHJT0u6SlJf5vaR0j6taT1ku6VdHhqPyLNr0/L60qo08zMukApAfG8pEkd2g7v5LZ/BM6KiJOAccBESROAbwM3R8T7gVeAmWn9mcArqf3mtJ6ZmeWolID4AvAtSXdL+m+S/oFOXqSOgtY02yf9BHAWsDC1z6PwnmuAqWmetLzBb7IzM8tXpwMiIjYDHwTuBwYBTwCf7uz2knpJagS2AQ9QCJeWiGhLqzQBw9L0MGBj+t42CuM+DczY5yxJqyStam5u7mwpZmbWCaVcpCYi/oNCQNxf6helbcdJGgAsAo4vdR8Z+5wLzAWor6+Pg92fmZm9pZQupi4RES0UbpU9DRggqT2kaoFNaXoTMBwgLX83sD3nUs3MerRcAkLSoHTm0P5O63OAtRSCov2lQ9OBJWl6aZonLf+3iPAZgplZjkrqYjoIQ4F5knpRCKX7ImKZpKeBeyRdDzwG3JHWvwP4Z0nrgZeBaTnVaWZmSS4BERFPACdntD8PvG1E2PS8xSdzKM3MzPYh92sQZmZ2aHBAmJlZJgeEmZllckCYmVkmB4SZmWVyQJiZWSYHhJmZZcrrQTkzy9mE388tmvtOxeqwQ5fPIMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDLlEhCShkt6UNLTkp6SdGVqf6+kByStS5/vSe2SdKuk9ZKekDQ+jzrNzOwteZ1BtAFfiojRwATgCkmjgdnAiogYCaxI8wCTgJHpZxZwe051mplZkktARMTmiFiTpncCa4FhwFRgXlptHnBBmp4KzI+CR4ABkobmUauZmRXkfg1CUh1wMvBrYEhEbE6LtgBD0vQwYGPRZk2preO+ZklaJWlVc3Nz2Wo2M+uJcg0ISf2A+4EvRsSrxcsiIoAoZX8RMTci6iOiftCgQV1YqZmZ5RYQkvpQCIcFEfHj1Ly1vesofW5L7ZuA4UWb16Y2MzPLSV53MQm4A1gbETcVLVoKTE/T04ElRe2XpruZJgA7irqizMwsB3m9k/p04C+A30pqTG3XAjcA90maCWwALkrLfgqcB6wHXgMuy6lOMzNLcgmIiPgloH0sbshYP4ArylqUmZm9Iz9JbWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWyQFhZmaZHBBmZpbJAWFmZpkcEGZmlskBYWZmmRwQZmaWKZeAkHSnpG2Snixqe6+kByStS5/vSe2SdKuk9ZKekDQ+jxrNzGxveZ1B/BMwsUPbbGBFRIwEVqR5gEnAyPQzC7g9pxrNzKxILgEREQ8BL3dongrMS9PzgAuK2udHwSPAAElD86jTzMzeUslrEEMiYnOa3gIMSdPDgI1F6zWlNjMzy1FVXKSOiACi1O0kzZK0StKq5ubmMlRmZtZzVTIgtrZ3HaXPbal9EzC8aL3a1PY2ETE3Iuojon7QoEFlLdbMrKepZEAsBaan6enAkqL2S9PdTBOAHUVdUWZmlpPeeXyJpB8BZwJHS2oCvg7cANwnaSawAbgorf5T4DxgPfAacFkeNZqZ2d5yCYiIuHgfixoy1g3givJWZGZm+1MVF6nNzKz6OCDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDI5IMzMLJMDwszMMjkgzMwskwPCzMwyOSDMzCyTA8LMzDL1rnQB+yJpInAL0Av4YUTcUOGSzA7Og9/a/zofnXPg25p1saoMCEm9gO8D5wBNwG8kLY2IpytbmVmZOQisilRrF9OHgPUR8XxEvA7cA0ytcE1mZj1KVZ5BAMOAjUXzTcCpHVeSNAuYlWZbJT2bQ21d7WjgpUoXkbOedsyVP97//D/y/sbKH3P+DtVjPnZfC6o1IDolIuYCcytdx8GQtCoi6itdR5562jH3tOMFH3N3Ua1dTJuA4UXztanNzMxyUq0B8RtgpKQRkg4HpgFLK1yTmVmPUpVdTBHRJum/Av+bwm2ud0bEUxUuq1wO6S6yA9TTjrmnHS/4mLsFRUSlazAzsypUrV1MZmZWYQ4IMzPL5ICoIpK+JCkkHV3pWspJ0o2SnpH0hKRFkgZUuqZykTRR0rOS1kuaXel6yk3ScEkPSnpa0lOSrqx0TXmR1EvSY5KWVbqWruKAqBKShgMfA35f6Vpy8AAwJiJOBH4H7GMAokNb0ZAxk4DRwMWSRle2qrJrA74UEaOBCcAVPeCY210JrK10EV3JAVE9bga