CUT.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "CUT.ipynb",
      "provenance": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/dvschultz/ml-art-colabs/blob/master/CUT.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TVR4bEt9Ptel"
      },
      "source": [
        "#CUT\n",
        "This new model uses contrastive learning (the hot technique of the moment) to better train unaligned image to image translation (i.e. CycleGAN)\n",
        "\n",
        "[GitHub](https://github.com/taesungp/contrastive-unpaired-translation)\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "If you find this notebook useful, consider signing up for my [Patreon](https://www.patreon.com/bustbright) or [YouTube channel](https://www.youtube.com/channel/UCaZuPdmZ380SFUMKHVsv_AA/join). You can also send me a one-time payment on [Venmo](https://venmo.com/Derrick-Schultz)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Jb21_nDZpRQj"
      },
      "source": [
        "## Setup"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0pNLe_yFHKrt",
        "outputId": "743103d1-70b8-446e-b52b-d0a660746743",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "!nvidia-smi -L"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "GPU 0: Tesla V100-SXM2-16GB (UUID: GPU-e51e1631-d8eb-e75c-09bb-75f625e70019)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OeP3P_xFH1KS",
        "outputId": "e7f460f2-16ca-4d81-9c37-b5e98c57db01",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gwhG3nHjpPc2"
      },
      "source": [
        "## Install repo"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mhz4EenMH3nR",
        "outputId": "7e6ce815-1301-44dd-8505-e05803b5b3fe",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "import os\n",
        "\n",
        "if os.path.isdir(\"/content/drive/MyDrive/colab-cut/CUT/\"):\n",
        "    %cd /content/drive/MyDrive/colab-cut/CUT/\n",
        "elif os.path.isdir(\"/content/drive/\"):\n",
        "    !mkdir /content/drive/MyDrive/colab-cut\n",
        "    %cd /content/drive/MyDrive/colab-cut\n",
        "    !git clone https://github.com/taesungp/contrastive-unpaired-translation CUT\n",
        "    %cd CUT\n",
        "else:\n",
        "    print('Drive not mounted, installing in /content/')  \n",
        "    !git clone https://github.com/taesungp/contrastive-unpaired-translation CUT\n",
        "    %cd CUT\n",
        "\n",
        "#install dependencies\n",
        "!pip install dominate visdom GPUtil"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/drive/MyDrive/colab-cut\n",
            "Cloning into 'CUT'...\n",
            "remote: Enumerating objects: 3, done.\u001b[K\n",
            "remote: Counting objects: 100% (3/3), done.\u001b[K\n",
            "remote: Compressing objects: 100% (3/3), done.\u001b[K\n",
            "remote: Total 238 (delta 0), reused 2 (delta 0), pack-reused 235\u001b[K\n",
            "Receiving objects: 100% (238/238), 17.90 MiB | 5.59 MiB/s, done.\n",
            "Resolving deltas: 100% (115/115), done.\n",
            "/content/drive/MyDrive/colab-cut/CUT\n",
            "Requirement already satisfied: dominate in /usr/local/lib/python3.7/dist-packages (2.6.0)\n",
            "Requirement already satisfied: visdom in /usr/local/lib/python3.7/dist-packages (0.1.8.9)\n",
            "Requirement already satisfied: GPUtil in /usr/local/lib/python3.7/dist-packages (1.4.0)\n",
            "Requirement already satisfied: websocket-client in /usr/local/lib/python3.7/dist-packages (from visdom) (0.57.0)\n",
            "Requirement already satisfied: pyzmq in /usr/local/lib/python3.7/dist-packages (from visdom) (22.0.3)\n",
            "Requirement already satisfied: jsonpatch in /usr/local/lib/python3.7/dist-packages (from visdom) (1.28)\n",
            "Requirement already satisfied: pillow in /usr/local/lib/python3.7/dist-packages (from visdom) (7.0.0)\n",
            "Requirement already satisfied: torchfile in /usr/local/lib/python3.7/dist-packages (from visdom) (0.1.0)\n",
            "Requirement already satisfied: tornado in /usr/local/lib/python3.7/dist-packages (from visdom) (5.1.1)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from visdom) (2.23.0)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from visdom) (1.4.1)\n",
            "Requirement already satisfied: numpy>=1.8 in /usr/local/lib/python3.7/dist-packages (from visdom) (1.19.5)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from visdom) (1.15.0)\n",
            "Requirement already satisfied: jsonpointer>=1.9 in /usr/local/lib/python3.7/dist-packages (from jsonpatch->visdom) (2.0)\n",
            "Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->visdom) (3.0.4)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->visdom) (2020.12.5)\n",
            "Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->visdom) (2.10)\n",
            "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->visdom) (1.24.3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n1tdZoGaMOGf"
      },
      "source": [
        "## Train\n",
        "Create a folder in `./datasets` and put two folders inside it. `trainA` should contain images from one domain and `trainB` should contain images from another domain. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uRI0XL6xK386"
      },
      "source": [
        "#example: download the russian blue cats to grumpy cats dataset\n",
        "!bash ./datasets/download_cut_dataset.sh grumpifycat"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vMeqDCKYJna9"
      },
      "source": [
        "!cp -avr /content/drive/MyDrive/colab-munit/MUNIT/datasets/tjpd2seguy/ /content/drive/MyDrive/colab-cut/CUT/datasets/desjpn2seguy/"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f9ItLTLEUSWG"
      },
      "source": [
        "This code repo is based on the [popular pix2pix and CycleGAN repo](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix). Many of the options and settings are the same. There are a few helpful pages in that repo that are applicable here as well:\n",
