From 0859359c10b824b9bc0e939b9ad442c700f5376f Mon Sep 17 00:00:00 2001 From: KrishnaDhakshin Date: Sun, 31 Mar 2019 01:22:12 +0530 Subject: [PATCH 1/8] notebook added --- notebooks/MHGAN_MNIST.ipynb | 14779 ++++++++++++++++++++++++++++++++++ 1 file changed, 14779 insertions(+) create mode 100644 notebooks/MHGAN_MNIST.ipynb diff --git a/notebooks/MHGAN_MNIST.ipynb b/notebooks/MHGAN_MNIST.ipynb new file mode 100644 index 0000000..499b978 --- /dev/null +++ b/notebooks/MHGAN_MNIST.ipynb @@ -0,0 +1,14779 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "MHGAN_MNIST.ipynb", + "version": "0.3.2", + "provenance": [], + "collapsed_sections": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "accelerator": "GPU" + }, + "cells": [ + { + "metadata": { + "id": "ReXdYV5Z6wNZ", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "!pip uninstall -y Pillow\n", + "!pip install Pillow==5.3.0\n", + "!pip install torchgan\n", + "!pip install tensorboardX" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "JxW667X0DwPE", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "import os\n", + "import random\n", + "import matplotlib.pyplot as plt\n", + "import matplotlib.animation as animation\n", + "import numpy as np\n", + "from IPython.display import HTML\n", + "import torch\n", + "import torch.nn as nn\n", + "import torchvision\n", + "from torch.optim import Adam\n", + "from torch.optim import SGD\n", + "import torch.nn as nn\n", + "import torch.utils.data as data\n", + "import torchvision.datasets as dsets\n", + "import torchvision.transforms as transforms\n", + "import torchvision.utils as vutils\n", + "import torchgan\n", + "from torchgan.models import *\n", + "from torchgan.losses import *\n", + "from torchgan.trainer import Trainer" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "XC8UpLxcEH-Y", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True)" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "yxOhiOGQGQON", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2)" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "My8wjH-oGYOZ", + "colab_type": "code", + "outputId": "3e7b1b78-d08d-43a1-c1c9-7a6c198ec5a7", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 498 + } + }, + "cell_type": "code", + "source": [ + "real_batch = next(iter(dataloader))\n", + "plt.figure(figsize=(8,8))\n", + "plt.axis(\"off\")\n", + "plt.title(\"Training Images\")\n", + "plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0)))\n", + "plt.show()" + ], + "execution_count": 5, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHhCAYAAAAibsTCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGIZh+In/B0GGiFVplUQhAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + } + ] + }, + { + "metadata": { + "id": "9Wn3sekuGf6T", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "acgan = {\n", + " \"generator\": {\n", + " \"name\": ACGANGenerator,\n", + " \"args\": {\n", + " \"encoding_dims\": 100,\n", + " \"num_classes\": 10,\n", + " \"out_channels\": 1,\n", + " \"step_channels\": 32,\n", + " \"out_size\":32,\n", + " \"nonlinearity\": nn.LeakyReLU(0.2),\n", + " \"last_nonlinearity\": nn.Tanh()\n", + " },\n", + " \"optimizer\": {\n", + " \"name\": Adam,\n", + " \"args\": {\n", + " \"lr\": 0.0009,\n", + " \"betas\": (0.5, 0.999)\n", + " }\n", + " }\n", + " },\n", + " \"discriminator\": {\n", + " \"name\": ACGANDiscriminator,\n", + " \"args\": {\n", + " \"in_channels\": 1,\n", + " \"step_channels\": 32,\n", + " \"in_size\": 32,\n", + " \"num_classes\": 10,\n", + " \"nonlinearity\": nn.LeakyReLU(0.2),\n", + " \"last_nonlinearity\": nn.Sigmoid()\n", + " },\n", + " \"optimizer\": {\n", + " \"name\": Adam,\n", + " \"args\": {\n", + " \"lr\": 0.0002,\n", + " \"betas\": (0.5, 0.999)\n", + " }\n", + " }\n", + " }\n", + "}" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "tVEjIUtxJdbc", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),]" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "LigSujXVJwZS", + "colab_type": "code", + "outputId": "e91c624f-267b-44b1-add6-114db8ffd0ae", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 51 + } + }, + "cell_type": "code", + "source": [ + "if torch.cuda.is_available():\n", + " device = torch.device(\"cuda:0\")\n", + " torch.backends.cudnn.deterministic = True\n", + " epochs = 20\n", + "else:\n", + " device = torch.device(\"cpu\")\n", + " epochs = 5\n", + "\n", + "print(\"Device: {}\".format(device))\n", + "print(\"Epochs: {}\".format(epochs))" + ], + "execution_count": 130, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Device: cuda:0\n", + "Epochs: 20\n" + ], + "name": "stdout" + } + ] + }, + { + "metadata": { + "id": "XegwsijSJ1jF", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device)" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "kTatr-LQJ9qa", + "colab_type": "code", + "outputId": "588aa7df-e61b-49ff-d00a-d383e7f0f39a", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1951 + } + }, + "cell_type": "code", + "source": [ + "trainer(dataloader)" + ], + "execution_count": 132, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Saving Model at './model/gan0.model'\n", + "Epoch 1 Summary\n", + "generator Mean Gradients : 0.5629363154607244\n", + "discriminator Mean Gradients : 21.085883638233142\n", + "Mean Running Discriminator Loss : 2.132794382729764\n", + "Mean Running Generator Loss : 1.117769119852006\n", + "Generating and Saving Images to ./images/epoch1_generator.png\n", + "\n", + "Saving Model at './model/gan1.model'\n", + "Epoch 2 Summary\n", + "generator Mean Gradients : 0.4621404109377145\n", + "discriminator Mean Gradients : 37.54858841070317\n", + "Mean Running Discriminator Loss : 2.0983149196102677\n", + "Mean Running Generator Loss : 1.0905873073554877\n", + "Generating and Saving Images to ./images/epoch2_generator.png\n", + "\n", + "Saving Model at './model/gan2.model'\n", + "Epoch 3 Summary\n", + "generator Mean Gradients : 0.42224539835597896\n", + "discriminator Mean Gradients : 34.49435525860745\n", + "Mean Running Discriminator Loss : 2.097173844763973\n", + "Mean Running Generator Loss : 1.0768492916637307\n", + "Generating and Saving Images to ./images/epoch3_generator.png\n", + "\n", + "Saving Model at './model/gan3.model'\n", + "Epoch 4 Summary\n", + "generator Mean Gradients : 0.3910194754263064\n", + "discriminator Mean Gradients : 30.64363928034132\n", + "Mean Running Discriminator Loss : 2.0992902942907326\n", + "Mean Running Generator Loss : 1.069386471527567\n", + "Generating and Saving Images to ./images/epoch4_generator.png\n", + "\n", + "Saving Model at './model/gan4.model'\n", + "Epoch 5 Summary\n", + "generator Mean Gradients : 0.3700217700678151\n", + "discriminator Mean Gradients : 28.378661746196947\n", + "Mean Running Discriminator Loss : 2.1001694201152206\n", + "Mean Running Generator Loss : 1.0651295932053504\n", + "Generating and Saving Images to ./images/epoch5_generator.png\n", + "\n", + "Saving Model at './model/gan0.model'\n", + "Epoch 6 Summary\n", + "generator Mean Gradients : 0.3731452710248436\n", + "discriminator Mean Gradients : 27.391221197105324\n", + "Mean Running Discriminator Loss : 2.098479413941725\n", + "Mean Running Generator Loss : 1.0625621958158502\n", + "Generating and Saving Images to ./images/epoch6_generator.png\n", + "\n", + "Saving Model at './model/gan1.model'\n", + "Epoch 7 Summary\n", + "generator Mean Gradients : 0.4097920223127502\n", + "discriminator Mean Gradients : 27.55014014898362\n", + "Mean Running Discriminator Loss : 2.092044219824006\n", + "Mean Running Generator Loss : 1.0619102914250553\n", + "Generating and Saving Images to ./images/epoch7_generator.png\n", + "\n", + "Saving Model at './model/gan2.model'\n", + "Epoch 8 Summary\n", + "generator Mean Gradients : 0.42416808925331695\n", + "discriminator Mean Gradients : 27.253160953803246\n", + "Mean Running Discriminator Loss : 2.082712709395362\n", + "Mean Running Generator Loss : 1.0626732571832915\n", + "Generating and Saving Images to ./images/epoch8_generator.png\n", + "\n", + "Saving Model at './model/gan3.model'\n", + "Epoch 9 Summary\n", + "generator Mean Gradients : 0.37972263266086353\n", + "discriminator Mean Gradients : 24.476701898901787\n", + "Mean Running Discriminator Loss : 2.0820189374751656\n", + "Mean Running Generator Loss : 1.0642426933875244\n", + "Generating and Saving Images to ./images/epoch9_generator.png\n", + "\n", + "Saving Model at './model/gan4.model'\n", + "Epoch 10 Summary\n", + "generator Mean Gradients : 0.3470192597153898\n", + "discriminator Mean Gradients : 22.522278636586094\n", + "Mean Running Discriminator Loss : 2.0774675978208657\n", + "Mean Running Generator Loss : 1.0654202675784448\n", + "Generating and Saving Images to ./images/epoch10_generator.png\n", + "\n" + ], + "name": "stdout" + }, + { + "output_type": "error", + "ename": "KeyboardInterrupt", + "evalue": "ignored", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 436\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 437\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 438\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 398\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreal_inputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 399\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 400\u001b[0;31m \u001b[0mlgen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgen_iter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_iter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 401\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'generator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mlgen\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 402\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'discriminator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain_iter\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 334\u001b[0m \u001b[0mgrad_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate_grads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 335\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mDiscriminatorLoss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 336\u001b[0;31m \u001b[0mcur_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_ops\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_arguments\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_arg_maps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 337\u001b[0m \u001b[0mloss_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcur_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 338\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mldis\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mcur_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/losses/auxclassifier.py\u001b[0m in \u001b[0;36mtrain_ops\u001b[0;34m(self, generator, discriminator, optimizer_discriminator, real_inputs, device, labels)\u001b[0m\n\u001b[1;32m 140\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 141\u001b[0m \u001b[0moptimizer_discriminator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m: " + ] + } + ] + }, + { + "metadata": { + "id": "a8wXZRm-KDVr", + "colab_type": "code", + "outputId": "cd8736b9-0fa3-4def-929a-84882f9fc74e", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1136 + } + }, + "cell_type": "code", + "source": [ + "fig = plt.figure(figsize=(8,8))\n", + "plt.axis(\"off\")\n", + "ims = [[plt.imshow(plt.imread(\"{}/epoch{}_generator.png\".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)]\n", + "ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True)\n", + "HTML(ani.to_jshtml())" + ], + "execution_count": 133, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "\n", + "