+DHT7uwYiYnlEtKXZRyg859Id9bghYyJic0SsSdM7KfzCHFbZqspPUi3w58APK11LV3JAVAFJU4FNEfF4pWupgM8CP6t0EWWSNWRMt/9l2U5SHXAy8OvKVpKL71L4A+/NShfSlaryOYjuSNL/Af4kY9FXgWspdC91G+90vBGxJK3zVQpdEgvyrM3KT1I/4H7gixHxaqXrKSdJk4FtEbFa0pmVrqcrOSByEhFnZ7VLGguMAB6XBIXuljWSPhQRW3IssUvt63jbSZoBTAYaovs+jNMjh4yR1IdCOCyIiB9Xup4cnA5MkXQeUAMcJemuiPhMhes6aH5QrspIehGoj4hDcVTITkkvg7oJ+LOIaK50PeUiqTeFi/ANFILhN8Cnu/GoAKjwV8484OWI+GKl68lbOoO4OiImV7qWruBrEFYJ3wP6Aw9IapT0g0oXVA7pQnz7kDFrgfu6czgkpwN/AZyV/m8b01/WdgjyGYSZmWXyGYSZmWVyQJiZWSYHhJmZZXJAmJlZJgeEmZllckCYmVkmB4SZmWVyQJiVSXovwjlp+npJ/7PSNZmVwmMxmZXP14HrJA2mMKrplArXY1YSP0ltVkaS/i/QDzgzvR/B7JDhLiazMkkj9Q4FXnc42KHIAWFWBpKGUnjPxVSgNY1ga3ZIcUCYdTFJRwI/pvBu5rXA31G4HmF2SPE1CDMzy+QzCDMzy+SAMDOzTA4IMzPL5IAwM7NMDggzM8vkgDAzs0wOCDMzy/T/AdtYBCyiWvLhAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch: 450 Loss D.: 0.6790764331817627\n",
            "Epoch: 450 Loss G.: 0.7356427907943726\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2so1hV6mUgst"
      },
      "source": [
        "Once the GAN has converged, we can generate samples using the generator and plot their histogram and compare it to the true distribution. Notice that the generator is simply mapping a single input (Gaussian random input) into a Heads (1.0) or Tails (0.0) by implicitly modelling the underlying generative process using its interaction with the discriminator. Below, we can see this mapping for the trained GAN. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 904
        },
        "id": "TBT1H6CWBLVy",
        "outputId": "c9105a52-0a7b-4384-bfcb-6bf3d35f9074"
      },
      "source": [
        "plt.figure()\r\n",
        "latent_space_samples = torch.randn(1000, 1)\r\n",
        "generated_samples = generator(latent_space_samples)\r\n",
        "latent_space_samples = latent_space_samples.detach()[:,0]\r\n",
        "generated_samples = np.round(generated_samples.detach()[:,0])\r\n",
        "plt.hist(train_data[:,0],bins=bins,alpha=0.5, label='True')\r\n",
        "plt.hist(generated_samples,bins=bins,alpha=0.5, label='Estimated')\r\n",
        "plt.title('Histogram');plt.xlabel('$x$');plt.ylabel('$count(x)$');plt.legend()\r\n",
        "plt.figure()\r\n",
        "plt.plot(latent_space_samples,generated_samples,'o')\r\n",