        "\n",
        "*   [Training and Testing Tips](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/docs/tips.md)\n",
        "*   [FAQ](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/docs/qa.md)\n",
        "\n",
        "There are also many other options to check out in the files contained in the `options` folder.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t_npjQWxMjkr"
      },
      "source": [
        "dataset_path = './datasets/desjpn2seguy'\n",
        "project_name = 'desjpn2seguy'\n",
        "load_size = 1080\n",
        "crop_size = 512"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6WEJ3xy9K8ck",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8aaa2ec5-eb45-499e-f727-2abb6dc15662"
      },
      "source": [
        "!python train.py --dataroot $dataset_path --name $project_name --CUT_mode CUT --load_size $load_size --crop_size $crop_size --display_id 0"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "----------------- Options ---------------\n",
            "                 CUT_mode: CUT                           \n",
            "               batch_size: 1                             \n",
            "                    beta1: 0.5                           \n",
            "                    beta2: 0.999                         \n",
            "          checkpoints_dir: ./checkpoints                 \n",
            "           continue_train: False                         \n",
            "                crop_size: 512                           \t[default: 256]\n",
            "                 dataroot: ./datasets/desjpn2seguy       \t[default: placeholder]\n",
            "             dataset_mode: unaligned                     \n",
            "                direction: AtoB                          \n",
            "              display_env: main                          \n",
            "             display_freq: 400                           \n",
            "               display_id: 0                             \t[default: None]\n",
            "            display_ncols: 4                             \n",
            "             display_port: 8097                          \n",
            "           display_server: http://localhost              \n",
            "          display_winsize: 256                           \n",
            "               easy_label: experiment_name               \n",
            "                    epoch: latest                        \n",
            "              epoch_count: 1                             \n",
            "          evaluation_freq: 5000                          \n",
            "        flip_equivariance: False                         \n",
            "                 gan_mode: lsgan                         \n",
            "                  gpu_ids: 0                             \n",
            "                init_gain: 0.02                          \n",
            "                init_type: xavier                        \n",
            "                 input_nc: 3                             \n",
            "                  isTrain: True                          \t[default: None]\n",
            "               lambda_GAN: 1.0                           \n",
            "               lambda_NCE: 1.0                           \n",
            "                load_size: 1080                          \t[default: 286]\n",
            "                       lr: 0.0002                        \n",
            "           lr_decay_iters: 50                            \n",
            "                lr_policy: linear                        \n",
            "         max_dataset_size: inf                           \n",
            "                    model: cut                           \n",
            "                 n_epochs: 200                           \n",
            "           n_epochs_decay: 200                           \n",
            "               n_layers_D: 3                             \n",
            "                     name: desjpn2seguy                  \t[default: experiment_name]\n",
            "                    nce_T: 0.07                          \n",
            "                  nce_idt: True                          \n",
            "nce_includes_all_negatives_from_minibatch: False                         \n",
            "               nce_layers: 0,4,8,12,16                   \n",
            "                      ndf: 64                            \n",
            "                     netD: basic                         \n",
            "                     netF: mlp_sample                    \n",
            "                  netF_nc: 256                           \n",
            "                     netG: resnet_9blocks                \n",
            "                      ngf: 64                            \n",
            "             no_antialias: False                         \n",
            "          no_antialias_up: False                         \n",
            "               no_dropout: True                          \n",
            "                  no_flip: False                         \n",
            "                  no_html: False                         \n",
            "                    normD: instance                      \n",
            "                    normG: instance                      \n",
            "              num_patches: 256                           \n",
            "              num_threads: 4                             \n",
            "                output_nc: 3                             \n",
            "                    phase: train                         \n",
            "                pool_size: 0                             \n",
            "               preprocess: resize_and_crop               \n",
            "          pretrained_name: None                          \n",
            "               print_freq: 100                           \n",