\n", + " \n", + "
\n", + " \n", + "
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
\n", + " Once \n", + " Loop \n", + " Reflect \n", + "
\n", + "
\n", + "\n", + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": { + "tags": [] + }, + "execution_count": 133 + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHVCAYAAAC9hYF5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkVVKKaVsoousUkopZZP/BwT3oBgGetiyAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + } + ] + }, + { + "metadata": { + "id": "YtjyFoEez8vl", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "gen = trainer.generator" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "t5ZrkNMVwmdI", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "dis = trainer.discriminator" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "dExgdnaG8MDl", + "colab_type": "code", + "outputId": "36d36215-ee6e-41e5-b46a-209c0645c9be", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 3637 + } + }, + "cell_type": "code", + "source": [ + "for i in range(10):\n", + " x = torch.randn([1,100], device=device)\n", + " for k in range(1000):\n", + " xk = torch.randn([1,100], device=device)\n", + " a = 1/dis(gen(x, torch.Tensor([i]).cuda()))\n", + " b = 1/dis(gen(xk, torch.Tensor([i]).cuda()))\n", + " d = (a-1)/(b-1)\n", + " p = torch.rand([1,1], device=device)\n", + " if (p < min(1, d)):\n", + " x = xk\n", + " image = gen(x, torch.Tensor([i]).cuda())\n", + " plt.figure()\n", + " plt.axis(\"off\")\n", + " plt.title(i)\n", + " plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0)))" + ], + "execution_count": 153, + "outputs": [ + { + "output_type": "stream", + "text": [ + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n" + ], + "name": "stderr" + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCRD4Bym7H3Oe+mRLAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngMDfvMqQ4k8FizsAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQCwBAmIJEBBLgIBYAgT+Ay0R0MvvvT5UAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxBIgIJYAAbEECIglQEAsAQL/A9upZjYozdLsAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAgTEEiDwD7l8ioHiDmDhAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + } + ] + } + ] +} \ No newline at end of file From 69676d00abd19bccccd15008c5660700c793a1c9 Mon Sep 17 00:00:00 2001 From: KrishnaDhakshin Date: Sun, 31 Mar 2019 01:26:53 +0530 Subject: [PATCH 2/8] Model added --- models/MHGAN_MNIST.py | 128 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 128 insertions(+) create mode 100644 models/MHGAN_MNIST.py diff --git a/models/MHGAN_MNIST.py b/models/MHGAN_MNIST.py new file mode 100644 index 0000000..505c969 --- /dev/null +++ b/models/MHGAN_MNIST.py @@ -0,0 +1,128 @@ +# -*- coding: utf-8 -*- +"""MHGAN_MNIST.ipynb + +Automatically generated by Colaboratory. + +Original file is located at + https://colab.research.google.com/drive/1WLBNRmz-EHOwK7Ew4zs_N4FZjZefKg5b +""" + +!pip uninstall -y Pillow +!pip install Pillow==5.3.0 +!pip install torchgan +!pip install tensorboardX + +import os +import random +import matplotlib.pyplot as plt +import matplotlib.animation as animation +import numpy as np +from IPython.display import HTML +import torch +import torch.nn as nn +import torchvision +from torch.optim import Adam +from torch.optim import SGD +import torch.nn as nn +import torch.utils.data as data +import torchvision.datasets as dsets +import torchvision.transforms as transforms +import torchvision.utils as vutils +import torchgan +from torchgan.models import * +from torchgan.losses import * +from torchgan.trainer import Trainer + +dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True) + +dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2) + +real_batch = next(iter(dataloader)) +plt.figure(figsize=(8,8)) +plt.axis("off") +plt.title("Training Images") +plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0))) +plt.show() + +acgan = { + "generator": { + "name": ACGANGenerator, + "args": { + "encoding_dims": 100, + "num_classes": 10, + "out_channels": 1, + "step_channels": 32, + "out_size":32, + "nonlinearity": nn.LeakyReLU(0.2), + "last_nonlinearity": nn.Tanh() + }, + "optimizer": { + "name": Adam, + "args": { + "lr": 0.0009, + "betas": (0.5, 0.999) + } + } + }, + "discriminator": { + "name": ACGANDiscriminator, + "args": { + "in_channels": 1, + "step_channels": 32, + "in_size": 32, + "num_classes": 10, + "nonlinearity": nn.LeakyReLU(0.2), + "last_nonlinearity": nn.Sigmoid() + }, + "optimizer": { + "name": Adam, + "args": { + "lr": 0.0002, + "betas": (0.5, 0.999) + } + } + } +} + +loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),] + +if torch.cuda.is_available(): + device = torch.device("cuda:0") + torch.backends.cudnn.deterministic = True + epochs = 20 +else: + device = torch.device("cpu") + epochs = 5 + +print("Device: {}".format(device)) +print("Epochs: {}".format(epochs)) + +trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device) + +trainer(dataloader) + +fig = plt.figure(figsize=(8,8)) +plt.axis("off") +ims = [[plt.imshow(plt.imread("{}/epoch{}_generator.png".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)] +ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True) +HTML(ani.to_jshtml()) + +gen = trainer.generator + +dis = trainer.discriminator + +for i in range(10): + x = torch.randn([1,100], device=device) + for k in range(1000): + xk = torch.randn([1,100], device=device) + a = 1/dis(gen(x, torch.Tensor([i]).cuda())) + b = 1/dis(gen(xk, torch.Tensor([i]).cuda())) + d = (a-1)/(b-1) + p = torch.rand([1,1], device=device) + if (p < min(1, d)): + x = xk + image = gen(x, torch.Tensor([i]).cuda()) + plt.figure() + plt.axis("off") + plt.title(i) + plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0))) \ No newline at end of file From 4aecb0c607797fb3c1eee484467cc45c02d50149 Mon Sep 17 00:00:00 2001 From: KrishnaDhakshin Date: Sun, 31 Mar 2019 01:30:22 +0530 Subject: [PATCH 3/8] modified --- models/MHGAN_MNIST.py | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/models/MHGAN_MNIST.py b/models/MHGAN_MNIST.py index 505c969..c2c6c28 100644 --- a/models/MHGAN_MNIST.py +++ b/models/MHGAN_MNIST.py @@ -1,17 +1,3 @@ -# -*- coding: utf-8 -*- -"""MHGAN_MNIST.ipynb - -Automatically generated by Colaboratory. - -Original file is located at - https://colab.research.google.com/drive/1WLBNRmz-EHOwK7Ew4zs_N4FZjZefKg5b -""" - -!pip uninstall -y Pillow -!pip install Pillow==5.3.0 -!pip install torchgan -!pip install tensorboardX - import os import random import matplotlib.pyplot as plt From efcdd5ca92fbc18fe01268666edefe8d05d25d7a Mon Sep 17 00:00:00 2001 From: KrishnaDhakshin Date: Wed, 15 May 2019 21:17:41 +0530 Subject: [PATCH 4/8] made directories --- models/MHGAN_MNIST.py | 114 - notebooks/MHGAN_MNIST.ipynb | 14779 ---------------------------------- 2 files changed, 14893 deletions(-) delete mode 100644 models/MHGAN_MNIST.py delete mode 100644 notebooks/MHGAN_MNIST.ipynb diff --git a/models/MHGAN_MNIST.py b/models/MHGAN_MNIST.py deleted file mode 100644 index c2c6c28..0000000 --- a/models/MHGAN_MNIST.py +++ /dev/null @@ -1,114 +0,0 @@ -import os -import random -import matplotlib.pyplot as plt -import matplotlib.animation as animation -import numpy as np -from IPython.display import HTML -import torch -import torch.nn as nn -import torchvision -from torch.optim import Adam -from torch.optim import SGD -import torch.nn as nn -import torch.utils.data as data -import torchvision.datasets as dsets -import torchvision.transforms as transforms -import torchvision.utils as vutils -import torchgan -from torchgan.models import * -from torchgan.losses import * -from torchgan.trainer import Trainer - -dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True) - -dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2) - -real_batch = next(iter(dataloader)) -plt.figure(figsize=(8,8)) -plt.axis("off") -plt.title("Training Images") -plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0))) -plt.show() - -acgan = { - "generator": { - "name": ACGANGenerator, - "args": { - "encoding_dims": 100, - "num_classes": 10, - "out_channels": 1, - "step_channels": 32, - "out_size":32, - "nonlinearity": nn.LeakyReLU(0.2), - "last_nonlinearity": nn.Tanh() - }, - "optimizer": { - "name": Adam, - "args": { - "lr": 0.0009, - "betas": (0.5, 0.999) - } - } - }, - "discriminator": { - "name": ACGANDiscriminator, - "args": { - "in_channels": 1, - "step_channels": 32, - "in_size": 32, - "num_classes": 10, - "nonlinearity": nn.LeakyReLU(0.2), - "last_nonlinearity": nn.Sigmoid() - }, - "optimizer": { - "name": Adam, - "args": { - "lr": 0.0002, - "betas": (0.5, 0.999) - } - } - } -} - -loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),] - -if torch.cuda.is_available(): - device = torch.device("cuda:0") - torch.backends.cudnn.deterministic = True - epochs = 20 -else: - device = torch.device("cpu") - epochs = 5 - -print("Device: {}".format(device)) -print("Epochs: {}".format(epochs)) - -trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device) - -trainer(dataloader) - -fig = plt.figure(figsize=(8,8)) -plt.axis("off") -ims = [[plt.imshow(plt.imread("{}/epoch{}_generator.png".