        "plt.xlabel('prior');plt.ylabel('output');plt.title('Learned density mapping')\r\n",
        "plt.figure()\r\n",
        "plt.hist(latent_space_samples[generated_samples==0.0],cumulative=True,bins=bins,alpha=0.5, label='Generates 0')\r\n",
        "plt.hist(latent_space_samples[generated_samples==1.0],cumulative=True,bins=bins,alpha=0.5, label='Generates 1')\r\n",
        "#plt.hist(latent_space_samples,cumulative=True,bins=bins,alpha=0.5, label='Cumulative')\r\n",
        "plt.title('Histogram');plt.xlabel('$z$');plt.ylabel('$count(z)$');plt.legend()\r\n",
        "\r\n",
        "plt.xlabel('prior');plt.ylabel('count');plt.title('Learned density mapping')\r\n",
        "plt.show()\r\n",
        "print('Generated Samples',torch.round(generated_samples[:50].T))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3df5xWZZ3/8ddbYCQVBRVJGRT7apqUgQ1q4ZIrKqImlr81A8Vl3WXJH5lRPb6b7ba76Vam5mKsmliWGqaQS/7IpPIHJiiZij9QIYZFRFREDXHos3+ca443wwzcwJz7nrnn/Xw87secc53rnPO575nH/Znrus65jiICMzMzgK2qHYCZmXUcTgpmZpZzUjAzs5yTgpmZ5ZwUzMws56RgZmY5JwWraZJukPStMusOlBSSuhcYzxmS7inq+NUi6VeSxlQ7DttyTgpWNkkLJR1e7Tg6s4i4KSKObF5PSWivasbUHiJiVERMrXYctuWcFKxDkdSt2jGYdWVOCrbFJG0laZKkFyStkHSrpB1Ltv9c0suSVkr6naRBJdtukDRZ0kxJbwN/m1okF0l6Iu1zi6SeJfscK2mepDckPSRp/5JtQyQ9JmmVpFuAfL9W4u4m6TuSXpX0InBMi+07SLpO0lJJSyR9qzlpSRor6YG0/+uSXpI0qmTfsZJeTHG8JOmM0v3S8u9S9T9KekvSKZKelPSZkuP0SPENaSX+QyU1SrpY0ispzuMlHS3pOUmvSfpaSf0DJT2cPrelkn4gqa5ke0j6Yor7VUn/KWmrkrgfTPuslPSMpBEl+86SdE6Zn82e6e9glaRfS7pa0k/a+j1ZZTkpWHuYCBwPfBrYDXgduLpk+6+AvYFdgMeAm1rsfzrwb0Av4IFUdjJwFLAnsD8wFrIvfeB64O+BnYAfAjMkbZ2+4O4AfgzsCPwcOGEDcf8dcCwwBGgATmyx/QagCdgr1TkSOKdk+0HAs8DOwGXAdcpsC1wJjIqIXsCngHktTx4Rw9PixyNiu4i4BbgR+HxJtaOBpRHxeBvv4YNkia8/8M/Af6f9PwH8DfD/Je2Z6q4FLkjxfhIYAfxji+N9Nn0WBwCjgbNbvN8X0v7fAH5RmvxbaPWzSdt+CvyB7Pd3CXBmG8ewaogIv/wq6wUsBA5vpXw+MKJkfVfgPaB7K3V7AwHskNZvAG5s5TyfL1m/DLgmLU8G/rVF/WfJEtJw4H8BlWx7CPhWG+/nN8C5JetHpti6A/2Ad4EPlGw/Dbg/LY8FFpRs2ybt+0FgW+ANsoT0gRbnHAs8ULIewF4l67sBq4Dt0/o04OI24j8U+AvQLa33Ssc7qKTOXOD4NvY/H7i9RSxHlaz/I3BfSdwtP9s/AGem5VnAOWV8NruTJdptSrb/BPhJtf++/cpebilYe9gDuD11S7xBliTWAv1SF823U9fSm2Rf+JD9B9lscSvHfLlk+R1gu5Jzfan5XOl8A8i+THcDlkT6pkkWbSDu3Vqcu7TuHkAPYGnJeX5I1tpZL8aIeCctbhcRbwOnAOem/f9H0r4biCMXEf8LPAicIKk3MIr1W1alVkTE2rT