            "         random_scale_max: 3.0                           \n",
            "             save_by_iter: False                         \n",
            "          save_epoch_freq: 5                             \n",
            "         save_latest_freq: 5000                          \n",
            "           serial_batches: False                         \n",
            "stylegan2_G_num_downsampling: 1                             \n",
            "                   suffix:                               \n",
            "         update_html_freq: 1000                          \n",
            "                  verbose: False                         \n",
            "----------------- End -------------------\n",
            "dataset [UnalignedDataset] was created\n",
            "model [CUTModel] was created\n",
            "The number of training images = 303\n",
            "create web directory ./checkpoints/desjpn2seguy/web...\n",
            "---------- Networks initialized -------------\n",
            "[Network G] Total number of parameters : 11.378 M\n",
            "[Network F] Total number of parameters : 0.560 M\n",
            "[Network D] Total number of parameters : 2.765 M\n",
            "-----------------------------------------------\n",
            "(epoch: 1, iters: 100, time: 0.265, data: 0.816) G_GAN: 0.274 D_real: 0.309 D_fake: 0.209 G: 4.490 NCE: 4.308 NCE_Y: 4.123 \n",
            "(epoch: 1, iters: 200, time: 0.349, data: 0.002) G_GAN: 0.355 D_real: 0.255 D_fake: 0.247 G: 3.734 NCE: 3.461 NCE_Y: 3.298 \n",
            "(epoch: 1, iters: 300, time: 0.400, data: 0.002) G_GAN: 0.200 D_real: 0.401 D_fake: 0.149 G: 3.175 NCE: 3.317 NCE_Y: 2.633 \n",
            "End of epoch 1 / 400 \t Time Taken: 151 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 2, iters: 97, time: 0.431, data: 0.003) G_GAN: 0.336 D_real: 0.213 D_fake: 0.245 G: 2.846 NCE: 2.358 NCE_Y: 2.662 \n",
            "(epoch: 2, iters: 197, time: 0.450, data: 0.002) G_GAN: 0.348 D_real: 0.292 D_fake: 0.229 G: 2.761 NCE: 2.256 NCE_Y: 2.570 \n",
            "(epoch: 2, iters: 297, time: 0.461, data: 0.002) G_GAN: 0.638 D_real: 0.111 D_fake: 0.485 G: 2.860 NCE: 2.591 NCE_Y: 1.854 \n",
            "End of epoch 2 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 3, iters: 94, time: 0.468, data: 0.001) G_GAN: 0.329 D_real: 0.452 D_fake: 0.055 G: 2.698 NCE: 2.425 NCE_Y: 2.313 \n",
            "(epoch: 3, iters: 194, time: 0.473, data: 0.002) G_GAN: 0.312 D_real: 0.262 D_fake: 0.192 G: 2.333 NCE: 1.938 NCE_Y: 2.106 \n",
            "(epoch: 3, iters: 294, time: 0.475, data: 0.002) G_GAN: 0.256 D_real: 0.382 D_fake: 0.170 G: 2.697 NCE: 3.135 NCE_Y: 1.748 \n",
            "End of epoch 3 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 4, iters: 91, time: 0.477, data: 0.003) G_GAN: 0.385 D_real: 0.203 D_fake: 0.234 G: 3.078 NCE: 2.992 NCE_Y: 2.393 \n",
            "(epoch: 4, iters: 191, time: 0.477, data: 0.002) G_GAN: 0.340 D_real: 0.398 D_fake: 0.062 G: 3.070 NCE: 3.566 NCE_Y: 1.893 \n",
            "(epoch: 4, iters: 291, time: 0.478, data: 0.003) G_GAN: 0.314 D_real: 0.296 D_fake: 0.142 G: 2.544 NCE: 2.776 NCE_Y: 1.685 \n",
            "End of epoch 4 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 5, iters: 88, time: 0.479, data: 0.002) G_GAN: 0.306 D_real: 0.319 D_fake: 0.243 G: 2.484 NCE: 2.382 NCE_Y: 1.973 \n",
            "(epoch: 5, iters: 188, time: 0.479, data: 0.003) G_GAN: 0.365 D_real: 0.106 D_fake: 0.499 G: 2.038 NCE: 2.048 NCE_Y: 1.299 \n",
            "(epoch: 5, iters: 288, time: 0.479, data: 0.002) G_GAN: 0.370 D_real: 0.221 D_fake: 0.291 G: 2.489 NCE: 2.613 NCE_Y: 1.624 \n",
            "saving the model at the end of epoch 5, iters 1515\n",
            "End of epoch 5 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 6, iters: 85, time: 0.479, data: 0.002) G_GAN: 0.251 D_real: 0.347 D_fake: 0.186 G: 1.910 NCE: 1.642 NCE_Y: 1.675 \n",
            "(epoch: 6, iters: 185, time: 0.479, data: 0.002) G_GAN: 0.237 D_real: 0.271 D_fake: 0.250 G: 2.026 NCE: 2.054 NCE_Y: 1.524 \n",
            "(epoch: 6, iters: 285, time: 0.479, data: 0.003) G_GAN: 0.192 D_real: 0.379 D_fake: 0.297 G: 1.780 NCE: 1.791 NCE_Y: 1.384 \n",
            "End of epoch 6 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 7, iters: 82, time: 0.480, data: 0.002) G_GAN: 0.182 D_real: 0.478 D_fake: 0.059 G: 1.625 NCE: 1.380 NCE_Y: 1.506 \n",
            "(epoch: 7, iters: 182, time: 0.479, data: 0.003) G_GAN: 0.301 D_real: 0.143 D_fake: 0.252 G: 1.742 NCE: 1.710 NCE_Y: 1.173 \n",
            "(epoch: 7, iters: 282, time: 0.479, data: 0.002) G_GAN: 0.426 D_real: 0.111 D_fake: 0.413 G: 1.863 NCE: 1.467 NCE_Y: 1.406 \n",
            "End of epoch 7 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 8, iters: 79, time: 0.479, data: 0.002) G_GAN: 0.258 D_real: 0.238 D_fake: 0.216 G: 1.819 NCE: 1.723 NCE_Y: 1.399 \n",
            "(epoch: 8, iters: 179, time: 0.479, data: 0.002) G_GAN: 0.357 D_real: 0.094 D_fake: 0.541 G: 1.948 NCE: 1.825 NCE_Y: 1.357 \n",
            "(epoch: 8, iters: 279, time: 0.478, data: 0.003) G_GAN: 0.327 D_real: 0.346 D_fake: 0.187 G: 1.929 NCE: 1.668 NCE_Y: 1.537 \n",
            "End of epoch 8 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 9, iters: 76, time: 0.478, data: 0.001) G_GAN: 0.311 D_real: 0.185 D_fake: 0.291 G: 2.036 NCE: 2.211 NCE_Y: 1.239 \n",
            "(epoch: 9, iters: 176, time: 0.478, data: 0.002) G_GAN: 0.362 D_real: 0.368 D_fake: 0.109 G: 1.944 NCE: 1.524 NCE_Y: 1.641 \n",
            "(epoch: 9, iters: 276, time: 0.478, data: 0.002) G_GAN: 0.368 D_real: 0.269 D_fake: 0.265 G: 2.046 NCE: 2.188 NCE_Y: 1.169 \n",
            "End of epoch 9 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 10, iters: 73, time: 0.478, data: 0.002) G_GAN: 0.412 D_real: 0.298 D_fake: 0.134 G: 1.750 NCE: 1.610 NCE_Y: 1.067 \n",
            "(epoch: 10, iters: 173, time: 0.478, data: 0.002) G_GAN: 0.374 D_real: 0.262 D_fake: 0.107 G: 1.750 NCE: 1.443 NCE_Y: 1.310 \n",