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)] -ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True) -HTML(ani.to_jshtml()) - -gen = trainer.generator - -dis = trainer.discriminator - -for i in range(10): - x = torch.randn([1,100], device=device) - for k in range(1000): - xk = torch.randn([1,100], device=device) - a = 1/dis(gen(x, torch.Tensor([i]).cuda())) - b = 1/dis(gen(xk, torch.Tensor([i]).cuda())) - d = (a-1)/(b-1) - p = torch.rand([1,1], device=device) - if (p < min(1, d)): - x = xk - image = gen(x, torch.Tensor([i]).cuda()) - plt.figure() - plt.axis("off") - plt.title(i) - plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0))) \ No newline at end of file diff --git a/notebooks/MHGAN_MNIST.ipynb b/notebooks/MHGAN_MNIST.ipynb deleted file mode 100644 index 499b978..0000000 --- a/notebooks/MHGAN_MNIST.ipynb +++ /dev/null @@ -1,14779 +0,0 @@ -{ - "nbformat": 4, - "nbformat_minor": 0, - "metadata": { - "colab": { - "name": "MHGAN_MNIST.ipynb", - "version": "0.3.2", - "provenance": [], - "collapsed_sections": [] - }, - "kernelspec": { - "name": "python3", - "display_name": "Python 3" - }, - "accelerator": "GPU" - }, - "cells": [ - { - "metadata": { - "id": "ReXdYV5Z6wNZ", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "!pip uninstall -y Pillow\n", - "!pip install Pillow==5.3.0\n", - "!pip install torchgan\n", - "!pip install tensorboardX" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "JxW667X0DwPE", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "import os\n", - "import random\n", - "import matplotlib.pyplot as plt\n", - "import matplotlib.animation as animation\n", - "import numpy as np\n", - "from IPython.display import HTML\n", - "import torch\n", - "import torch.nn as nn\n", - "import torchvision\n", - "from torch.optim import Adam\n", - "from torch.optim import SGD\n", - "import torch.nn as nn\n", - "import torch.utils.data as data\n", - "import torchvision.datasets as dsets\n", - "import torchvision.transforms as transforms\n", - "import torchvision.utils as vutils\n", - "import torchgan\n", - "from torchgan.models import *\n", - "from torchgan.losses import *\n", - "from torchgan.trainer import Trainer" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "XC8UpLxcEH-Y", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True)" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "yxOhiOGQGQON", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2)" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "My8wjH-oGYOZ", - "colab_type": "code", - "outputId": "3e7b1b78-d08d-43a1-c1c9-7a6c198ec5a7", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 498 - } - }, - "cell_type": "code", - "source": [ - "real_batch = next(iter(dataloader))\n", - "plt.figure(figsize=(8,8))\n", - "plt.axis(\"off\")\n", - "plt.title(\"Training Images\")\n", - "plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0)))\n", - "plt.show()" - ], - "execution_count": 5, - "outputs": [ - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHhCAYAAAAibsTCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGIZh+In/B0GGiFVplUQhAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - } - ] - }, - { - "metadata": { - "id": "9Wn3sekuGf6T", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "acgan = {\n", - " \"generator\": {\n", - " \"name\": ACGANGenerator,\n", - " \"args\": {\n", - " \"encoding_dims\": 100,\n", - " \"num_classes\": 10,\n", - " \"out_channels\": 1,\n", - " \"step_channels\": 32,\n", - " \"out_size\":32,\n", - " \"nonlinearity\": nn.LeakyReLU(0.2),\n", - " \"last_nonlinearity\": nn.Tanh()\n", - " },\n", - " \"optimizer\": {\n", - " \"name\": Adam,\n", - " \"args\": {\n", - " \"lr\": 0.0009,\n", - " \"betas\": (0.5, 0.999)\n", - " }\n", - " }\n", - " },\n", - " \"discriminator\": {\n", - " \"name\": ACGANDiscriminator,\n", - " \"args\": {\n", - " \"in_channels\": 1,\n", - " \"step_channels\": 32,\n", - " \"in_size\": 32,\n", - " \"num_classes\": 10,\n", - " \"nonlinearity\": nn.LeakyReLU(0.2),\n", - " \"last_nonlinearity\": nn.Sigmoid()\n", - " },\n", - " \"optimizer\": {\n", - " \"name\": Adam,\n", - " \"args\": {\n", - " \"lr\": 0.0002,\n", - " \"betas\": (0.5, 0.999)\n", - " }\n", - " }\n", - " }\n", - "}" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "tVEjIUtxJdbc", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),]" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "LigSujXVJwZS", - "colab_type": "code", - "outputId": "e91c624f-267b-44b1-add6-114db8ffd0ae", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 51 - } - }, - "cell_type": "code", - "source": [ - "if torch.cuda.is_available():\n", - " device = torch.device(\"cuda:0\")\n", - " torch.backends.cudnn.deterministic = True\n", - " epochs = 20\n", - "else:\n", - " device = torch.device(\"cpu\")\n", - " epochs = 5\n", - "\n", - "print(\"Device: {}\".format(device))\n", - "print(\"Epochs: {}\".format(epochs))" - ], - "execution_count": 130, - "outputs": [ - { - "output_type": "stream", - "text": [ - "Device: cuda:0\n", - "Epochs: 20\n" - ], - "name": "stdout" - } - ] - }, - { - "metadata": { - "id": "XegwsijSJ1jF", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device)" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "kTatr-LQJ9qa", - "colab_type": "code", - "outputId": "588aa7df-e61b-49ff-d00a-d383e7f0f39a", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1951 - } - }, - "cell_type": "code", - "source": [ - "trainer(dataloader)" - ], - "execution_count": 132, - "outputs": [ - { - "output_type": "stream", - "text": [ - "Saving Model at './model/gan0.model'\n", - "Epoch 1 Summary\n", - "generator Mean Gradients : 0.5629363154607244\n", - "discriminator Mean Gradients : 21.085883638233142\n", - "Mean Running Discriminator Loss : 2.132794382729764\n", - "Mean Running Generator Loss : 1.117769119852006\n", - "Generating and Saving Images to ./images/epoch1_generator.png\n", - "\n", - "Saving Model at './model/gan1.model'\n", - "Epoch 2 Summary\n", - "generator Mean Gradients : 0.4621404109377145\n", - "discriminator Mean Gradients : 37.54858841070317\n", - "Mean Running Discriminator Loss : 2.0983149196102677\n", - "Mean Running Generator Loss : 1.0905873073554877\n", - "Generating and Saving Images to ./images/epoch2_generator.png\n", - "\n", - "Saving Model at './model/gan2.model'\n", - "Epoch 3 Summary\n", - "generator Mean Gradients : 0.42224539835597896\n", - "discriminator Mean Gradients : 34.49435525860745\n", - "Mean Running Discriminator Loss : 2.097173844763973\n", - "Mean Running Generator Loss : 1.0768492916637307\n", - "Generating and Saving Images to ./images/epoch3_generator.png\n", - "\n", - "Saving Model at './model/gan3.model'\n", - "Epoch 4 Summary\n", - "generator Mean Gradients : 0.3910194754263064\n", - "discriminator Mean Gradients : 30.64363928034132\n", - "Mean Running Discriminator Loss : 2.0992902942907326\n", - "Mean Running Generator Loss : 1.069386471527567\n", - "Generating and Saving Images to ./images/epoch4_generator.png\n", - "\n", - "Saving Model at './model/gan4.model'\n", - "Epoch 5 Summary\n", - "generator Mean Gradients : 0.3700217700678151\n", - "discriminator Mean Gradients : 28.378661746196947\n", - "Mean Running Discriminator Loss : 2.1001694201152206\n", - "Mean Running Generator Loss : 1.0651295932053504\n", - "Generating and Saving Images to ./images/epoch5_generator.png\n", - "\n", - "Saving Model at './model/gan0.model'\n", - "Epoch 6 Summary\n", - "generator Mean Gradients : 0.3731452710248436\n", - "discriminator Mean Gradients : 27.391221197105324\n", - "Mean Running Discriminator Loss : 2.098479413941725\n", - "Mean Running Generator Loss : 1.0625621958158502\n", - "Generating and Saving Images to ./images/epoch6_generator.png\n", - "\n", - "Saving Model at './model/gan1.model'\n", - "Epoch 7 Summary\n", - "generator Mean Gradients : 0.4097920223127502\n", - "discriminator Mean Gradients : 27.55014014898362\n", - "Mean Running Discriminator Loss : 2.092044219824006\n", - "Mean Running Generator Loss : 1.0619102914250553\n", - "Generating and Saving Images to ./images/epoch7_generator.png\n", - "\n", - "Saving Model at './model/gan2.model'\n", - "Epoch 8 Summary\n", - "generator Mean Gradients : 0.42416808925331695\n", - "discriminator Mean Gradients : 27.253160953803246\n", - "Mean Running Discriminator Loss : 2.082712709395362\n", - "Mean Running Generator Loss : 1.0626732571832915\n", - "Generating and Saving Images to ./images/epoch8_generator.png\n", - "\n", - "Saving Model at './model/gan3.model'\n", - "Epoch 9 Summary\n", - "generator Mean Gradients : 0.37972263266086353\n", - "discriminator Mean Gradients : 24.476701898901787\n", - "Mean Running Discriminator Loss : 2.0820189374751656\n", - "Mean Running Generator Loss : 1.0642426933875244\n", - "Generating and Saving Images to ./images/epoch9_generator.png\n", - "\n", - "Saving Model at './model/gan4.model'\n", - "Epoch 10 Summary\n", - "generator Mean Gradients : 0.3470192597153898\n", - "discriminator Mean Gradients : 22.522278636586094\n", - "Mean Running Discriminator Loss : 2.0774675978208657\n", - "Mean Running Generator Loss : 1.0654202675784448\n", - "Generating and Saving Images to ./images/epoch10_generator.png\n", - "\n" - ], - "name": "stdout" - }, - { - "output_type": "error", - "ename": "KeyboardInterrupt", - "evalue": "ignored", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 436\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 437\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 438\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 398\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreal_inputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 399\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 400\u001b[0;31m \u001b[0mlgen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgen_iter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_iter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 401\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'generator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mlgen\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 402\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'discriminator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain_iter\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 334\u001b[0m \u001b[0mgrad_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate_grads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 335\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mDiscriminatorLoss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 336\u001b[0;31m \u001b[0mcur_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_ops\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_arguments\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_arg_maps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 337\u001b[0m \u001b[0mloss_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcur_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 338\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mldis\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mcur_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/losses/auxclassifier.py\u001b[0m in \u001b[0;36mtrain_ops\u001b[0;34m(self, generator, discriminator, optimizer_discriminator, real_inputs, device, labels)\u001b[0m\n\u001b[1;32m 140\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 141\u001b[0m \u001b[0moptimizer_discriminator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;31mKeyboardInterrupt\u001b[0m: " - ] - } - ] - }, - { - "metadata": { - "id": "a8wXZRm-KDVr", - "colab_type": "code", - "outputId": "cd8736b9-0fa3-4def-929a-84882f9fc74e", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1136 - } - }, - "cell_type": "code", - "source": [ - "fig = plt.figure(figsize=(8,8))\n", - "plt.axis(\"off\")\n", - "ims = [[plt.imshow(plt.imread(\"{}/epoch{}_generator.png\".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)]\n", - "ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True)\n", - "HTML(ani.to_jshtml())" - ], - "execution_count": 133, - "outputs": [ - { - "output_type": "execute_result", - "data": { - "text/html": [ - "\n", - "\n", - "\n", - "\n", - "