8l/RzWcn2v5A+O0kflnRn6sp7E/h31v09wPqfx24l6619tqXbS7X62aT6r5WUtTynVZmTgrWHxWRdJb1LXj0jYglZ19Bo4HBgB2Bg2kcl+2/KVL2LgX9rca5tIuJnwFKgf0k3BWT/mbZlKVlCaa3uYrKWws4l59k+IgZRhoi4OyKOIGs1PUPWrVOuqWRdQCcBD6fPsT1MTrHsHRHbA19j3d8DrP95/G/Jemufben2ciwFdpS0TRvntCpzUrBN1UNSz5JXd+Aa4N8k7QEgqa+k0al+L7Iv1xVk3Qj/voXn/2/gXEkHNfffSzpGUi/gYbKuiS+mAdrPAQdu4Fi3prr1kvoAk5o3RMRS4B7gu5K2VzaY/v8kfXpjAUrqJ2l0Glt4F3gL+Gsb1ZcBH2pRdgdZn/55ZGMM7aUX8CbwVmq5/EMrdb4sqY+kAen8t5Rs24X3P9uTgI8AMzclgIhYBMwBLpFUJ+mTwGc2sptVkJOCbaqZZF0Sza9LgCuAGcA9klYBs8kGGiH7UlsELAGeTts2W0TMIRsg/gHZgPYC0iB0RKwBPpfWXyPrwvnFBg7338DdwB/JBsBb1v0CUJfifp2sf3/XMsLcCriQ7L/o18jGO1r7Aobs85uauqhOTu/jL8BtZIPsG4p/U11E1nJbRfbeb2mlznSycYh5wP8A15Vse4TsgoFXyS4MODEiVmxGHGeQDXSvAL6V4nh3M45jBdC6XYRm1hFI+mfgwxHx+Y1Wbr9zBlnX0oJWto0lG0g+pIDz3gI8ExHfaO9j26ZzS8Gsg0mXeY4DplQ7liJIGpq64raSdBTZmNMd1Y7LMk4KZh2IpL8jG+T+VUT8bmP1O6kPkl3C+hbZ/Rz/EG3fh2EV5u4jMzPLuaVgZma5wqYIroSdd945Bg4cWO0wzMw6lblz574aEX1b29apk8LAgQOZM2dOtcMwM+tUJLV5p7+7j8zMLOekYGZmOScFMzPLdeoxhda89957NDY2snr16mqH0iX17NmT+vp6evToUe1QzGwz1FxSaGxspFevXgwcOJB1J3S0okUEK1asoLGxkT333HPjO5hZh1Nz3UerV69mp512ckKoAknstNNObqWZdWI1lxQAJ4Qq8mdv1rnVZFIwM7PNU3NjCi1dfu9z7Xq8C4748EbrLFu2jAsuuIDZs2fTp08f6urquPjii/nsZz/brrGUa9asWdTV1fGpT31qi4911113cd5557F27VrOOeccJk2atPGdzKzTqPmkUGkRwfHHH8+YMWP46U9/CsCiRYuYMWNGoedtamqie/fWf52zZs1iu+222+KksHbtWiZMmMC9995LfX09Q4cO5bjjjmO//fbbouOalWrvf+RqVTn/oG6OwrqPJO0jaV7J601J50vaUdK9kp5PP/uk+pJ0paQFkp6QdEBRsRXpN7/5DXV1dZx77rl52R577MHEiROB7Iv1y1/+MkOHDmX//ffnhz/8IZB9cR966KGceOKJ7Lvvvpxxxhk0z2A7d+5cPv3pT/OJT3yCkSNHsnTpUgAOPfRQzj//fBoaGrjiiiv45S9/yUEHHcSQIUM4/PDDWbZsGQsXLuSaa67h8ssvZ/Dgwfz+979n+fLlnHDCCQwdOpShQ4fy4IMPAvDb3/6WwYMHM3jwYIYMGcKqVavWeW9/+MMf2GuvvfjQhz5EXV0dp556KtOnTy/8M7Xad/m9z+Uvq67CWgoR8SwwGEBSN7LHMd5O9hzc+yLi25ImpfWvAKPIHvW3N9mjHCfz/iMdO42nnnqKAw5oO59dd9117LDDDjz66KO8++67DBs2jCOPPBKAxx9/nKeeeorddtuNYcOG8eCDD3LQQQcxceJEpk+fTt++fbnlllv4+te/zvXXXw/AmjVr8vmfXn/9dWbPno0krr32Wi677DK++93vcu6557Lddttx0UUXAXD66adzwQUXcMghh/DnP/+ZkSNHMn/+fL7zne9w9dVXM2zYMN566y169uy5TuxLlixhwID3n7FeX1/PI4880q6fn3UdTgAdU6W6j0YAL0TEovRA90NT+VSyh218hezpSzdG9u/xbEm9Je2aHqDeaU2YMIEHHniAuro6Hn30Ue655x6eeOIJpk2bBsDKlSt5/vnnqaur48ADD6S+vh6AwYMHs3DhQnr37s2TTz7JEUccAWQtjV13ff8xwaecckq+3NjYyCmnnMLSpUtZs2ZNm/cK/PrXv+bpp5/O1998803eeusthg0bxoUXXsgZZ5zB5z73uTwWs/biRNDxVSopnAr8LC33K/mifxnol5b7kz1xqlljKlsnKUgaD4wH2H333YuKd7MNGjSI2267LV+/+uqrefXVV2loaACyMYerrrqKkSNHrrPfrFmz2HrrrfP1bt260dTUREQwaNAgHn744VbPt+222+bLEydO5MILL+S4445j1qxZXHLJJa3u89e//pXZs2ev1xKYNGkSxxxzDDNnzmTYsGHcfffd7Lvvvvn2/v37s3jx+7+ixsZG+vfvv5FPxLoaf/F3boVfkiqpDjgO+HnLbalVsEmPfouIKRHREBENffu2Oh14VR122GGsXr2ayZMn52XvvPNOvjxy5EgmT57Me++9B8Bzzz3H22+/3ebx9tlnH5YvX54nhffee4+nnnqq1borV67Mv6SnTp2al/fq1Wud8YEjjzySq666Kl+fN28eAC+88AIf+9jH+MpXvsLQoUN55pln1jn+0KFDef7553nppZdYs2YNN998M8cdd9yGPxDrEjwmUDsq0VIYBTwWEcvS+rLmbiFJuwKvpPIlwICS/epT2RYpaoS+LZK44447uOCCC7jsssvo27cv2267LZdeeikA55xzDgsXLuSAAw4gIujbty933NH2M8vr6uqYNm0aX/ziF1m5ciVNTU2cf/75DBo0aL26l1xyCSeddBJ9+vThsMMO46WXXgLgM5/5DCeeeCLTp0/nqquu4sorr2TChAnsv//+NDU1MXz4cK655hq+//3vc//997PVVlsxaNAgRo0atc7xu3fvzg9+8ANGjhzJ2rVrOfvss1uNw8w6r8Kf0SzpZuDuiPhRWv9PYEXJQPOOEXGxpGOAfwKOJhtgvjIiDtzQsRsaGqLlQ3bmz5/PRz7ykSLeipXJv4Ouwa2C6tqSf3glzY2Ihta2FdpSkLQtcATw9yXF3wZulTQOWAScnMpnkiWEBcA7wFlFxmZmm86JoPYVmhQi4m1gpxZlK8iuRmpZN4AJRcZjZmYb5juazWyD3DroWjwhnpmZ5ZwUzMws56RgZma52h9TuP8/2vd4f/vVjVap5amzzz77bO6880522WUXnnzyyXaIzjoijyN0XW4ptLPmqbOHDx/Oiy++yNy5c7n55ptpbGws9LxNTU1tbps1axYPPfRQu5xn7Nix3HXXXe1yLDPreJwU2lktT50NMHz4cHbcccdCP0OrDk9VYdAVuo8qrJanzjaz2uekUDBPnW1mnYmTQjur5amzzaz2eUyhndXy1NlWezyOYC3VfkuhjEtI21MtT50NcNpppzFr1ixeffVV6uvr+eY3v8m4cePa6dMzs2orfOrsInnq7I7Jv4POwy2EzquoqbPdfWRmZrna7z4ys3W4dWAbUpMthc7cJdbZ+bM369xqLin07NmTFStW+MupCiKCFStW+KY3s06s5rqP6uvraWxsZPny5dUOpUvq2bOnb3oz68RqLin06NGjzTt5zcxsw2qu+8jMzDZfoS0FSb2Ba4GPAgGcDTwL3AIMBBYCJ0fE65IEXAEcDbwDjI2Ix4qMz6yr8BVHVq6iWwpXAHdFxL7Ax4H5wCTgvojYG7gvrQOMAvZOr/HA5PUPZ2ZmRSosKUjaARgOXAcQEWsi4g1gNNA8Mc9U4Pi0PBq4MTKzgd6SdsXMzCqmyJbCnsBy4EeSHpd0raRtgX4RsTTVeRnol5b7A4tL9m9MZeuQNF7SHElzfIWRmVn7KnJMoTtwADAxIh6RdAXvdxUBEBEhaZNuKIiIKcAUyOY+aq9gzWqNxxFscxTZUmgEGiPikbQ+jSxJLGvuFko/X0nblwADSvavT2VmZlYhhSWFiHgZWCxpn1Q0AngamAGMSWVjgOlpeQbwBWUOBlaWdDOZmVkFFH3z2kTgJkl1wIvAWWSJ6FZJ44BFwMmp7kyyy1EXkF2SelbBsZmZWQuFJoWImAe0Nmf3iFbqBjChyHjMzGzDam6aC7OuzIPLtqU8zYWZmeWcFMzMLOekYGZmOScFMzPLOSmYmVnOScHMzHJOCmZmlvN9CmadnO9NsPbkloKZmeWcFMzMLOekYGZmOScFMzPLOSmYmVnOScHMzHJOCmZmlnNSMDOznJOCmZnlfEezWSfku5itKG4pmJlZrtCkIGmhpD9JmidpTirbUdK9kp5PP/ukckm6UtICSU9IOqDI2MzMbH2VaCn8bUQMjoiGtD4JuC8i9gbuS+sAo4C902s8MLkCsZmZWYlqdB+NBqam5anA8SXlN0ZmNtBb0q5ViM/MrMsqOikEcI+kuZLGp7J+EbE0Lb8M9EvL/YHFJfs2prJ1SBovaY6kOcuXLy8qbjOzLqnoq48OiYglknYB7pX0TOnGiAhJsSkHjIgpwBSAhoaGTdrXzMw2rNCWQkQsST9fAW4HDgSWNXcLpZ+vpOpLgAElu9enMjMzq5DCkoKkbSX1al4GjgSeBGYAY1K1McD0tDwD+EK6CulgYGVJN5OZmVVAkd1H/YDbJTWf56cRcZekR4FbJY0DFgEnp/ozgaOBBcA7wFkFxmbWqfhmNauUwpJCRLwIfLyV8hXAiFbKA5hQVDxmZrZxvqPZzMxyTgpmZpZzUjAzs5yTgpmZ5ZwUzMws56RgZmY5JwUzM8v5yWtmHZRvWLNqcEvBzMxyTgpmZpZzUjAzs5yTgpmZ5ZwUzMws56RgZmY5JwUzM8s5KZiZWa6spCDpvHLKzMyscyv3juYxwBUtysa2UmZmW8B3MVu1bTApSDoNOB3YU9KMkk29gNeKDMzMzCpvYy2Fh4ClwM7Ad0vKVwFPFBWUmZlVxwaTQkQsAhYBn9zcE0jqBswBlkTEsZL2BG4GdgLmAmdGxBpJWwM3Ap8AVgCnRMTCzT2vmZltunIHmldJejO9VktaK+nNMs9xHjC/ZP1S4PKI2At4HRiXyscBr6fyy1M9MzOroLKSQkT0iojtI2J74APACcB/bWw/SfXAMcC1aV3AYcC0VGUqcHxaHp3WSdtHpPpmZlYhm3yfQmTuAEaWUf37wMXAX9P6TsAbEdGU1huB/mm5P7A4naMJWJnqr0PSeElzJM1Zvnz5poZvZmYbUNYlqZI+V7K6FdAArN7IPscCr0TEXEmHbnaELUTEFGAKQENDQ7TXcc3MrPz7FD5TstwELCTr7tmQYcBxko4GegLbk93X0FtS99QaqAeWpPpLgAFAo6TuwA5kA85mZlYhZSWFiDhrUw8cEV8FvgqQWgoXRcQZkn4OnEh2BdIYYHraZUZafzht/01EuCVgZlZB5V599CFJv5S0XNIrkqZL+tBmnvMrwIWSFpCNGVyXyq8DdkrlFwKTNvP4Zma2mcrtPvopcDXw2bR+KvAz4KBydo6IWcCstPwicGArdVYDJ5UZj1nN8NQW1pGUe/XRNhHx44hoSq+fkI0TmJlZDSm3pfArSZPIxgECOAWYKWlHgIjwPEhmZjWg3KRwcvr59y3KTyVLEps7vmBmZh1IuUnhI6nPPyepZ8syMzPr3ModU3iozDIzM+vENvY8hQ+STT/xAUlDgOa5iLYHtik4NjMzq7CNdR+NJHvCWj3wvZLyVcDXCorJzMyqZGPPU5gKTJV0QkTcVqGYzMysSsodaP6opEEtCyPiX9o5HjMzq6Jyk8JbJcs9gWNZ98E5ZmZWA8qdEK/0+cxI+g5wdyERmZlZ1ZTbUmhpG7LBZzPbDJ7vyDqqch+y8yeyO5chu7dhF+BfiwrKzMyqo9yWwrFAH+BvgN7AzIiYW1hUZmZWFeXe0Twa+DGwM9AD+JGkiYVFZWZmVVFuS+Ec4OCIeBtA0qVkT0i7qqjAzMys8sptKQhYW7K+lvenvDAzsxpRbkvhR8Ajkm5P68fz/mM0zcysRpR7n8L3JM0CDklFZ0XE44VFZWZmVVH2fQoR8RjwWIGxmJlZlW3uzWsbJakn8Dtg63SeaRHxDUl7kj3WcydgLnBmRKyRtDVwI/AJYAVwSkQsLCo+s0rzDWvWGZQ70Lw53gUOi4iPA4OBoyQdDFwKXB4RewGvA+NS/XHA66n88lTPzMwqqLCkEJnmifR6pFcAhwHTUvlUskFryO6FmJqWpwEjJPkKJzOzCiqypYCkbpLmAa8A9wIvAG9ERFOq0kj2ZDfSz8UAaftKsi6mlsccL2mOpDnLly8vMnwzsy6n0KQQEWsjYjDZ5HkHAvu2wzGnRERDRDT07dt3i2M0M7P3FZoUmkXEG8D9wCeB3pKaB7jrgSVpeQkwACBt34FswNnMzCqksKQgqa+k3mn5A8ARZA/muR84MVUbA0xPyzPSOmn7byIiMDOziinsklRgV7LnO3cjSz63RsSdkp4Gbpb0LeBx3r8z+jrgx5IWAK8BpxYYm9mG3f8f7X7Ig//shq+1p+8UctTCkkJEPAEMaaX8RbLxhZblq4GTiorHzMw2riJjCmZm1jkU2X1k1uU9/KK7jKxzcUvBzMxybimYQSEDy2adkVsKZmaWc1IwM7Ock4KZmeWcFMzMLOekYGZmOScFMzPLOSmYmVnOScHMzHK+ec2snXlqC+vMnBSs6/JdzGbrcfeRmZnlnBTMzCznpGBmZjknBTMzyzkpmJlZzknBzMxyhSUFSQMk3S/paUlPSTovle8o6V5Jz6effVK5JF0paYGkJyQdUFRsZmbWuiJbCk3AlyJiP+BgYIKk/YBJwH0RsTdwX1oHGAXsnV7jgckFxmZmZq0oLClExNKIeCwtrwLmA/2B0cDUVG0qcHxaHg3cGJnZQG9JuxYVn5mZra8idzRLGggMAR4B+kXE0rTpZaBfWu4PLC7ZrTGVLS0pQ9J4spYEu+++e2ExW40q6C5mT21htaLwgWZJ2wG3AedHxJul2yIigNiU40XElIhoiIiGvn37tmOkZmZWaFKQ1IMsIdwUEb9Ixcuau4XSz1dS+RJgQMnu9anMzMwqpMirjwRcB8yPiO+VbJoBjEnLY4DpJeVfSFchHQysLOlmMjOzCihyTGEYcCbwJ0nzUtnXgG8Dt0oaBywCTk7bZgJHAwuAd4CzCozNzMxaUVhSiIgHALWxeUQr9QOYUFQ8Zma2cb6j2czMck4KZmaW85PXrPb5CWtmZXNLwczMck4KZmaWc1IwM7Ock4KZmeU80Gy1qQKDy54Ez2qRWwpmZpZzUjAzs5yTgpmZ5ZwUzMws56RgZmY5JwUzM8s5KZiZWc73KVjt8MR3ZlvMLQUzM8s5KZiZWc7dR2abwFNbWK1zS8HMzHKFtRQkXQ8cC7wSER9NZTsCtwADgYXAyRHxuiQBVwBHA+8AYyPisaJisxriwWWzdlVkS+EG4KgWZZOA+yJib+C+tA4wCtg7vcYDkwuMy8zM2lBYUoiI3wGvtSgeDUxNy1OB40vKb4zMbKC3pF2Lis3MzFpX6TGFfhGxNC2/DPRLy/2BxSX1GlPZeiSNlzRH0pzly5cXF6mZWRdUtYHmiAggNmO/KRHREBENffv2LSAyM7Ouq9KXpC6TtGtELE3dQ6+k8iXAgJJ69anMbH0eXDYrTKVbCjOAMWl5DDC9pPwLyhwMrCzpZjIzswop8pLUnwGHAjtLagS+AXwbuFXSOGARcHKqPpPsctQFZJeknlVUXGZm1rbCkkJEnNbGphGt1A1gQlGxmJlZeXxHs5mZ5ZwUzMws5wnxrHOo4hVHngTPuhK3FMzMLOekYGZmOScFMzPLeUzBOi7fuWxWcW4pmJlZzknBzMxy7j6yjsVdRmZV5ZaCmZnl3FKw6nPrwKzDcFKw6nAiMOuQnBTMWuGpLayr8piCmZnl3FKwynGXkVmH55aCmZnl3FKw4rhlYNbpOClY+3IiMOvUnBRsyzkRmNWMDpUUJB0FXAF0A66NiG9XOSQr5S9/s5rXYZKCpG7A1cARQCPwqKQZEfF0dSPr4pwIzLqUDpMUgAOBBRHxIoCkm4HRgJNCe/KXvJltQEdKCv2BxSXrjcBBLStJGg+MT6tvSXq2ArG1t52BV6sdRIV1tffc1d4v+D1X1jnf3ZK992hrQ0dKCmWJiCnAlGrHsSUkzYmIhmrHUUld7T13tfcLfs+1oiPdvLYEGFCyXp/KzMysQjpSUngU2FvSnpLqgFOBGVWOycysS+kw3UcR0STpn4C7yS5JvT4inqpyWEXp1N1fm6mrveeu9n7B77kmKCKqHYOZmXUQHan7yMzMqsxJwczMck4KVSbpS5JC0s7VjqVIkv5T0jOSnpB0u6Te1Y6pKJKOkvSspAWSJlU7nqJJGiDpfklPS3pK0nnVjqlSJHWT9LikO6sdS3txUqgiSQOAI4E/VzuWCrgX+GhE7A88B3y1yvEUomS6llHAfsBpkvarblSFawK+FBH7AQcDE7rAe252HjC/2kG0JyeF6rocuBio+dH+iLgnIprS6myy+1BqUT5dS0SsAZqna6lZEbE0Ih5Ly6vIviT7Vzeq4kmqB44Brq12LO3JSaFKJI0GlkTEH6sdSxWcDfyq2kEUpLXpWmr+C7KZpIHAEOCR6kZSEd8n+6fur9UOpD11mPsUapGkXwMfbGXT14GvkXUd1YwNvd+ImJ7qfJ2su+GmSsZmxZO0HXAbcH5EvFnteIok6VjglYiYK+nQasfTnpwUChQRh7dWLuljwJ7AHyVB1pXymKQDI+LlCobYrtp6v80kjQWOBUZE7d4g0yWna5HUgywh3BQRv6h2PBUwDDhO0tFAT2B7ST+JiM9XOa4t5pvXOgBJC4GGiKjZGSbTA5S+B3w6IpZXO56iSOpONpA+giwZPAqcXsN356PsP5upwGsRcX6146m01FK4KCKOrXYs7cFjClYpPwB6AfdKmifpmmoHVIQ0mN48Xct84NZaTgjJMOBM4LD0u52X/oO2TsgtBTMzy7mlYGZmOScFMzPLOSmYmVnOScHMzHJOCmZmlnNSMCuApH+RtMGb+cw6Il+SatbOJHWLiLVF72NWBLcUzDaBpIHpuRA3SZovaZqkbSQtlHSppMeAkyTdIOnEtM+INOf+nyRdL2nrVL7OPtV8X2bNnBTMNt0+wH9FxEeAN4F/TOUrIuKAiLi5uaKknsANwCkR8TGy+cb+oeRY6+1jVk1OCmabbnFEPJiWfwIckpZvaaXuPsBLEfFcWp8KDC/Z3to+ZlXjpGC26VoOxDWvv70Zx9qcfcwK46Rgtul2l/TJtHw68MAG6j4LDJS0V1o/E/htkcGZbQknBbNN9yzZc4jnA32AyW1VjIjVwFnAzyX9iewpXTU5Q6zVBl+SarYJ0uMm74yIj1Y5FLNCuKVgZmY5txTMzCznloKZmeWcFMzMLOekYGZmOScFMzPLOSmYmVnu/+ID2jUAAAAFSURBVADsAVu5OhVCyQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Generated Samples tensor([-0., -0., -0., 0., -0., 1., 1., -0., -0., 0., -0., -0., -0., -0., -0., -0., -0., -0.,\n",
            "        -0., 1., 1., -0., -0., -0., -0., -0., -0., 0., -0., -0., 1., 1., -0., -0., -0., 1.,\n",
            "        -0., -0., -0., -0., -0., -0., -0., 1., 1., -0., -0., -0., -0., -0.])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3A9zlP9CNwqK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c8aac8a1-181d-4fca-91b7-32b9129b7868"
      },
      "source": [
        "latent_space_samples[generated_samples==1.0].shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([286])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ab3uR_eFYlUj"
      },
      "source": [
        "## Exercise:\r\n",
        "\r\n",
        "See if you can train a model with a diffenrent probability of heads.\r\n",
        "\r\n",
        "See if you can train a GAN to model more complex 1-D density estimation problems. \r\n",
        "\r\n",
        "See if you can change the loss function and architectures of the generator and discriminator networks to learn the desnsity more efficiently.\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "### Acknowledgements\r\n",
        "\r\n",
        "Some parts of this tutorial have been inspired from the source below which is strongly recommended for reading.\r\n",
        "\r\n",
        "Raevskiy, Mikhail. “Write Your First Generative Adversarial Network Model on PyTorch.” Medium, August 31, 2020. https://medium.com/dev-genius/write-your-first-generative-adversarial-network-model-on-pytorch-7dc0c7c892c7.\r\n"
      ]
    }
  ]
}