            "(epoch: 10, iters: 273, time: 0.478, data: 0.002) G_GAN: 0.213 D_real: 0.374 D_fake: 0.075 G: 1.628 NCE: 1.560 NCE_Y: 1.268 \n",
            "saving the model at the end of epoch 10, iters 3030\n",
            "End of epoch 10 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 11, iters: 70, time: 0.479, data: 0.002) G_GAN: 0.529 D_real: 0.155 D_fake: 0.178 G: 2.087 NCE: 1.920 NCE_Y: 1.196 \n",
            "(epoch: 11, iters: 170, time: 0.479, data: 0.002) G_GAN: 0.227 D_real: 0.583 D_fake: 0.151 G: 1.919 NCE: 2.136 NCE_Y: 1.247 \n",
            "(epoch: 11, iters: 270, time: 0.479, data: 0.002) G_GAN: 0.392 D_real: 0.225 D_fake: 0.236 G: 1.807 NCE: 1.651 NCE_Y: 1.180 \n",
            "End of epoch 11 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 12, iters: 67, time: 0.479, data: 0.003) G_GAN: 0.295 D_real: 0.326 D_fake: 0.122 G: 1.724 NCE: 1.745 NCE_Y: 1.112 \n",
            "(epoch: 12, iters: 167, time: 0.479, data: 0.003) G_GAN: 0.383 D_real: 0.339 D_fake: 0.119 G: 1.532 NCE: 1.248 NCE_Y: 1.051 \n",
            "(epoch: 12, iters: 267, time: 0.480, data: 0.003) G_GAN: 0.226 D_real: 0.530 D_fake: 0.143 G: 2.054 NCE: 2.285 NCE_Y: 1.370 \n",
            "End of epoch 12 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 13, iters: 64, time: 0.480, data: 0.002) G_GAN: 0.265 D_real: 0.359 D_fake: 0.146 G: 1.731 NCE: 1.846 NCE_Y: 1.087 \n",
            "(epoch: 13, iters: 164, time: 0.480, data: 0.003) G_GAN: 0.257 D_real: 0.367 D_fake: 0.123 G: 1.678 NCE: 1.674 NCE_Y: 1.169 \n",
            "(epoch: 13, iters: 264, time: 0.480, data: 0.003) G_GAN: 0.275 D_real: 0.208 D_fake: 0.261 G: 1.617 NCE: 1.578 NCE_Y: 1.106 \n",
            "End of epoch 13 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 14, iters: 61, time: 0.480, data: 0.003) G_GAN: 0.409 D_real: 0.197 D_fake: 0.222 G: 1.737 NCE: 1.511 NCE_Y: 1.145 \n",
            "(epoch: 14, iters: 161, time: 0.480, data: 0.002) G_GAN: 0.308 D_real: 0.282 D_fake: 0.134 G: 1.612 NCE: 1.449 NCE_Y: 1.159 \n",
            "(epoch: 14, iters: 261, time: 0.480, data: 0.002) G_GAN: 0.368 D_real: 0.243 D_fake: 0.145 G: 1.530 NCE: 1.281 NCE_Y: 1.042 \n",
            "End of epoch 14 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 15, iters: 58, time: 0.480, data: 0.003) G_GAN: 0.318 D_real: 0.143 D_fake: 0.323 G: 1.488 NCE: 1.139 NCE_Y: 1.201 \n",
            "(epoch: 15, iters: 158, time: 0.480, data: 0.002) G_GAN: 0.354 D_real: 0.201 D_fake: 0.238 G: 1.495 NCE: 1.332 NCE_Y: 0.948 \n",
            "(epoch: 15, iters: 258, time: 0.480, data: 0.002) G_GAN: 0.305 D_real: 0.175 D_fake: 0.244 G: 1.642 NCE: 1.600 NCE_Y: 1.075 \n",
            "saving the model at the end of epoch 15, iters 4545\n",
            "End of epoch 15 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 16, iters: 55, time: 0.480, data: 0.003) G_GAN: 0.256 D_real: 0.412 D_fake: 0.116 G: 1.377 NCE: 1.212 NCE_Y: 1.030 \n",
            "(epoch: 16, iters: 155, time: 0.480, data: 0.002) G_GAN: 0.340 D_real: 0.147 D_fake: 0.248 G: 1.891 NCE: 1.967 NCE_Y: 1.135 \n",
            "(epoch: 16, iters: 255, time: 0.480, data: 0.003) G_GAN: 0.297 D_real: 0.254 D_fake: 0.298 G: 1.803 NCE: 1.808 NCE_Y: 1.203 \n",
            "End of epoch 16 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 17, iters: 52, time: 0.480, data: 0.002) G_GAN: 0.189 D_real: 0.273 D_fake: 0.205 G: 1.506 NCE: 1.537 NCE_Y: 1.096 \n",
            "(epoch: 17, iters: 152, time: 0.480, data: 0.003) G_GAN: 0.326 D_real: 0.270 D_fake: 0.185 G: 1.501 NCE: 1.216 NCE_Y: 1.134 \n",
            "saving the latest model (epoch 17, total_iters 5000)\n",
            "desjpn2seguy\n",
            "(epoch: 17, iters: 252, time: 0.480, data: 0.002) G_GAN: 0.369 D_real: 0.271 D_fake: 0.238 G: 1.764 NCE: 1.602 NCE_Y: 1.188 \n",
            "End of epoch 17 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 18, iters: 49, time: 0.481, data: 0.003) G_GAN: 0.379 D_real: 0.186 D_fake: 0.381 G: 1.796 NCE: 1.724 NCE_Y: 1.112 \n",
            "(epoch: 18, iters: 149, time: 0.481, data: 0.002) G_GAN: 0.283 D_real: 0.317 D_fake: 0.167 G: 1.375 NCE: 1.143 NCE_Y: 1.039 \n",
            "(epoch: 18, iters: 249, time: 0.481, data: 0.003) G_GAN: 0.436 D_real: 0.173 D_fake: 0.248 G: 1.672 NCE: 1.303 NCE_Y: 1.169 \n",
            "End of epoch 18 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 19, iters: 46, time: 0.481, data: 0.003) G_GAN: 0.217 D_real: 0.134 D_fake: 0.448 G: 1.636 NCE: 1.826 NCE_Y: 1.013 \n",
            "(epoch: 19, iters: 146, time: 0.480, data: 0.002) G_GAN: 0.238 D_real: 0.271 D_fake: 0.189 G: 1.593 NCE: 1.752 NCE_Y: 0.958 \n",
            "(epoch: 19, iters: 246, time: 0.480, data: 0.003) G_GAN: 0.191 D_real: 0.373 D_fake: 0.171 G: 1.598 NCE: 1.600 NCE_Y: 1.215 \n",
            "End of epoch 19 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 20, iters: 43, time: 0.481, data: 0.002) G_GAN: 0.350 D_real: 0.328 D_fake: 0.243 G: 1.750 NCE: 1.840 NCE_Y: 0.959 \n",
            "(epoch: 20, iters: 143, time: 0.480, data: 0.003) G_GAN: 0.258 D_real: 0.195 D_fake: 0.384 G: 1.628 NCE: 1.643 NCE_Y: 1.098 \n",
            "(epoch: 20, iters: 243, time: 0.480, data: 0.003) G_GAN: 0.212 D_real: 0.294 D_fake: 0.192 G: 1.936 NCE: 2.394 NCE_Y: 1.054 \n",
            "saving the model at the end of epoch 20, iters 6060\n",
            "End of epoch 20 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 21, iters: 40, time: 0.481, data: 0.002) G_GAN: 0.323 D_real: 0.118 D_fake: 0.550 G: 1.536 NCE: 1.367 NCE_Y: 1.059 \n",
            "(epoch: 21, iters: 140, time: 0.480, data: 0.003) G_GAN: 0.330 D_real: 0.209 D_fake: 0.118 G: 1.671 NCE: 1.711 NCE_Y: 0.971 \n",
            "(epoch: 21, iters: 240, time: 0.480, data: 0.002) G_GAN: 0.276 D_real: 0.317 D_fake: 0.125 G: 1.396 NCE: 1.274 NCE_Y: 0.967 \n",
            "End of epoch 21 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 22, iters: 37, time: 0.481, data: 0.002) G_GAN: 0.435 D_real: 0.458 D_fake: 0.086 G: 1.562 NCE: 1.251 NCE_Y: 1.003 \n",