\n", - " \n", - "
\n", - " \n", - "
\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
\n", - " Once \n", - " Loop \n", - " Reflect \n", - "
\n", - "
\n", - "\n", - "\n", - "\n" - ], - "text/plain": [ - "" - ] - }, - "metadata": { - "tags": [] - }, - "execution_count": 133 - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHVCAYAAAC9hYF5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkVVKKaVsoousUkopZZP/BwT3oBgGetiyAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - } - ] - }, - { - "metadata": { - "id": "YtjyFoEez8vl", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "gen = trainer.generator" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "t5ZrkNMVwmdI", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "dis = trainer.discriminator" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "dExgdnaG8MDl", - "colab_type": "code", - "outputId": "36d36215-ee6e-41e5-b46a-209c0645c9be", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 3637 - } - }, - "cell_type": "code", - "source": [ - "for i in range(10):\n", - " x = torch.randn([1,100], device=device)\n", - " for k in range(1000):\n", - " xk = torch.randn([1,100], device=device)\n", - " a = 1/dis(gen(x, torch.Tensor([i]).cuda()))\n", - " b = 1/dis(gen(xk, torch.Tensor([i]).cuda()))\n", - " d = (a-1)/(b-1)\n", - " p = torch.rand([1,1], device=device)\n", - " if (p < min(1, d)):\n", - " x = xk\n", - " image = gen(x, torch.Tensor([i]).cuda())\n", - " plt.figure()\n", - " plt.axis(\"off\")\n", - " plt.title(i)\n", - " plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0)))" - ], - "execution_count": 153, - "outputs": [ - { - "output_type": "stream", - "text": [ - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n" - ], - "name": "stderr" - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCRD4Bym7H3Oe+mRLAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngMDfvMqQ4k8FizsAAAAASUVORK5CYII=\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAACqNJREFUeJzt3U2IlWUDxvH7TE5laQiHjGLQWlk0\nUJREMdDCjAiyDyLEvmyKcBNaRDggum9hhFFEiyD6WFi0TIRsEQnSbKpFaIWOBU1F4mhMqU0+7+Z9\nId7F3JfMnDNzZn6/9cVzP+L05wnPmafVNE1TAJhW31zfAEAvEEuAgFgCBMQSICCWAAGxBAiIJT3n\n119/LcPDw2XdunVlw4YNZXR0dK5viUVALOk5IyMj5c477yyfffZZ2bFjR3nvvffm+pZYBFo+lE4v\nGR8fLw888EA5ePBg6e/vn+vbYRHxZElPOXz4cBkYGCi7d+8u99xzT3n88cfLt99+O9e3xSIglvSU\n06dPl++++66sXbu27N+/v9x///3lueeeK1NTU3N9ayxwYklPWb58eWm322X9+vWllFIeeeSRcurU\nqTI2Nja3N8aCJ5b0lGuuuaZMTk6W8+fPl1JKabVapa+vr/T1+VGms/yE0VPWrFlTVq5cWT788MNS\nSin79u0rV1xxRVm1atUc3xkLnX8Np+f88MMPZWRkpJw8ebK02+2ya9euMjg4ONe3xQInlgAB/xsO\nEBBLgIBYAgTEEiAglgCBJd04pNVqdeMYgBmZ7sNBniwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVA\nQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEOjKayVgtj322GPR7v333+/wnbBYeLIE\nCIglQEAsAQJiCRAQS4CAWAIExBIgIJYAgVbTNE3HD2m1On0Ei8z4+Hi0u+6666qbM2fOzPR2WCCm\ny6EnS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRDwm9KZd3bu3Fnd/Pnnn9G1fOCc2eLJEiAglgAB\nsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiDgGzx01bp166qbTZs2VTfbtm2bjduBmCdLgIBYAgTEEiAg\nlgABsQQIiCVAQCwBAmIJEGg1TdN0/JBWq9NHMMe2b98e7TZv3lzdnD9/vroZHByMzoMLMV0OPVkC\nBMQSICCWAAGxBAiIJUBALAECYgkQEEuAgN+UTtX69eurmwcffDC6Vrvdrm6Ghoaia0E3ebIECIgl\nQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECPgGzyJ34403Vjc7duyobiYmJqLz7rjjjurm6NGj0bWg\nmzxZAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgECraZqm44e0Wp0+gv/z7LPPRrv77ruvujl+/Hh1\ns3Xr1ug8mM+my6EnS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRDwm9J70NDQUHUzPDwcXWtqaqq6\n2bhxY3QtWMg8WQIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIEfIOnB23fvr26GRwcjK61c+fO\n6ubMmTPRtci89tpr0e7YsWPVzSuvvDLT2yHkyRIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIEWk3T\nNB0/pNXq9BGLyi+//FLdnDt3LrrWqlWrZno7/Mstt9xS3ezbty+6VvJlgN27d1c3e/bsic6jlOly\n6MkSICCWAAGxBAiIJUBALAECYgkQEEuAgFgCBPym9Hnm7rvvrm76+/urm4mJidm4HS7Qpk2bqpuL\nLrooutbVV19d3ezatau6+eqrr6LzPv/882i3WHmyBAiIJUBALAECYgkQEEuAgFgCBMQSICCWAAGx\nBAj4Bs88s3Xr1uom+QbI6OjobNwO/3LzzTdXNw8//HB1s3Tp0tm4nVJKKZdffnl18+qrr0bXeuut\nt6qbN998M7rWQuTJEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgR8KH2eWbt2bXVz7ty56ubAgQOz\ncTuLwpdffhnt1qxZU91ccskl1c0///wTnXf27Nnq5vvvv69ufvzxx+i8p59+uro5duxYdbN///7o\nvF7jyRIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIEWk3TNB0/pNXq9BHz3rJly6LdiRMnqpvff/+9\nuhkYGIjO68Jf/wW79957q5u33347utaKFSuqm76+7JlhamqqupmcnKxuPvjgg+i8559/PtrVvPHG\nG9HuiSeeqG6SP9+7774bnffSSy9Fu26a7r8HT5YAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYA\nAa+V6JKNGzdGu+TbTidPnqxu5uM3c0opZdu2bdXNyMhIddNut6PzTp8+Xd0cP348utbLL79c3ezd\nuze6VjclryoppZSLL764uunv769uDh06FJ3XazxZAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIDX\nSswzv/32W3Vz8ODB6uahhx6ajduZdV9//XV1c+2111Y3Y2Nj0Xlbtmypbhbqh6j/J3kNSSmlXHrp\npdXNkSNHqptbb701Om8+8loJgBkSS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRDwWol55tNPP61u\nxsfHu3AnF2b16tXRbmBgoLqZmJiobm666aboPEq57LLLot3k5GR1Mzw8PNPb6VmeLAECYgkQEEuA\ngFgCBMQSICCWAAGxBAiIJUDAh9LnmUcffbS6eeaZZ6qbjz/+ODovec3DN998U90MDg5G5/3888/V\nzdGjR6NrUcpTTz1V3aSvdfnjjz+qm+RnYaHyZAkQEEuAgFgCBMQSICCWAAGxBAiIJUBALAECraZp\nmo4fEn4oltlz6NChaLd06dLqZsWKFdXNqVOnovPa7XZ189NPP1U3t99+e3ReLztw4EB1c/3111c3\ny5Yti847cuRIdXPbbbdF1+pV0+XQkyVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAwGslFqj0\nGy579+6tbu66667qZuXKldF5S5bUf+SWL19e3Xz00UfReVu2bKluTpw4EV1rw4YN1c2TTz5Z3axe\nvTo674Ybbqhuzp49W90cPnw4Om/Pnj3RbrHyZAkQEEuAgFgCBMQSICCWAAGxBAiIJUBALAECXitB\n1QsvvDArm1JKueqqq6qbv//+u7pJPoxdSikTExPVzV9//RVda2xsrLq58sorq5vkz1dK9gqOF198\nsbr55JNPovPwWgmAGRNLgIBYAgTEEiAglgABsQQIiCVAQCwBAj6Uzqx4/fXXo93Q0FB188UXX1Q3\no6Oj0XnvvPNOtINSfCgdYMbEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgR8gwfgv3yDB2CGxBIg\nIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYA\nAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEE\nCIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIgl\nQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4DAkm4c0jRNN44B6BhPlgABsQQI\niCVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVA\nQCwBAmIJEBBLgIBYAgT+Ay0R0MvvvT5UAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxBIgIJYAAbEECIglQEAsAQL/A9upZjYozdLsAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAgTEEiDwD7l8ioHiDmDhAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - } - ] - } - ] -} \ No newline at end of file From ee1983a51f7250781cfe21e27645653507970cfc Mon Sep 17 00:00:00 2001 From: KrishnaDhakshin Date: Wed, 15 May 2019 21:23:47 +0530 Subject: [PATCH 5/8] directories made properly --- models/mhgan/MHGAN_MNIST.py | 114 + notebooks/mhgan/MHGAN_MNIST.ipynb | 14779 ++++++++++++++++++++++++++++ 2 files changed, 14893 insertions(+) create mode 100644 models/mhgan/MHGAN_MNIST.py create mode 100644 notebooks/mhgan/MHGAN_MNIST.ipynb diff --git a/models/mhgan/MHGAN_MNIST.py b/models/mhgan/MHGAN_MNIST.py new file mode 100644 index 0000000..c2c6c28 --- /dev/null +++ b/models/mhgan/MHGAN_MNIST.py @@ -0,0 +1,114 @@ +import os +import random +import matplotlib.pyplot as plt +import matplotlib.animation as animation +import numpy as np +from IPython.display import HTML +import torch +import torch.nn as nn +import torchvision +from torch.optim import Adam +from torch.optim import SGD +import torch.nn as nn +import torch.utils.data as data +import torchvision.datasets as dsets +import torchvision.transforms as transforms +import torchvision.utils as vutils +import torchgan +from torchgan.models import * +from torchgan.losses import * +from torchgan.trainer import Trainer + +dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True) + +dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2) + +real_batch = next(iter(dataloader)) +plt.figure(figsize=(8,8)) +plt.axis("off") +plt.title("Training Images") +plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0))) +plt.show() + +acgan = { + "generator": { + "name": ACGANGenerator, + "args": { + "encoding_dims": 100, + "num_classes": 10, + "out_channels": 1, + "step_channels": 32, + "out_size":32, + "nonlinearity": nn.LeakyReLU(0.2), + "last_nonlinearity": nn.Tanh() + }, + "optimizer": { + "name": Adam, + "args": { + "lr": 0.0009, + "betas": (0.5, 0.999) + } + } + }, + "discriminator": { + "name": ACGANDiscriminator, + "args": { + "in_channels": 1, + "step_channels": 32, + "in_size": 32, + "num_classes": 10, + "nonlinearity": nn.LeakyReLU(0.2), + "last_nonlinearity": nn.Sigmoid() + }, + "optimizer": { + "name": Adam, + "args": { + "lr": 0.0002, + "betas": (0.5, 0.999) + } + } + } +} + +loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),] + +if torch.cuda.is_available(): + device = torch.device("cuda:0") + torch.backends.cudnn.deterministic = True + epochs = 20 +else: + device = torch.device("cpu") + epochs = 5 + +print("Device: {}".format(device)) +print("Epochs: {}".format(epochs)) + +trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device) + +trainer(dataloader) + +fig = plt.figure(figsize=(8,8)) +plt.axis("off") +ims = [[plt.imshow(plt.imread("{}/epoch{}_generator.png".