            "(epoch: 22, iters: 137, time: 0.481, data: 0.002) G_GAN: 0.435 D_real: 0.188 D_fake: 0.237 G: 1.993 NCE: 1.991 NCE_Y: 1.125 \n",
            "(epoch: 22, iters: 237, time: 0.480, data: 0.002) G_GAN: 0.225 D_real: 0.291 D_fake: 0.188 G: 1.704 NCE: 1.250 NCE_Y: 1.707 \n",
            "End of epoch 22 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 23, iters: 34, time: 0.480, data: 0.002) G_GAN: 0.231 D_real: 0.154 D_fake: 0.365 G: 1.605 NCE: 1.836 NCE_Y: 0.913 \n",
            "(epoch: 23, iters: 134, time: 0.480, data: 0.003) G_GAN: 0.179 D_real: 0.580 D_fake: 0.038 G: 1.844 NCE: 1.694 NCE_Y: 1.637 \n",
            "(epoch: 23, iters: 234, time: 0.480, data: 0.002) G_GAN: 0.374 D_real: 0.282 D_fake: 0.230 G: 1.614 NCE: 1.482 NCE_Y: 0.999 \n",
            "End of epoch 23 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 24, iters: 31, time: 0.481, data: 0.003) G_GAN: 0.256 D_real: 0.290 D_fake: 0.238 G: 1.556 NCE: 1.580 NCE_Y: 1.020 \n",
            "(epoch: 24, iters: 131, time: 0.481, data: 0.003) G_GAN: 0.278 D_real: 0.175 D_fake: 0.495 G: 1.550 NCE: 1.536 NCE_Y: 1.008 \n",
            "(epoch: 24, iters: 231, time: 0.481, data: 0.003) G_GAN: 0.739 D_real: 0.026 D_fake: 0.422 G: 2.607 NCE: 1.895 NCE_Y: 1.842 \n",
            "End of epoch 24 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 25, iters: 28, time: 0.481, data: 0.002) G_GAN: 0.353 D_real: 0.401 D_fake: 0.158 G: 1.737 NCE: 1.669 NCE_Y: 1.099 \n",
            "(epoch: 25, iters: 128, time: 0.481, data: 0.003) G_GAN: 0.296 D_real: 0.282 D_fake: 0.088 G: 1.612 NCE: 1.522 NCE_Y: 1.110 \n",
            "(epoch: 25, iters: 228, time: 0.481, data: 0.003) G_GAN: 0.375 D_real: 0.149 D_fake: 0.369 G: 1.558 NCE: 1.218 NCE_Y: 1.147 \n",
            "saving the model at the end of epoch 25, iters 7575\n",
            "End of epoch 25 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 26, iters: 25, time: 0.481, data: 0.003) G_GAN: 0.247 D_real: 0.271 D_fake: 0.214 G: 1.511 NCE: 1.374 NCE_Y: 1.154 \n",
            "(epoch: 26, iters: 125, time: 0.481, data: 0.002) G_GAN: 0.346 D_real: 0.105 D_fake: 0.395 G: 1.670 NCE: 1.500 NCE_Y: 1.148 \n",
            "(epoch: 26, iters: 225, time: 0.481, data: 0.003) G_GAN: 0.268 D_real: 0.227 D_fake: 0.188 G: 1.580 NCE: 1.515 NCE_Y: 1.109 \n",
            "End of epoch 26 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 27, iters: 22, time: 0.481, data: 0.003) G_GAN: 0.599 D_real: 0.127 D_fake: 0.342 G: 1.887 NCE: 1.517 NCE_Y: 1.058 \n",
            "(epoch: 27, iters: 122, time: 0.481, data: 0.003) G_GAN: 0.691 D_real: 0.030 D_fake: 0.538 G: 2.266 NCE: 2.093 NCE_Y: 1.056 \n",
            "(epoch: 27, iters: 222, time: 0.481, data: 0.002) G_GAN: 0.394 D_real: 0.397 D_fake: 0.153 G: 1.599 NCE: 1.330 NCE_Y: 1.079 \n",
            "End of epoch 27 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 28, iters: 19, time: 0.481, data: 0.003) G_GAN: 0.259 D_real: 0.473 D_fake: 0.048 G: 1.469 NCE: 1.375 NCE_Y: 1.046 \n",
            "(epoch: 28, iters: 119, time: 0.481, data: 0.002) G_GAN: 0.271 D_real: 0.364 D_fake: 0.176 G: 1.600 NCE: 1.694 NCE_Y: 0.963 \n",
            "(epoch: 28, iters: 219, time: 0.481, data: 0.002) G_GAN: 0.330 D_real: 0.317 D_fake: 0.123 G: 1.577 NCE: 1.247 NCE_Y: 1.247 \n",
            "End of epoch 28 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 29, iters: 16, time: 0.481, data: 0.002) G_GAN: 0.794 D_real: 0.078 D_fake: 0.280 G: 2.206 NCE: 1.677 NCE_Y: 1.146 \n",
            "(epoch: 29, iters: 116, time: 0.481, data: 0.002) G_GAN: 0.469 D_real: 0.057 D_fake: 0.355 G: 1.909 NCE: 1.578 NCE_Y: 1.302 \n",
            "(epoch: 29, iters: 216, time: 0.481, data: 0.003) G_GAN: 0.319 D_real: 0.416 D_fake: 0.195 G: 1.489 NCE: 1.366 NCE_Y: 0.973 \n",
            "End of epoch 29 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 30, iters: 13, time: 0.481, data: 0.003) G_GAN: 0.196 D_real: 0.415 D_fake: 0.129 G: 1.328 NCE: 1.255 NCE_Y: 1.010 \n",
            "(epoch: 30, iters: 113, time: 0.481, data: 0.002) G_GAN: 0.260 D_real: 0.276 D_fake: 0.234 G: 1.679 NCE: 1.781 NCE_Y: 1.057 \n",
            "(epoch: 30, iters: 213, time: 0.480, data: 0.003) G_GAN: 0.550 D_real: 0.122 D_fake: 0.235 G: 1.855 NCE: 1.422 NCE_Y: 1.188 \n",
            "saving the model at the end of epoch 30, iters 9090\n",
            "End of epoch 30 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 31, iters: 10, time: 0.480, data: 0.003) G_GAN: 0.338 D_real: 0.442 D_fake: 0.186 G: 1.628 NCE: 1.598 NCE_Y: 0.981 \n",
            "(epoch: 31, iters: 110, time: 0.481, data: 0.002) G_GAN: 0.447 D_real: 0.301 D_fake: 0.100 G: 1.576 NCE: 1.343 NCE_Y: 0.914 \n",
            "(epoch: 31, iters: 210, time: 0.480, data: 0.001) G_GAN: 0.345 D_real: 0.274 D_fake: 0.288 G: 1.626 NCE: 1.506 NCE_Y: 1.055 \n",
            "End of epoch 31 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 32, iters: 7, time: 0.480, data: 0.002) G_GAN: 0.256 D_real: 0.248 D_fake: 0.421 G: 1.391 NCE: 1.275 NCE_Y: 0.994 \n",
            "(epoch: 32, iters: 107, time: 0.480, data: 0.001) G_GAN: 0.437 D_real: 0.218 D_fake: 0.195 G: 1.656 NCE: 1.437 NCE_Y: 1.000 \n",
            "(epoch: 32, iters: 207, time: 0.480, data: 0.002) G_GAN: 0.226 D_real: 0.226 D_fake: 0.284 G: 1.517 NCE: 1.399 NCE_Y: 1.182 \n",
            "End of epoch 32 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 33, iters: 4, time: 0.480, data: 0.002) G_GAN: 0.553 D_real: 0.144 D_fake: 0.213 G: 2.136 NCE: 1.534 NCE_Y: 1.632 \n",
            "(epoch: 33, iters: 104, time: 0.480, data: 0.003) G_GAN: 0.526 D_real: 0.118 D_fake: 0.268 G: 2.519 NCE: 1.457 NCE_Y: 2.530 \n",
            "(epoch: 33, iters: 204, time: 0.480, data: 0.003) G_GAN: 0.274 D_real: 0.325 D_fake: 0.330 G: 1.571 NCE: 1.392 NCE_Y: 1.202 \n",
            "End of epoch 33 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 34, iters: 1, time: 0.480, data: 0.003) G_GAN: 0.202 D_real: 0.576 D_fake: 0.054 G: 1.589 NCE: 1.746 NCE_Y: 1.027 \n",
            "saving the latest model (epoch 34, total_iters 10000)\n",
            "desjpn2seguy\n",
            "(epoch: 34, iters: 101, time: 0.480, data: 0.005) G_GAN: 0.228 D_real: 0.399 D_fake: 0.218 G: 1.622 NCE: 1.809 NCE_Y: 0.979 \n",
            "(epoch: 34, iters: 201, time: 0.480, data: 0.003) G_GAN: 0.448 D_real: 0.131 D_fake: 0.341 G: 1.684 NCE: 1.346 NCE_Y: 1.126 \n",
            "(epoch: 34, iters: 301, time: 0.480, data: 0.003) G_GAN: 0.232 D_real: 0.455 D_fake: 0.138 G: 1.396 NCE: 1.393 NCE_Y: 0.936 \n",
            "End of epoch 34 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 35, iters: 98, time: 0.480, data: 0.003) G_GAN: 0.448 D_real: 0.173 D_fake: 0.099 G: 1.563 NCE: 1.131 NCE_Y: 1.098 \n",
            "(epoch: 35, iters: 198, time: 0.480, data: 0.002) G_GAN: 0.414 D_real: 0.051 D_fake: 0.334 G: 1.545 NCE: 1.352 NCE_Y: 0.910 \n",
            "(epoch: 35, iters: 298, time: 0.480, data: 0.003) G_GAN: 0.317 D_real: 0.296 D_fake: 0.164 G: 1.670 NCE: 1.655 NCE_Y: 1.052 \n",
            "saving the model at the end of epoch 35, iters 10605\n",
            "End of epoch 35 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 36, iters: 95, time: 0.480, data: 0.003) G_GAN: 0.321 D_real: 0.228 D_fake: 0.137 G: 1.586 NCE: 1.549 NCE_Y: 0.982 \n",
            "(epoch: 36, iters: 195, time: 0.480, data: 0.003) G_GAN: 0.200 D_real: 0.227 D_fake: 0.323 G: 1.561 NCE: 1.773 NCE_Y: 0.950 \n",
            "(epoch: 36, iters: 295, time: 0.480, data: 0.002) G_GAN: 0.457 D_real: 0.058 D_fake: 0.364 G: 1.765 NCE: 1.216 NCE_Y: 1.400 \n",
            "End of epoch 36 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 37, iters: 92, time: 0.480, data: 0.002) G_GAN: 0.235 D_real: 0.397 D_fake: 0.217 G: 1.709 NCE: 1.883 NCE_Y: 1.066 \n",
            "(epoch: 37, iters: 192, time: 0.480, data: 0.002) G_GAN: 0.412 D_real: 0.272 D_fake: 0.257 G: 1.605 NCE: 1.485 NCE_Y: 0.902 \n",
            "(epoch: 37, iters: 292, time: 0.480, data: 0.002) G_GAN: 0.331 D_real: 0.224 D_fake: 0.205 G: 1.481 NCE: 1.226 NCE_Y: 1.074 \n",
            "End of epoch 37 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 38, iters: 89, time: 0.480, data: 0.002) G_GAN: 0.271 D_real: 0.312 D_fake: 0.202 G: 1.468 NCE: 1.425 NCE_Y: 0.968 \n",
            "(epoch: 38, iters: 189, time: 0.480, data: 0.003) G_GAN: 0.373 D_real: 0.255 D_fake: 0.078 G: 2.155 NCE: 1.777 NCE_Y: 1.787 \n",
            "(epoch: 38, iters: 289, time: 0.480, data: 0.002) G_GAN: 0.184 D_real: 0.541 D_fake: 0.044 G: 1.341 NCE: 1.301 NCE_Y: 1.013 \n",
            "End of epoch 38 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 39, iters: 86, time: 0.480, data: 0.003) G_GAN: 0.671 D_real: 0.217 D_fake: 0.318 G: 2.114 NCE: 1.860 NCE_Y: 1.025 \n",
            "(epoch: 39, iters: 186, time: 0.480, data: 0.002) G_GAN: 0.335 D_real: 0.151 D_fake: 0.379 G: 1.384 NCE: 1.177 NCE_Y: 0.922 \n",
            "(epoch: 39, iters: 286, time: 0.480, data: 0.002) G_GAN: 0.396 D_real: 0.113 D_fake: 0.254 G: 1.480 NCE: 1.290 NCE_Y: 0.877 \n",
            "End of epoch 39 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 40, iters: 83, time: 0.480, data: 0.003) G_GAN: 0.216 D_real: 0.295 D_fake: 0.144 G: 1.337 NCE: 1.261 NCE_Y: 0.982 \n",
            "(epoch: 40, iters: 183, time: 0.480, data: 0.003) G_GAN: 0.481 D_real: 0.196 D_fake: 0.258 G: 1.852 NCE: 1.593 NCE_Y: 1.148 \n",
            "(epoch: 40, iters: 283, time: 0.480, data: 0.002) G_GAN: 0.460 D_real: 0.269 D_fake: 0.422 G: 2.321 NCE: 1.737 NCE_Y: 1.986 \n",
            "saving the model at the end of epoch 40, iters 12120\n",
            "End of epoch 40 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 41, iters: 80, time: 0.480, data: 0.003) G_GAN: 0.276 D_real: 0.528 D_fake: 0.102 G: 1.548 NCE: 1.463 NCE_Y: 1.081 \n",
            "(epoch: 41, iters: 180, time: 0.480, data: 0.002) G_GAN: 0.382 D_real: 0.254 D_fake: 0.249 G: 1.729 NCE: 1.672 NCE_Y: 1.021 \n",
            "(epoch: 41, iters: 280, time: 0.480, data: 0.003) G_GAN: 0.214 D_real: 0.444 D_fake: 0.142 G: 1.447 NCE: 1.563 NCE_Y: 0.904 \n",
            "End of epoch 41 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 42, iters: 77, time: 0.480, data: 0.002) G_GAN: 0.461 D_real: 0.180 D_fake: 0.193 G: 1.571 NCE: 1.254 NCE_Y: 0.965 \n",
            "(epoch: 42, iters: 177, time: 0.480, data: 0.003) G_GAN: 0.436 D_real: 0.368 D_fake: 0.012 G: 1.612 NCE: 1.434 NCE_Y: 0.919 \n",
            "(epoch: 42, iters: 277, time: 0.480, data: 0.002) G_GAN: 0.416 D_real: 0.185 D_fake: 0.529 G: 1.702 NCE: 1.621 NCE_Y: 0.952 \n",
            "End of epoch 42 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 43, iters: 74, time: 0.480, data: 0.003) G_GAN: 0.284 D_real: 0.253 D_fake: 0.206 G: 1.438 NCE: 1.400 NCE_Y: 0.906 \n",
            "(epoch: 43, iters: 174, time: 0.480, data: 0.002) G_GAN: 0.435 D_real: 0.304 D_fake: 0.151 G: 1.623 NCE: 1.440 NCE_Y: 0.936 \n",
            "(epoch: 43, iters: 274, time: 0.480, data: 0.002) G_GAN: 0.158 D_real: 0.576 D_fake: 0.064 G: 1.262 NCE: 1.244 NCE_Y: 0.962 \n",
            "End of epoch 43 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 44, iters: 71, time: 0.481, data: 0.002) G_GAN: 0.456 D_real: 0.193 D_fake: 0.109 G: 1.861 NCE: 1.383 NCE_Y: 1.428 \n",
            "(epoch: 44, iters: 171, time: 0.481, data: 0.003) G_GAN: 0.237 D_real: 0.246 D_fake: 0.523 G: 1.307 NCE: 1.291 NCE_Y: 0.850 \n",
            "(epoch: 44, iters: 271, time: 0.480, data: 0.002) G_GAN: 0.253 D_real: 0.315 D_fake: 0.125 G: 1.429 NCE: 1.404 NCE_Y: 0.949 \n",
            "End of epoch 44 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 45, iters: 68, time: 0.481, data: 0.002) G_GAN: 0.183 D_real: 0.154 D_fake: 0.676 G: 1.641 NCE: 1.935 NCE_Y: 0.980 \n",
            "(epoch: 45, iters: 168, time: 0.481, data: 0.002) G_GAN: 0.273 D_real: 0.316 D_fake: 0.127 G: 1.359 NCE: 1.250 NCE_Y: 0.922 \n",
            "(epoch: 45, iters: 268, time: 0.481, data: 0.003) G_GAN: 0.184 D_real: 0.549 D_fake: 0.102 G: 1.436 NCE: 1.398 NCE_Y: 1.106 \n",
            "saving the model at the end of epoch 45, iters 13635\n",