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)] +ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True) +HTML(ani.to_jshtml()) + +gen = trainer.generator + +dis = trainer.discriminator + +for i in range(10): + x = torch.randn([1,100], device=device) + for k in range(1000): + xk = torch.randn([1,100], device=device) + a = 1/dis(gen(x, torch.Tensor([i]).cuda())) + b = 1/dis(gen(xk, torch.Tensor([i]).cuda())) + d = (a-1)/(b-1) + p = torch.rand([1,1], device=device) + if (p < min(1, d)): + x = xk + image = gen(x, torch.Tensor([i]).cuda()) + plt.figure() + plt.axis("off") + plt.title(i) + plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0))) \ No newline at end of file diff --git a/notebooks/mhgan/MHGAN_MNIST.ipynb b/notebooks/mhgan/MHGAN_MNIST.ipynb new file mode 100644 index 0000000..499b978 --- /dev/null +++ b/notebooks/mhgan/MHGAN_MNIST.ipynb @@ -0,0 +1,14779 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "MHGAN_MNIST.ipynb", + "version": "0.3.2", + "provenance": [], + "collapsed_sections": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "accelerator": "GPU" + }, + "cells": [ + { + "metadata": { + "id": "ReXdYV5Z6wNZ", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "!pip uninstall -y Pillow\n", + "!pip install Pillow==5.3.0\n", + "!pip install torchgan\n", + "!pip install tensorboardX" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "JxW667X0DwPE", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "import os\n", + "import random\n", + "import matplotlib.pyplot as plt\n", + "import matplotlib.animation as animation\n", + "import numpy as np\n", + "from IPython.display import HTML\n", + "import torch\n", + "import torch.nn as nn\n", + "import torchvision\n", + "from torch.optim import Adam\n", + "from torch.optim import SGD\n", + "import torch.nn as nn\n", + "import torch.utils.data as data\n", + "import torchvision.datasets as dsets\n", + "import torchvision.transforms as transforms\n", + "import torchvision.utils as vutils\n", + "import torchgan\n", + "from torchgan.models import *\n", + "from torchgan.losses import *\n", + "from torchgan.trainer import Trainer" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "XC8UpLxcEH-Y", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True)" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "yxOhiOGQGQON", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2)" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "My8wjH-oGYOZ", + "colab_type": "code", + "outputId": "3e7b1b78-d08d-43a1-c1c9-7a6c198ec5a7", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 498 + } + }, + "cell_type": "code", + "source": [ + "real_batch = next(iter(dataloader))\n", + "plt.figure(figsize=(8,8))\n", + "plt.axis(\"off\")\n", + "plt.title(\"Training Images\")\n", + "plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0)))\n", + "plt.show()" + ], + "execution_count": 5, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHhCAYAAAAibsTCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGIZh+In/B0GGiFVplUQhAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + } + ] + }, + { + "metadata": { + "id": "9Wn3sekuGf6T", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "acgan = {\n", + " \"generator\": {\n", + " \"name\": ACGANGenerator,\n", + " \"args\": {\n", + " \"encoding_dims\": 100,\n", + " \"num_classes\": 10,\n", + " \"out_channels\": 1,\n", + " \"step_channels\": 32,\n", + " \"out_size\":32,\n", + " \"nonlinearity\": nn.LeakyReLU(0.2),\n", + " \"last_nonlinearity\": nn.Tanh()\n", + " },\n", + " \"optimizer\": {\n", + " \"name\": Adam,\n", + " \"args\": {\n", + " \"lr\": 0.0009,\n", + " \"betas\": (0.5, 0.999)\n", + " }\n", + " }\n", + " },\n", + " \"discriminator\": {\n", + " \"name\": ACGANDiscriminator,\n", + " \"args\": {\n", + " \"in_channels\": 1,\n", + " \"step_channels\": 32,\n", + " \"in_size\": 32,\n", + " \"num_classes\": 10,\n", + " \"nonlinearity\": nn.LeakyReLU(0.2),\n", + " \"last_nonlinearity\": nn.Sigmoid()\n", + " },\n", + " \"optimizer\": {\n", + " \"name\": Adam,\n", + " \"args\": {\n", + " \"lr\": 0.0002,\n", + " \"betas\": (0.5, 0.999)\n", + " }\n", + " }\n", + " }\n", + "}" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "tVEjIUtxJdbc", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),]" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "LigSujXVJwZS", + "colab_type": "code", + "outputId": "e91c624f-267b-44b1-add6-114db8ffd0ae", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 51 + } + }, + "cell_type": "code", + "source": [ + "if torch.cuda.is_available():\n", + " device = torch.device(\"cuda:0\")\n", + " torch.backends.cudnn.deterministic = True\n", + " epochs = 20\n", + "else:\n", + " device = torch.device(\"cpu\")\n", + " epochs = 5\n", + "\n", + "print(\"Device: {}\".format(device))\n", + "print(\"Epochs: {}\".format(epochs))" + ], + "execution_count": 130, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Device: cuda:0\n", + "Epochs: 20\n" + ], + "name": "stdout" + } + ] + }, + { + "metadata": { + "id": "XegwsijSJ1jF", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device)" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "kTatr-LQJ9qa", + "colab_type": "code", + "outputId": "588aa7df-e61b-49ff-d00a-d383e7f0f39a", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1951 + } + }, + "cell_type": "code", + "source": [ + "trainer(dataloader)" + ], + "execution_count": 132, + "outputs": [ + { + "output_type": "stream", + "text": [ + "Saving Model at './model/gan0.model'\n", + "Epoch 1 Summary\n", + "generator Mean Gradients : 0.5629363154607244\n", + "discriminator Mean Gradients : 21.085883638233142\n", + "Mean Running Discriminator Loss : 2.132794382729764\n", + "Mean Running Generator Loss : 1.117769119852006\n", + "Generating and Saving Images to ./images/epoch1_generator.png\n", + "\n", + "Saving Model at './model/gan1.model'\n", + "Epoch 2 Summary\n", + "generator Mean Gradients : 0.4621404109377145\n", + "discriminator Mean Gradients : 37.54858841070317\n", + "Mean Running Discriminator Loss : 2.0983149196102677\n", + "Mean Running Generator Loss : 1.0905873073554877\n", + "Generating and Saving Images to ./images/epoch2_generator.png\n", + "\n", + "Saving Model at './model/gan2.model'\n", + "Epoch 3 Summary\n", + "generator Mean Gradients : 0.42224539835597896\n", + "discriminator Mean Gradients : 34.49435525860745\n", + "Mean Running Discriminator Loss : 2.097173844763973\n", + "Mean Running Generator Loss : 1.0768492916637307\n", + "Generating and Saving Images to ./images/epoch3_generator.png\n", + "\n", + "Saving Model at './model/gan3.model'\n", + "Epoch 4 Summary\n", + "generator Mean Gradients : 0.3910194754263064\n", + "discriminator Mean Gradients : 30.64363928034132\n", + "Mean Running Discriminator Loss : 2.0992902942907326\n", + "Mean Running Generator Loss : 1.069386471527567\n", + "Generating and Saving Images to ./images/epoch4_generator.png\n", + "\n", + "Saving Model at './model/gan4.model'\n", + "Epoch 5 Summary\n", + "generator Mean Gradients : 0.3700217700678151\n", + "discriminator Mean Gradients : 28.378661746196947\n", + "Mean Running Discriminator Loss : 2.1001694201152206\n", + "Mean Running Generator Loss : 1.0651295932053504\n", + "Generating and Saving Images to ./images/epoch5_generator.png\n", + "\n", + "Saving Model at './model/gan0.model'\n", + "Epoch 6 Summary\n", + "generator Mean Gradients : 0.3731452710248436\n", + "discriminator Mean Gradients : 27.391221197105324\n", + "Mean Running Discriminator Loss : 2.098479413941725\n", + "Mean Running Generator Loss : 1.0625621958158502\n", + "Generating and Saving Images to ./images/epoch6_generator.png\n", + "\n", + "Saving Model at './model/gan1.model'\n", + "Epoch 7 Summary\n", + "generator Mean Gradients : 0.4097920223127502\n", + "discriminator Mean Gradients : 27.55014014898362\n", + "Mean Running Discriminator Loss : 2.092044219824006\n", + "Mean Running Generator Loss : 1.0619102914250553\n", + "Generating and Saving Images to ./images/epoch7_generator.png\n", + "\n", + "Saving Model at './model/gan2.model'\n", + "Epoch 8 Summary\n", + "generator Mean Gradients : 0.42416808925331695\n", + "discriminator Mean Gradients : 27.253160953803246\n", + "Mean Running Discriminator Loss : 2.082712709395362\n", + "Mean Running Generator Loss : 1.0626732571832915\n", + "Generating and Saving Images to ./images/epoch8_generator.png\n", + "\n", + "Saving Model at './model/gan3.model'\n", + "Epoch 9 Summary\n", + "generator Mean Gradients : 0.37972263266086353\n", + "discriminator Mean Gradients : 24.476701898901787\n", + "Mean Running Discriminator Loss : 2.0820189374751656\n", + "Mean Running Generator Loss : 1.0642426933875244\n", + "Generating and Saving Images to ./images/epoch9_generator.png\n", + "\n", + "Saving Model at './model/gan4.model'\n", + "Epoch 10 Summary\n", + "generator Mean Gradients : 0.3470192597153898\n", + "discriminator Mean Gradients : 22.522278636586094\n", + "Mean Running Discriminator Loss : 2.0774675978208657\n", + "Mean Running Generator Loss : 1.0654202675784448\n", + "Generating and Saving Images to ./images/epoch10_generator.png\n", + "\n" + ], + "name": "stdout" + }, + { + "output_type": "error", + "ename": "KeyboardInterrupt", + "evalue": "ignored", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 436\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 437\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 438\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 398\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreal_inputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 399\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 400\u001b[0;31m \u001b[0mlgen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgen_iter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_iter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 401\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'generator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mlgen\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 402\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'discriminator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain_iter\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 334\u001b[0m \u001b[0mgrad_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate_grads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 335\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mDiscriminatorLoss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 336\u001b[0;31m \u001b[0mcur_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_ops\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_arguments\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_arg_maps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 337\u001b[0m \u001b[0mloss_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcur_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 338\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mldis\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mcur_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/losses/auxclassifier.py\u001b[0m in \u001b[0;36mtrain_ops\u001b[0;34m(self, generator, discriminator, optimizer_discriminator, real_inputs, device, labels)\u001b[0m\n\u001b[1;32m 140\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 141\u001b[0m \u001b[0moptimizer_discriminator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", + "\u001b[0;31mKeyboardInterrupt\u001b[0m: " + ] + } + ] + }, + { + "metadata": { + "id": "a8wXZRm-KDVr", + "colab_type": "code", + "outputId": "cd8736b9-0fa3-4def-929a-84882f9fc74e", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1136 + } + }, + "cell_type": "code", + "source": [ + "fig = plt.figure(figsize=(8,8))\n", + "plt.axis(\"off\")\n", + "ims = [[plt.imshow(plt.imread(\"{}/epoch{}_generator.png\".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)]\n", + "ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True)\n", + "HTML(ani.to_jshtml())" + ], + "execution_count": 133, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/html": [ + "\n", + "\n", + "\n", + "\n", + "
\n", + " \n", + "
\n", + " \n", + "
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
\n", + " Once \n", + " Loop \n", + " Reflect \n", + "
\n", + "
\n", + "\n", + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": { + "tags": [] + }, + "execution_count": 133 + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHVCAYAAAC9hYF5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkVVKKaVsoousUkopZZP/BwT3oBgGetiyAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + } + ] + }, + { + "metadata": { + "id": "YtjyFoEez8vl", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "gen = trainer.generator" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "t5ZrkNMVwmdI", + "colab_type": "code", + "colab": {} + }, + "cell_type": "code", + "source": [ + "dis = trainer.discriminator" + ], + "execution_count": 0, + "outputs": [] + }, + { + "metadata": { + "id": "dExgdnaG8MDl", + "colab_type": "code", + "outputId": "36d36215-ee6e-41e5-b46a-209c0645c9be", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 3637 + } + }, + "cell_type": "code", + "source": [ + "for i in range(10):\n", + " x = torch.randn([1,100], device=device)\n", + " for k in range(1000):\n", + " xk = torch.randn([1,100], device=device)\n", + " a = 1/dis(gen(x, torch.Tensor([i]).cuda()))\n", + " b = 1/dis(gen(xk, torch.Tensor([i]).cuda()))\n", + " d = (a-1)/(b-1)\n", + " p = torch.rand([1,1], device=device)\n", + " if (p < min(1, d)):\n", + " x = xk\n", + " image = gen(x, torch.Tensor([i]).cuda())\n", + " plt.figure()\n", + " plt.axis(\"off\")\n", + " plt.title(i)\n", + " plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0)))" + ], + "execution_count": 153, + "outputs": [ + { + "output_type": "stream", + "text": [ + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", + "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n" + ], + "name": "stderr" + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCRD4Bym7H3Oe+mRLAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngMDfvMqQ4k8FizsAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQCwBAmIJEBBLgIBYAgT+Ay0R0MvvvT5UAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxBIgIJYAAbEECIglQEAsAQL/A9upZjYozdLsAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAgTEEiDwD7l8ioHiDmDhAAAAAElFTkSuQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + }, + { + "output_type": "display_data", + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [] + } + } + ] + } + ] +} \ No newline at end of file From a648273c67bd0e6b550b7e865affec275f0a5480 Mon Sep 17 00:00:00 2001 From: k-k-d Date: Tue, 11 Jun 2019 00:13:24 +0530 Subject: [PATCH 6/8] Updated Directory Name --- models/mhgan/MHGAN_MNIST.py => mhgan/mhgan.py | 0 notebooks/mhgan/MHGAN_MNIST.ipynb | 14779 ---------------- 2 files changed, 14779 deletions(-) rename models/mhgan/MHGAN_MNIST.py => mhgan/mhgan.py (100%) delete mode 100644 notebooks/mhgan/MHGAN_MNIST.ipynb diff --git a/models/mhgan/MHGAN_MNIST.py b/mhgan/mhgan.py similarity index 100% rename from models/mhgan/MHGAN_MNIST.py rename to mhgan/mhgan.py diff --git a/notebooks/mhgan/MHGAN_MNIST.ipynb b/notebooks/mhgan/MHGAN_MNIST.ipynb deleted file mode 100644 index 499b978..0000000 --- a/notebooks/mhgan/MHGAN_MNIST.ipynb +++ /dev/null @@ -1,14779 +0,0 @@ -{ - "nbformat": 4, - "nbformat_minor": 0, - "metadata": { - "colab": { - "name": "MHGAN_MNIST.ipynb", - "version": "0.3.2", - "provenance": [], - "collapsed_sections": [] - }, - "kernelspec": { - "name": "python3", - "display_name": "Python 3" - }, - "accelerator": "GPU" - }, - "cells": [ - { - "metadata": { - "id": "ReXdYV5Z6wNZ", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "!pip uninstall -y Pillow\n", - "!pip install Pillow==5.3.0\n", - "!pip install torchgan\n", - "!pip install tensorboardX" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "JxW667X0DwPE", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "import os\n", - "import random\n", - "import matplotlib.pyplot as plt\n", - "import matplotlib.animation as animation\n", - "import numpy as np\n", - "from IPython.display import HTML\n", - "import torch\n", - "import torch.nn as nn\n", - "import torchvision\n", - "from torch.optim import Adam\n", - "from torch.optim import SGD\n", - "import torch.nn as nn\n", - "import torch.utils.data as data\n", - "import torchvision.datasets as dsets\n", - "import torchvision.transforms as transforms\n", - "import torchvision.utils as vutils\n", - "import torchgan\n", - "from torchgan.models import *\n", - "from torchgan.losses import *\n", - "from torchgan.trainer import Trainer" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "XC8UpLxcEH-Y", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True)" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "yxOhiOGQGQON", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2)" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "My8wjH-oGYOZ", - "colab_type": "code", - "outputId": "3e7b1b78-d08d-43a1-c1c9-7a6c198ec5a7", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 498 - } - }, - "cell_type": "code", - "source": [ - "real_batch = next(iter(dataloader))\n", - "plt.figure(figsize=(8,8))\n", - "plt.axis(\"off\")\n", - "plt.title(\"Training Images\")\n", - "plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0)))\n", - "plt.show()" - ], - "execution_count": 5, - "outputs": [ - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHhCAYAAAAibsTCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGIZh+In/B0GGiFVplUQhAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - } - ] - }, - { - "metadata": { - "id": "9Wn3sekuGf6T", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "acgan = {\n", - " \"generator\": {\n", - " \"name\": ACGANGenerator,\n", - " \"args\": {\n", - " \"encoding_dims\": 100,\n", - " \"num_classes\": 10,\n", - " \"out_channels\": 1,\n", - " \"step_channels\": 32,\n", - " \"out_size\":32,\n", - " \"nonlinearity\": nn.LeakyReLU(0.2),\n", - " \"last_nonlinearity\": nn.Tanh()\n", - " },\n", - " \"optimizer\": {\n", - " \"name\": Adam,\n", - " \"args\": {\n", - " \"lr\": 0.0009,\n", - " \"betas\": (0.5, 0.999)\n", - " }\n", - " }\n", - " },\n", - " \"discriminator\": {\n", - " \"name\": ACGANDiscriminator,\n", - " \"args\": {\n", - " \"in_channels\": 1,\n", - " \"step_channels\": 32,\n", - " \"in_size\": 32,\n", - " \"num_classes\": 10,\n", - " \"nonlinearity\": nn.LeakyReLU(0.2),\n", - " \"last_nonlinearity\": nn.Sigmoid()\n", - " },\n", - " \"optimizer\": {\n", - " \"name\": Adam,\n", - " \"args\": {\n", - " \"lr\": 0.0002,\n", - " \"betas\": (0.5, 0.999)\n", - " }\n", - " }\n", - " }\n", - "}" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "tVEjIUtxJdbc", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),]" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "LigSujXVJwZS", - "colab_type": "code", - "outputId": "e91c624f-267b-44b1-add6-114db8ffd0ae", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 51 - } - }, - "cell_type": "code", - "source": [ - "if torch.cuda.is_available():\n", - " device = torch.device(\"cuda:0\")\n", - " torch.backends.cudnn.deterministic = True\n", - " epochs = 20\n", - "else:\n", - " device = torch.device(\"cpu\")\n", - " epochs = 5\n", - "\n", - "print(\"Device: {}\".format(device))\n", - "print(\"Epochs: {}\".format(epochs))" - ], - "execution_count": 130, - "outputs": [ - { - "output_type": "stream", - "text": [ - "Device: cuda:0\n", - "Epochs: 20\n" - ], - "name": "stdout" - } - ] - }, - { - "metadata": { - "id": "XegwsijSJ1jF", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "trainer = Trainer(acgan, loss, sample_size=64, epochs=epochs, device=device)" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "kTatr-LQJ9qa", - "colab_type": "code", - "outputId": "588aa7df-e61b-49ff-d00a-d383e7f0f39a", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1951 - } - }, - "cell_type": "code", - "source": [ - "trainer(dataloader)" - ], - "execution_count": 132, - "outputs": [ - { - "output_type": "stream", - "text": [ - "Saving Model at './model/gan0.model'\n", - "Epoch 1 Summary\n", - "generator Mean Gradients : 0.5629363154607244\n", - "discriminator Mean Gradients : 21.085883638233142\n", - "Mean Running Discriminator Loss : 2.132794382729764\n", - "Mean Running Generator Loss : 1.117769119852006\n", - "Generating and Saving Images to ./images/epoch1_generator.png\n", - "\n", - "Saving Model at './model/gan1.model'\n", - "Epoch 2 Summary\n", - "generator Mean Gradients : 0.4621404109377145\n", - "discriminator Mean Gradients : 37.54858841070317\n", - "Mean Running Discriminator Loss : 2.0983149196102677\n", - "Mean Running Generator Loss : 1.0905873073554877\n", - "Generating and Saving Images to ./images/epoch2_generator.png\n", - "\n", - "Saving Model at './model/gan2.model'\n", - "Epoch 3 Summary\n", - "generator Mean Gradients : 0.42224539835597896\n", - "discriminator Mean Gradients : 34.49435525860745\n", - "Mean Running Discriminator Loss : 2.097173844763973\n", - "Mean Running Generator Loss : 1.0768492916637307\n", - "Generating and Saving Images to ./images/epoch3_generator.png\n", - "\n", - "Saving Model at './model/gan3.model'\n", - "Epoch 4 Summary\n", - "generator Mean Gradients : 0.3910194754263064\n", - "discriminator Mean Gradients : 30.64363928034132\n", - "Mean Running Discriminator Loss : 2.0992902942907326\n", - "Mean Running Generator Loss : 1.069386471527567\n", - "Generating and Saving Images to ./images/epoch4_generator.png\n", - "\n", - "Saving Model at './model/gan4.model'\n", - "Epoch 5 Summary\n", - "generator Mean Gradients : 0.3700217700678151\n", - "discriminator Mean Gradients : 28.378661746196947\n", - "Mean Running Discriminator Loss : 2.1001694201152206\n", - "Mean Running Generator Loss : 1.0651295932053504\n", - "Generating and Saving Images to ./images/epoch5_generator.png\n", - "\n", - "Saving Model at './model/gan0.model'\n", - "Epoch 6 Summary\n", - "generator Mean Gradients : 0.3731452710248436\n", - "discriminator Mean Gradients : 27.391221197105324\n", - "Mean Running Discriminator Loss : 2.098479413941725\n", - "Mean Running Generator Loss : 1.0625621958158502\n", - "Generating and Saving Images to ./images/epoch6_generator.png\n", - "\n", - "Saving Model at './model/gan1.model'\n", - "Epoch 7 Summary\n", - "generator Mean Gradients : 0.4097920223127502\n", - "discriminator Mean Gradients : 27.55014014898362\n", - "Mean Running Discriminator Loss : 2.092044219824006\n", - "Mean Running Generator Loss : 1.0619102914250553\n", - "Generating and Saving Images to ./images/epoch7_generator.png\n", - "\n", - "Saving Model at './model/gan2.model'\n", - "Epoch 8 Summary\n", - "generator Mean Gradients : 0.42416808925331695\n", - "discriminator Mean Gradients : 27.253160953803246\n", - "Mean Running Discriminator Loss : 2.082712709395362\n", - "Mean Running Generator Loss : 1.0626732571832915\n", - "Generating and Saving Images to ./images/epoch8_generator.png\n", - "\n", - "Saving Model at './model/gan3.model'\n", - "Epoch 9 Summary\n", - "generator Mean Gradients : 0.37972263266086353\n", - "discriminator Mean Gradients : 24.476701898901787\n", - "Mean Running Discriminator Loss : 2.0820189374751656\n", - "Mean Running Generator Loss : 1.0642426933875244\n", - "Generating and Saving Images to ./images/epoch9_generator.png\n", - "\n", - "Saving Model at './model/gan4.model'\n", - "Epoch 10 Summary\n", - "generator Mean Gradients : 0.3470192597153898\n", - "discriminator Mean Gradients : 22.522278636586094\n", - "Mean Running Discriminator Loss : 2.0774675978208657\n", - "Mean Running Generator Loss : 1.0654202675784448\n", - "Generating and Saving Images to ./images/epoch10_generator.png\n", - "\n" - ], - "name": "stdout" - }, - { - "output_type": "error", - "ename": "KeyboardInterrupt", - "evalue": "ignored", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", - "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtrainer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataloader\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 436\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 437\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata_loader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 438\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, data_loader, **kwargs)\u001b[0m\n\u001b[1;32m 398\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreal_inputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 399\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 400\u001b[0;31m \u001b[0mlgen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgen_iter\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_iter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 401\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'generator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mlgen\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 402\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_information\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'discriminator_losses'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/trainer/trainer.py\u001b[0m in \u001b[0;36mtrain_iter\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 334\u001b[0m \u001b[0mgrad_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate_grads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 335\u001b[0m \u001b[0;32melif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mDiscriminatorLoss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 336\u001b[0;31m \u001b[0mcur_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_ops\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_arguments\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_arg_maps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 337\u001b[0m \u001b[0mloss_logs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcur_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 338\u001b[0m \u001b[0mldis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mldis\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mcur_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdis_iter\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torchgan/losses/auxclassifier.py\u001b[0m in \u001b[0;36mtrain_ops\u001b[0;34m(self, generator, discriminator, optimizer_discriminator, real_inputs, device, labels)\u001b[0m\n\u001b[1;32m 140\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 141\u001b[0m \u001b[0moptimizer_discriminator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", - "\u001b[0;31mKeyboardInterrupt\u001b[0m: " - ] - } - ] - }, - { - "metadata": { - "id": "a8wXZRm-KDVr", - "colab_type": "code", - "outputId": "cd8736b9-0fa3-4def-929a-84882f9fc74e", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 1136 - } - }, - "cell_type": "code", - "source": [ - "fig = plt.figure(figsize=(8,8))\n", - "plt.axis(\"off\")\n", - "ims = [[plt.imshow(plt.imread(\"{}/epoch{}_generator.png\".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)]\n", - "ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True)\n", - "HTML(ani.to_jshtml())" - ], - "execution_count": 133, - "outputs": [ - { - "output_type": "execute_result", - "data": { - "text/html": [ - "\n", - "\n", - "\n", - "\n", - "
\n", - " \n", - "
\n", - " \n", - "
\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
\n", - " Once \n", - " Loop \n", - " Reflect \n", - "
\n", - "
\n", - "\n", - "\n", - "\n" - ], - "text/plain": [ - "" - ] - }, - "metadata": { - "tags": [] - }, - "execution_count": 133 - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAHVCAYAAAC9hYF5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkVVKKaVsoousUkopZZP/BwT3oBgGetiyAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - } - ] - }, - { - "metadata": { - "id": "YtjyFoEez8vl", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "gen = trainer.generator" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "t5ZrkNMVwmdI", - "colab_type": "code", - "colab": {} - }, - "cell_type": "code", - "source": [ - "dis = trainer.discriminator" - ], - "execution_count": 0, - "outputs": [] - }, - { - "metadata": { - "id": "dExgdnaG8MDl", - "colab_type": "code", - "outputId": "36d36215-ee6e-41e5-b46a-209c0645c9be", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 3637 - } - }, - "cell_type": "code", - "source": [ - "for i in range(10):\n", - " x = torch.randn([1,100], device=device)\n", - " for k in range(1000):\n", - " xk = torch.randn([1,100], device=device)\n", - " a = 1/dis(gen(x, torch.Tensor([i]).cuda()))\n", - " b = 1/dis(gen(xk, torch.Tensor([i]).cuda()))\n", - " d = (a-1)/(b-1)\n", - " p = torch.rand([1,1], device=device)\n", - " if (p < min(1, d)):\n", - " x = xk\n", - " image = gen(x, torch.Tensor([i]).cuda())\n", - " plt.figure()\n", - " plt.axis(\"off\")\n", - " plt.title(i)\n", - " plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0)))" - ], - "execution_count": 153, - "outputs": [ - { - "output_type": "stream", - "text": [ - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n", - "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n" - ], - "name": "stderr" - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCRD4Bym7H3Oe+mRLAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAChJJREFUeJzt3U+IldUfx/Fzx3EclNpkao5OQbVI\nQ6FFVAqiIggNRVFBuQlauaiF/9ok0S6IrFXtwgFnpyiIg0YxFOQiatcirKlFMZSLUXKaxcw0t82P\n33LOJ+6dud65r9f6yzkHg3dHeZ77NJrNZrMAsKS+Th8AoBuIJUBALAECYgkQEEuAgFgCBMSSrnPt\n2rXy/PPPl8OHD5dXX3213Lhxo9NHogc0PGdJN5mamiovvvhiuXDhQhkaGiqjo6Pl8uXL5fz5850+\nGqucmyVdpb+/v3z44YdlaGiolFLK008/XX799dcOn4pe0N/pA8B/sWnTprJp06ZSSikLCwvl4sWL\n5eDBgx0+Fb3AzZKuNDo6Wvbs2VO+++67cuLEiU4fhx7g3yzpWs1ms1y5cqWcOXOmjI+Pl8HBwU4f\niVXMzZKuMjk5Wa5fv15KKaXRaJSRkZHy999/+3dLlp1Y0lWmp6fLqVOnyp9//llKKeX7778v8/Pz\nZfv27R0+Gaudv4bTdcbGxsrY2FhZXFwsAwMD5fjx42Xfvn2dPharnFgCBPw1HCAglgABsQQIiCVA\nQCwBAivybnij0ViJbQBastTDQW6WAAGxBAiIJUBALAECYgkQEEuAgFgCBMQSICCWAAGxBAiIJUBA\nLAECYgkQEEuAgFgCBMQSICCWAAGxBAiIJUBALAECYgkQEEuAgFgCBMQSICCWAAGxBAiIJUCgv9MH\nYHmcPHkymvvggw+W+SSwOrhZAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgECj2Ww2l32TRmO5t+gp\nH330UXVmZGQkWuvRRx9t9TiwaiyVQzdLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIDPSnSh\nZ555pjqzdevWaK1jx45VZ86cOROtBauZmyVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIeCi9C83O\nzlZnbt++Ha21e/fuVo8DPcHNEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgQazWazueybNBrLvUVP\nSf48P/nkk2itqamp6sxXX31Vnfn666+j/dol3W/btm3Vmenp6WitF154oTrz22+/RWtxd1oqh26W\nAAGxBAiIJUBALAECYgkQEEuAgFgCBMQSICCWAAGflehCyUtXR48ejdbatWtXdebWrVvRWu2yfv36\n6swTTzwRrbW4uFiduXTpUrSWt3N6m5slQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECPisBF3p448/\njubm5uaqM6dOnYrW6u+vv8Nx7ty56szGjRuj/QYHB6sze/fujdYi47MSAC0SS4CAWAIExBIgIJYA\nAbEECIglQEAsAQIeSofQe++9V5158803qzMDAwPRfjMzM9WZLVu2RGuR8VA6QIvEEiAglgABsQQI\niCVAQCwBAmIJEBBLgIBYAgTqv5MPq9z27dujueRtmeRTEGvWrIn2u3XrVnXm9ddfr86cPXs22o+l\nuVkCBMQSICCWAAGxBAiIJUBALAECYgkQEEuAgM9K0PNee+21aO7dd9+tzjzyyCOtHuf/bt682ZaZ\n3bt3t+M4PcFnJQBaJJYAAbEECIglQEAsAQJiCRAQS4CAWAIEPJTOqnby5MnqzJEjR6K1duzYUZ3p\n71/Zjw/Mz89XZx5++OFord9//73V43Q9D6UDtEgsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQMAb\nPD3u3nvvrc4cPHiwOvP4449H+83OzlZn7rvvvurM8PBwtN9TTz1VnRkaGorWWrduXTR3t/n222+j\nueTParXzBg9Ai8QSICCWAAGxBAiIJUBALAECYgkQEEuAwMr+Bj53ndOnT1dndu7cWZ2ZnJyM9hsf\nH6/OHDhwoDrzxRdfRPstLCxUZ5577rlorTVr1rRlZnFxMdqvr69+l0le+Hjsscei/ViamyVAQCwB\nAmIJEBBLgIBYAgTEEiAglgABsQQI+KV0aKPDhw9XZ1566aVorWQu+aX75MH8UkoZGRmpznz++efR\nWt3KL6UDtEgsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQMBnJaCNrl692paZUkpZv359deaVV16p\nziSfuiillP3791dnVvsbPEtxswQIiCVAQCwBAmIJEBBLgIBYAgTEEiAglgABn5WALjY3N1ed6e/P\n3j15+eWXqzMXLlyI1upWPisB0CKxBAiIJUBALAECYgkQEEuAgFgCBMQSICCWAAGflYAu1tdXv++k\nL+nNzMy0epxVzc0SICCWAAGxBAiIJUBALAECYgkQEEuAgFgCBHxWArpY8lmJ+fn5aK0NGza0epyu\n57MSAC0SS4CAWAIExBIgIJYAAbEECIglQEAsAQJ+KR3uUseOHavOLCwsVGe+/PLLdhyn57lZAgTE\nEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgS8wbNC0p/sf/LJJ6szExMTrR6HDnrjjTeiuXfeeac6\nMzMzU515//33o/1YmpslQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECHgofYV888030dyDDz5Ynfns\ns8+qM8ePH4/2I7d58+bqzPj4eHXmgQceiPa75557qjOXL1+uzly/fj3aj6W5WQIExBIgIJYAAbEE\nCIglQEAsAQJiCRAQS4BAo9lsNpd9k0ZjubfoqMHBwerM7du3o7UGBgZaPc5/2u+ff/6pziT//dau\nXRvtl6z1ww8/VGdmZ2ej/RYXF6szW7Zsida6//77qzMbN26M1kr89ddf1ZmHHnqoOnPnzp02nKY3\nLJVDN0uAgFgCBMQSICCWAAGxBAiIJUBALAECYgkQEEuAgM9KtMG6deuqM3/88Ue01vDwcHUmeekq\n+SRBKaXMzc21Zb/kTaBSsjeU+vrq/w//6aefov2St6uSt2BKyd5SSv6spqamov3efvvt6oy3c1aO\nmyVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQI+KzEXeaXX36pzvz888/Vmenp6Wi/5KH0nTt3Vmcm\nJyej/SYmJqozn376abRWuxw6dCiae/bZZ6szO3bsqM689dZb0X4//vhjNEf7+KwEQIvEEiAglgAB\nsQQIiCVAQCwBAmIJEBBLgICH0gH+x0PpAC0SS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CA\nWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIE\nxBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQEAsAQJiCRAQS4CAWAIExBIg\nIJYAAbEECPSvxCbNZnMltgFYNm6WAAGxBAiIJUBALAECYgkQEEuAgFgCBMQSICCWAAGxBAiIJUBA\nLAECYgkQEEuAgFgCBMQSICCWAAGxBAiIJUBALAECYgkQEEuAgFgCBP4FQN3H+ylf4HIAAAAASUVO\nRK5CYII=\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAACbJJREFUeJzt3U9o1/UDx/HP9+cmtrE2GDgMFsUK\nBpUhGCISJAUbFSXGqpug5L9bnlREQvDSKbBbeVOsdYmBo0MHd4nAQ6ATPLRGTUgaCZZbbOj32+XH\nD/od9n79fvv73R6Po7z4vD+nJ5/h58O31mg0GhUAC/rXat8AQDMQS4CAWAIExBIgIJYAAbEECLSs\n9g3A/+LOnTvVwMBA1dvb+59/2759e/Xxxx+v4l2xEYglTaenp6f65ptvVvs22GD8GQ4QEEuazoMH\nD6rjx49Xg4OD1aFDh6qJiYnVviU2ALGkqbS3t1dvvvlmdfr06Wp0dLTas2dPdfz48erhw4erfWus\nczXfhtPMGo1GtXPnzurLL7+snnnmmdW+HdYxT5Y0lfv371dTU1P/+Ld6vV61tPi/SpaXWNJUbt68\nWR04cKC6d+9eVVVVNTw8XG3btu0frxLBcvBnOE3n888/r7766quqVqtVPT091dmzZ6u+vr7Vvi3W\nObEECPgzHCAglgABsQQIiCVAQCwBAivyJm+tVluJYwAWZaGXgzxZAgTEEiAglgABsQQIiCVAQCwB\nAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJ\nEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBL\ngIBYAgTEEiAglgABsQQIiCVAQCwBAi2rfQPA8jp8+HC0e+WVV4qbo0ePFjd//PFHdF6z8WQJEBBL\ngIBYAgTEEiAglgABsQQIiCVAQCwBAl5KZ8157733ipu7d+9G1xobG1vs7Sy5tra2aNfd3V3cvPPO\nO8XNBx98EJ3X3t5e3AwNDRU3Fy9ejM5rNp4sAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQMBL6SyJ\n69evR7sXXnihuNm0aVNx02g0ovNmZ2eLm1qtFl3rzz//LG4ePXpU3MzMzETndXZ2FjcTExPFzdtv\nvx2d9+OPP0a7jcqTJUBALAECYgkQEEuAgFgCBMQSICCWAAGxBAiIJUDAFzwbXE9PT3Fz69at4qar\nqys6L/nypl6vFzfJlzJVlX0tMzU1FV3ru+++K24mJyeLm6tXr0bn/fTTT9GOleHJEiAglgABsQQI\niCVAQCwBAmIJEBBLgIBYAgS8lL7BffLJJ8VNe3t7cXP//v3ovEuXLhU3V65cKW6+//776DxYKp4s\nAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQMBL6RtcX19fcZO8cH7+/PnovAsXLkQ7WGs8WQIExBIg\nIJYAAbEECIglQEAsAQJiCRAQS4CAWAIEfMGzTr3++uvRrl6vFzcTExPFzbVr16LzoFl5sgQIiCVA\nQCwBAmIJEBBLgIBYAgTEEiAglgABL6WvU3v37o12Tz/9dHEzPz9f3Bw8eDA678MPP4x2K2n37t3R\n7vHHHy9ubty4Udz8+uuv0XmsLZ4sAQJiCRAQS4CAWAIExBIgIJYAAbEECIglQKDWaDQay35Irbbc\nR/BfRkZGot2LL75Y3HR0dBQ3v/zyS3Te3NxccTM6OlrcPPfcc9F5W7duLW6effbZ6Fp37twpbq5e\nvVrcnDt3LjqPlbdQDj1ZAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgT8rMQ69dZbb0W7oaGh\n4mbXrl3FzczMTHTe5ORkcZN8VdTb2xud19/fX9y0t7dH12pra4t2JenPSnz22WdLch5Lw5MlQEAs\nAQJiCRAQS4CAWAIExBIgIJYAAbEECPhZCda15CcjxsbGomt1dXUVN9PT08XN+Ph4dN4bb7wR7Vg6\nflYCYJHEEiAglgABsQQIiCVAQCwBAmIJEBBLgICX0iF0+/bt4iZ5cb1er0fnjYyMFDdHjx6NrkXG\nS+kAiySWAAGxBAiIJUBALAECYgkQEEuAgFgCBMQSIOALHgi9//77xc3Zs2eLm82bN0fnJbsnn3wy\nuhYZX/AALJJYAgTEEiAglgABsQQIiCVAQCwBAmIJEGhZ7RuAZvHFF18UN88//3xxMzAwEJ332GOP\nRTtWhidLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIAveGAJnTlzprhJf1Yi+dInudb8/Hx0\nHgvzZAkQEEuAgFgCBMQSICCWAAGxBAiIJUBALAECXkqHFfbDDz9Eu3379hU3586dK25OnjwZncfC\nPFkCBMQSICCWAAGxBAiIJUBALAECYgkQEEuAgJfSYYX9/vvv0a6jo6O42bVr12Jvh5AnS4CAWAIE\nxBIgIJYAAbEECIglQEAsAQJiCRAQS4CAL3ia0GuvvVbcfPvttytwJ/w/duzYEe1aW1uLm+7u7sXe\nDiFPlgABsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiDgpfQ15qWXXipuPvroo+LGS+lr1+DgYLTbsmVL\ncbNt27bF3g4hT5YAAbEECIglQEAsAQJiCRAQS4CAWAIExBIgUGs0Go1lP6RWW+4j1o27d+8WN52d\nncXN8PBwdN6BAweiHZnx8fHi5qmnnoquNTc3V9xMT08XN/39/dF5VNVCOfRkCRAQS4CAWAIExBIg\nIJYAAbEECIglQEAsAQJiCRDwBc8a8/PPPxc33d3dxc3s7Gx03okTJ4qbS5cuRddai1pbW4ubI0eO\nRNc6duxYcfPEE08UN7dv347O+/TTT4uby5cvR9ci4wsegEUSS4CAWAIExBIgIJYAAbEECIglQEAs\nAQJeSm9Chw4dKm72798fXaterxc3Dx8+XJLrVFVV9fT0FDebN28ubpKf1qiqqtq0aVNxk77A/9tv\nvxU3169fL25OnToVncfK81I6wCKJJUBALAECYgkQEEuAgFgCBMQSICCWAAEvpW9wycvd7777bnHz\n6quvRue9/PLLxc21a9eKm9HR0ei85GX5r7/+OrrWX3/9Fe1oXl5KB1gksQQIiCVAQCwBAmIJEBBL\ngIBYAgTEEiAglgABX/AA/JsveAAWSSwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBL\ngIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBY\nAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgEDLShzSaDRW4hiAZePJEiAglgABsQQIiCVAQCwBAmIJ\nEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBLgIBYAgTEEiAglgABsQQIiCVAQCwBAmIJEBBL\ngMDfvMqQ4k8FizsAAAAASUVORK5CYII=\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQCwBAmIJEBBLgIBYAgT+Ay0R0MvvvT5UAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxBIgIJYAAbEECIglQEAsAQL/A9upZjYozdLsAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAgTEEiDwD7l8ioHiDmDhAAAAAElFTkSuQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - }, - { - "output_type": "display_data", - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAFZCAYAAAARqQ0OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "tags": [] - } - } - ] - } - ] -} \ No newline at end of file From 53800d107150fe0d20b1dd079755b1f0240715bf Mon Sep 17 00:00:00 2001 From: k-k-d Date: Tue, 11 Jun 2019 00:43:44 +0530 Subject: [PATCH 7/8] isort and yapf done readme updated --- README.md | 4 +++ mhgan/mhgan.py | 85 ++++++++++++++++++++++++++++++++------------------ 2 files changed, 58 insertions(+), 31 deletions(-) diff --git a/README.md b/README.md index 86625ae..e669796 100644 --- a/README.md +++ b/README.md @@ -12,6 +12,10 @@ Collection of Generative Adversarial Networks developed using TorchGAN [Link to Paper](https://arxiv.org/pdf/1810.04714.pdf)\ Requires the `torchgan` master. +3. Metropolis-Hastings GAN (MHGAN)\ + [Link to Paper](https://arxiv.org/pdf/1811.11357.pdf)\ + Requires the `torchgan` master. + ## Contribution Guidelines We are open to accepting any model that you have built. The only things diff --git a/mhgan/mhgan.py b/mhgan/mhgan.py index c2c6c28..23fe5a4 100644 --- a/mhgan/mhgan.py +++ b/mhgan/mhgan.py @@ -1,33 +1,45 @@ import os import random -import matplotlib.pyplot as plt + import matplotlib.animation as animation +import matplotlib.pyplot as plt import numpy as np -from IPython.display import HTML import torch import torch.nn as nn -import torchvision -from torch.optim import Adam -from torch.optim import SGD -import torch.nn as nn import torch.utils.data as data +import torchgan +import torchvision import torchvision.datasets as dsets import torchvision.transforms as transforms import torchvision.utils as vutils -import torchgan -from torchgan.models import * +from IPython.display import HTML +from torch.optim import SGD, Adam from torchgan.losses import * +from torchgan.models import * from torchgan.trainer import Trainer -dataset = dsets.MNIST(root='./mnist', train=True, transform=transforms.Compose([transforms.Resize((32, 32)), transforms.ToTensor(), transforms.Normalize(mean = (0.5,), std = (0.5,))]), download=True) +dataset = dsets.MNIST(root='./mnist', + train=True, + transform=transforms.Compose([ + transforms.Resize((32, 32)), + transforms.ToTensor(), + transforms.Normalize(mean=(0.5, ), std=(0.5, )) + ]), + download=True) -dataloader = data.DataLoader(dataset, batch_size=64, shuffle=True, num_workers=2) +dataloader = data.DataLoader(dataset, + batch_size=64, + shuffle=True, + num_workers=2) real_batch = next(iter(dataloader)) -plt.figure(figsize=(8,8)) +plt.figure(figsize=(8, 8)) plt.axis("off") plt.title("Training Images") -plt.imshow(np.transpose(vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(),(1,2,0))) +plt.imshow( + np.transpose( + vutils.make_grid(real_batch[0][:64], padding=2, normalize=True).cpu(), + (1, 2, 0))) plt.show() acgan = { @@ -38,7 +50,7 @@ "num_classes": 10, "out_channels": 1, "step_channels": 32, - "out_size":32, + "out_size": 32, "nonlinearity": nn.LeakyReLU(0.2), "last_nonlinearity": nn.Tanh() }, @@ -70,7 +82,12 @@ } } -loss = [MinimaxDiscriminatorLoss(), MinimaxGeneratorLoss(), AuxiliaryClassifierGeneratorLoss(), AuxiliaryClassifierDiscriminatorLoss(),] +loss = [ + MinimaxDiscriminatorLoss(), + MinimaxGeneratorLoss(), + AuxiliaryClassifierGeneratorLoss(), + AuxiliaryClassifierDiscriminatorLoss(), +] if torch.cuda.is_available(): device = torch.device("cuda:0") @@ -87,10 +104,16 @@ trainer(dataloader) -fig = plt.figure(figsize=(8,8)) +fig = plt.figure(figsize=(8, 8)) plt.axis("off") -ims = [[plt.imshow(plt.imread("{}/epoch{}_generator.png".format(trainer.recon, i)))] for i in range(1, trainer.epochs + 1)] -ani = animation.ArtistAnimation(fig, ims, interval=1000, repeat_delay=1000, blit=True) +ims = [[ + plt.imshow(plt.imread("{}/epoch{}_generator.png".format(trainer.recon, i))) +] for i in range(1, trainer.epochs + 1)] +ani = animation.ArtistAnimation(fig, + ims, + interval=1000, + repeat_delay=1000, + blit=True) HTML(ani.to_jshtml()) gen = trainer.generator @@ -98,17 +121,17 @@ dis = trainer.discriminator for i in range(10): - x = torch.randn([1,100], device=device) - for k in range(1000): - xk = torch.randn([1,100], device=device) - a = 1/dis(gen(x, torch.Tensor([i]).cuda())) - b = 1/dis(gen(xk, torch.Tensor([i]).cuda())) - d = (a-1)/(b-1) - p = torch.rand([1,1], device=device) - if (p < min(1, d)): - x = xk - image = gen(x, torch.Tensor([i]).cuda()) - plt.figure() - plt.axis("off") - plt.title(i) - plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(),(1,2,0))) \ No newline at end of file + x = torch.randn([1, 100], device=device) + for k in range(1000): + xk = torch.randn([1, 100], device=device) + a = 1 / dis(gen(x, torch.Tensor([i]).cuda())) + b = 1 / dis(gen(xk, torch.Tensor([i]).cuda())) + d = (a - 1) / (b - 1) + p = torch.rand([1, 1], device=device) + if (p < min(1, d)): + x = xk + image = gen(x, torch.Tensor([i]).cuda()) + plt.figure() + plt.axis("off") + plt.title(i) + plt.imshow(np.transpose(vutils.make_grid(image.detach()).cpu(), (1, 2, 0))) From 303c751aa56e971f14224164c70f37bd0410998a Mon Sep 17 00:00:00 2001 From: k-k-d Date: Tue, 11 Jun 2019 13:45:59 +0530 Subject: [PATCH 8/8] removed unnecessary imports --- mhgan/mhgan.py | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/mhgan/mhgan.py b/mhgan/mhgan.py index 23fe5a4..b44641a 100644 --- a/mhgan/mhgan.py +++ b/mhgan/mhgan.py @@ -1,21 +1,18 @@ -import os -import random - import matplotlib.animation as animation import matplotlib.pyplot as plt import numpy as np import torch import torch.nn as nn import torch.utils.data as data -import torchgan -import torchvision import torchvision.datasets as dsets import torchvision.transforms as transforms import torchvision.utils as vutils from IPython.display import HTML -from torch.optim import SGD, Adam -from torchgan.losses import * -from torchgan.models import * +from torch.optim import Adam +from torchgan.losses import (AuxiliaryClassifierDiscriminatorLoss, + AuxiliaryClassifierGeneratorLoss, + MinimaxDiscriminatorLoss, MinimaxGeneratorLoss) +from torchgan.models import ACGANDiscriminator, ACGANGenerator from torchgan.trainer import Trainer dataset = dsets.MNIST(root='./mnist',