            "End of epoch 45 / 400 \t Time Taken: 149 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 46, iters: 65, time: 0.481, data: 0.002) G_GAN: 0.340 D_real: 0.289 D_fake: 0.192 G: 1.528 NCE: 1.393 NCE_Y: 0.981 \n",
            "(epoch: 46, iters: 165, time: 0.481, data: 0.003) G_GAN: 0.216 D_real: 0.386 D_fake: 0.090 G: 1.468 NCE: 1.520 NCE_Y: 0.984 \n",
            "(epoch: 46, iters: 265, time: 0.480, data: 0.003) G_GAN: 0.409 D_real: 0.134 D_fake: 0.254 G: 1.532 NCE: 1.366 NCE_Y: 0.881 \n",
            "End of epoch 46 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 47, iters: 62, time: 0.481, data: 0.003) G_GAN: 0.408 D_real: 0.167 D_fake: 0.419 G: 1.499 NCE: 1.234 NCE_Y: 0.947 \n",
            "(epoch: 47, iters: 162, time: 0.481, data: 0.003) G_GAN: 0.419 D_real: 0.389 D_fake: 0.102 G: 1.552 NCE: 1.237 NCE_Y: 1.030 \n",
            "(epoch: 47, iters: 262, time: 0.481, data: 0.003) G_GAN: 0.496 D_real: 0.278 D_fake: 0.151 G: 1.714 NCE: 1.309 NCE_Y: 1.127 \n",
            "End of epoch 47 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 48, iters: 59, time: 0.481, data: 0.003) G_GAN: 0.443 D_real: 0.192 D_fake: 0.308 G: 1.606 NCE: 1.374 NCE_Y: 0.953 \n",
            "(epoch: 48, iters: 159, time: 0.481, data: 0.004) G_GAN: 0.250 D_real: 0.207 D_fake: 0.290 G: 1.414 NCE: 1.387 NCE_Y: 0.941 \n",
            "(epoch: 48, iters: 259, time: 0.481, data: 0.002) G_GAN: 0.541 D_real: 0.231 D_fake: 0.085 G: 1.823 NCE: 1.448 NCE_Y: 1.114 \n",
            "End of epoch 48 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 49, iters: 56, time: 0.481, data: 0.003) G_GAN: 0.373 D_real: 0.170 D_fake: 0.314 G: 1.519 NCE: 1.233 NCE_Y: 1.060 \n",
            "(epoch: 49, iters: 156, time: 0.481, data: 0.003) G_GAN: 0.340 D_real: 0.114 D_fake: 0.326 G: 1.711 NCE: 1.707 NCE_Y: 1.036 \n",
            "(epoch: 49, iters: 256, time: 0.481, data: 0.002) G_GAN: 0.604 D_real: 0.049 D_fake: 0.617 G: 1.892 NCE: 1.674 NCE_Y: 0.903 \n",
            "End of epoch 49 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 50, iters: 53, time: 0.480, data: 0.002) G_GAN: 0.801 D_real: 0.061 D_fake: 0.099 G: 2.229 NCE: 1.609 NCE_Y: 1.248 \n",
            "(epoch: 50, iters: 153, time: 0.480, data: 0.003) G_GAN: 0.373 D_real: 0.186 D_fake: 0.397 G: 1.497 NCE: 1.167 NCE_Y: 1.081 \n",
            "saving the latest model (epoch 50, total_iters 15000)\n",
            "desjpn2seguy\n",
            "(epoch: 50, iters: 253, time: 0.480, data: 0.002) G_GAN: 0.278 D_real: 0.430 D_fake: 0.146 G: 1.476 NCE: 1.458 NCE_Y: 0.939 \n",
            "saving the model at the end of epoch 50, iters 15150\n",
            "End of epoch 50 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 51, iters: 50, time: 0.481, data: 0.003) G_GAN: 0.648 D_real: 0.095 D_fake: 0.251 G: 1.718 NCE: 1.211 NCE_Y: 0.928 \n",
            "(epoch: 51, iters: 150, time: 0.481, data: 0.003) G_GAN: 0.174 D_real: 0.381 D_fake: 0.320 G: 1.429 NCE: 1.606 NCE_Y: 0.905 \n",
            "(epoch: 51, iters: 250, time: 0.481, data: 0.002) G_GAN: 0.188 D_real: 0.627 D_fake: 0.093 G: 1.289 NCE: 1.257 NCE_Y: 0.945 \n",
            "End of epoch 51 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 52, iters: 47, time: 0.481, data: 0.003) G_GAN: 0.246 D_real: 0.271 D_fake: 0.716 G: 1.430 NCE: 1.379 NCE_Y: 0.989 \n",
            "(epoch: 52, iters: 147, time: 0.480, data: 0.002) G_GAN: 0.358 D_real: 0.075 D_fake: 0.393 G: 1.647 NCE: 1.621 NCE_Y: 0.957 \n",
            "(epoch: 52, iters: 247, time: 0.480, data: 0.002) G_GAN: 0.390 D_real: 0.341 D_fake: 0.069 G: 1.529 NCE: 1.372 NCE_Y: 0.907 \n",
            "End of epoch 52 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 53, iters: 44, time: 0.480, data: 0.003) G_GAN: 0.456 D_real: 0.156 D_fake: 0.194 G: 1.569 NCE: 1.345 NCE_Y: 0.881 \n",
            "(epoch: 53, iters: 144, time: 0.480, data: 0.003) G_GAN: 0.484 D_real: 0.173 D_fake: 0.232 G: 1.713 NCE: 1.593 NCE_Y: 0.866 \n",
            "(epoch: 53, iters: 244, time: 0.480, data: 0.004) G_GAN: 0.360 D_real: 0.244 D_fake: 0.278 G: 1.511 NCE: 1.349 NCE_Y: 0.953 \n",
            "End of epoch 53 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 54, iters: 41, time: 0.480, data: 0.003) G_GAN: 0.471 D_real: 0.097 D_fake: 0.371 G: 1.713 NCE: 1.537 NCE_Y: 0.947 \n",
            "(epoch: 54, iters: 141, time: 0.480, data: 0.003) G_GAN: 0.538 D_real: 0.172 D_fake: 0.423 G: 1.623 NCE: 1.318 NCE_Y: 0.852 \n",
            "(epoch: 54, iters: 241, time: 0.480, data: 0.002) G_GAN: 0.398 D_real: 0.232 D_fake: 0.114 G: 1.452 NCE: 1.226 NCE_Y: 0.882 \n",
            "End of epoch 54 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 55, iters: 38, time: 0.480, data: 0.003) G_GAN: 0.388 D_real: 0.147 D_fake: 0.377 G: 1.550 NCE: 1.470 NCE_Y: 0.855 \n",
            "(epoch: 55, iters: 138, time: 0.481, data: 0.002) G_GAN: 0.282 D_real: 0.330 D_fake: 0.106 G: 1.408 NCE: 1.307 NCE_Y: 0.946 \n",
            "(epoch: 55, iters: 238, time: 0.481, data: 0.002) G_GAN: 0.392 D_real: 0.280 D_fake: 0.190 G: 1.542 NCE: 1.404 NCE_Y: 0.897 \n",
            "saving the model at the end of epoch 55, iters 16665\n",
            "End of epoch 55 / 400 \t Time Taken: 149 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 56, iters: 35, time: 0.481, data: 0.002) G_GAN: 0.768 D_real: 0.164 D_fake: 0.120 G: 1.950 NCE: 1.213 NCE_Y: 1.150 \n",
            "(epoch: 56, iters: 135, time: 0.481, data: 0.002) G_GAN: 0.818 D_real: 0.023 D_fake: 0.573 G: 1.922 NCE: 1.243 NCE_Y: 0.965 \n",
            "(epoch: 56, iters: 235, time: 0.481, data: 0.002) G_GAN: 0.357 D_real: 0.079 D_fake: 0.350 G: 1.436 NCE: 1.258 NCE_Y: 0.900 \n",
            "End of epoch 56 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 57, iters: 32, time: 0.481, data: 0.003) G_GAN: 0.462 D_real: 0.138 D_fake: 0.320 G: 1.872 NCE: 1.591 NCE_Y: 1.229 \n",
            "(epoch: 57, iters: 132, time: 0.481, data: 0.001) G_GAN: 0.412 D_real: 0.380 D_fake: 0.177 G: 1.607 NCE: 1.470 NCE_Y: 0.920 \n",
            "(epoch: 57, iters: 232, time: 0.481, data: 0.004) G_GAN: 0.424 D_real: 0.226 D_fake: 0.198 G: 1.615 NCE: 1.281 NCE_Y: 1.101 \n",
            "End of epoch 57 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 58, iters: 29, time: 0.481, data: 0.002) G_GAN: 0.315 D_real: 0.118 D_fake: 0.290 G: 1.726 NCE: 1.670 NCE_Y: 1.151 \n",
            "(epoch: 58, iters: 129, time: 0.481, data: 0.003) G_GAN: 0.794 D_real: 0.387 D_fake: 0.033 G: 2.140 NCE: 1.650 NCE_Y: 1.043 \n",
            "(epoch: 58, iters: 229, time: 0.481, data: 0.003) G_GAN: 0.366 D_real: 0.217 D_fake: 0.256 G: 1.600 NCE: 1.601 NCE_Y: 0.867 \n",
            "End of epoch 58 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 59, iters: 26, time: 0.481, data: 0.003) G_GAN: 0.366 D_real: 0.349 D_fake: 0.127 G: 1.554 NCE: 1.473 NCE_Y: 0.904 \n",
            "(epoch: 59, iters: 126, time: 0.481, data: 0.002) G_GAN: 0.329 D_real: 0.440 D_fake: 0.272 G: 1.541 NCE: 1.363 NCE_Y: 1.062 \n",
            "(epoch: 59, iters: 226, time: 0.481, data: 0.003) G_GAN: 0.482 D_real: 0.175 D_fake: 0.180 G: 1.580 NCE: 1.242 NCE_Y: 0.956 \n",
            "End of epoch 59 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 60, iters: 23, time: 0.481, data: 0.004) G_GAN: 0.242 D_real: 0.214 D_fake: 0.261 G: 1.321 NCE: 1.251 NCE_Y: 0.906 \n",
            "(epoch: 60, iters: 123, time: 0.481, data: 0.003) G_GAN: 0.391 D_real: 0.193 D_fake: 0.227 G: 1.553 NCE: 1.335 NCE_Y: 0.988 \n",
            "(epoch: 60, iters: 223, time: 0.481, data: 0.003) G_GAN: 0.274 D_real: 0.349 D_fake: 0.277 G: 1.550 NCE: 1.474 NCE_Y: 1.077 \n",
            "saving the model at the end of epoch 60, iters 18180\n",
            "End of epoch 60 / 400 \t Time Taken: 149 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 61, iters: 20, time: 0.481, data: 0.003) G_GAN: 0.413 D_real: 0.143 D_fake: 0.273 G: 1.500 NCE: 1.299 NCE_Y: 0.874 \n",
            "(epoch: 61, iters: 120, time: 0.481, data: 0.002) G_GAN: 0.654 D_real: 0.190 D_fake: 0.029 G: 1.851 NCE: 1.450 NCE_Y: 0.943 \n",
            "(epoch: 61, iters: 220, time: 0.480, data: 0.003) G_GAN: 0.268 D_real: 0.334 D_fake: 0.105 G: 1.362 NCE: 1.198 NCE_Y: 0.992 \n",
            "End of epoch 61 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 62, iters: 17, time: 0.481, data: 0.002) G_GAN: 0.427 D_real: 0.190 D_fake: 0.233 G: 1.483 NCE: 1.159 NCE_Y: 0.954 \n",
            "(epoch: 62, iters: 117, time: 0.481, data: 0.003) G_GAN: 0.574 D_real: 0.108 D_fake: 0.192 G: 1.721 NCE: 1.242 NCE_Y: 1.051 \n",
            "(epoch: 62, iters: 217, time: 0.481, data: 0.003) G_GAN: 0.367 D_real: 0.411 D_fake: 0.067 G: 1.571 NCE: 1.272 NCE_Y: 1.136 \n",
            "End of epoch 62 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 63, iters: 14, time: 0.481, data: 0.003) G_GAN: 0.370 D_real: 0.233 D_fake: 0.203 G: 1.462 NCE: 1.283 NCE_Y: 0.903 \n",
            "(epoch: 63, iters: 114, time: 0.481, data: 0.002) G_GAN: 0.441 D_real: 0.261 D_fake: 0.280 G: 1.516 NCE: 1.165 NCE_Y: 0.986 \n",
            "(epoch: 63, iters: 214, time: 0.481, data: 0.003) G_GAN: 0.420 D_real: 0.492 D_fake: 0.077 G: 1.478 NCE: 1.254 NCE_Y: 0.863 \n",
            "End of epoch 63 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 64, iters: 11, time: 0.482, data: 0.003) G_GAN: 0.597 D_real: 0.064 D_fake: 0.277 G: 1.741 NCE: 1.379 NCE_Y: 0.909 \n",
            "(epoch: 64, iters: 111, time: 0.482, data: 0.003) G_GAN: 0.358 D_real: 0.307 D_fake: 0.389 G: 1.536 NCE: 1.231 NCE_Y: 1.125 \n",
            "(epoch: 64, iters: 211, time: 0.482, data: 0.002) G_GAN: 0.284 D_real: 0.249 D_fake: 0.263 G: 1.386 NCE: 1.285 NCE_Y: 0.918 \n",
            "End of epoch 64 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 65, iters: 8, time: 0.482, data: 0.003) G_GAN: 0.915 D_real: 0.387 D_fake: 0.459 G: 2.299 NCE: 1.317 NCE_Y: 1.451 \n",
            "(epoch: 65, iters: 108, time: 0.482, data: 0.001) G_GAN: 0.231 D_real: 0.149 D_fake: 0.378 G: 1.533 NCE: 1.292 NCE_Y: 1.312 \n",
            "(epoch: 65, iters: 208, time: 0.481, data: 0.003) G_GAN: 0.402 D_real: 0.166 D_fake: 0.346 G: 1.780 NCE: 1.559 NCE_Y: 1.198 \n",
            "saving the model at the end of epoch 65, iters 19695\n",
            "End of epoch 65 / 400 \t Time Taken: 149 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 66, iters: 5, time: 0.481, data: 0.002) G_GAN: 0.404 D_real: 0.334 D_fake: 0.260 G: 1.488 NCE: 1.239 NCE_Y: 0.931 \n",
            "(epoch: 66, iters: 105, time: 0.481, data: 0.003) G_GAN: 0.551 D_real: 0.161 D_fake: 0.136 G: 1.755 NCE: 1.278 NCE_Y: 1.130 \n",
            "(epoch: 66, iters: 205, time: 0.481, data: 0.003) G_GAN: 0.461 D_real: 0.204 D_fake: 0.127 G: 1.606 NCE: 1.399 NCE_Y: 0.891 \n",
            "End of epoch 66 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 67, iters: 2, time: 0.481, data: 0.003) G_GAN: 0.602 D_real: 0.236 D_fake: 0.106 G: 1.706 NCE: 1.224 NCE_Y: 0.983 \n",
            "saving the latest model (epoch 67, total_iters 20000)\n",
            "desjpn2seguy\n",
            "(epoch: 67, iters: 102, time: 0.481, data: 0.001) G_GAN: 0.353 D_real: 0.116 D_fake: 0.216 G: 1.572 NCE: 1.438 NCE_Y: 1.000 \n",
            "(epoch: 67, iters: 202, time: 0.481, data: 0.001) G_GAN: 0.428 D_real: 0.293 D_fake: 0.165 G: 1.533 NCE: 1.307 NCE_Y: 0.903 \n",
            "(epoch: 67, iters: 302, time: 0.481, data: 0.001) G_GAN: 0.470 D_real: 0.266 D_fake: 0.304 G: 1.863 NCE: 1.813 NCE_Y: 0.973 \n",
            "End of epoch 67 / 400 \t Time Taken: 149 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 68, iters: 99, time: 0.481, data: 0.003) G_GAN: 0.491 D_real: 0.237 D_fake: 0.444 G: 1.624 NCE: 1.340 NCE_Y: 0.926 \n",
            "(epoch: 68, iters: 199, time: 0.481, data: 0.002) G_GAN: 0.446 D_real: 0.564 D_fake: 0.069 G: 1.624 NCE: 1.489 NCE_Y: 0.866 \n",
            "(epoch: 68, iters: 299, time: 0.481, data: 0.003) G_GAN: 0.513 D_real: 0.109 D_fake: 0.295 G: 1.726 NCE: 1.500 NCE_Y: 0.925 \n",
            "End of epoch 68 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 69, iters: 96, time: 0.482, data: 0.003) G_GAN: 0.398 D_real: 0.224 D_fake: 0.131 G: 1.462 NCE: 1.247 NCE_Y: 0.880 \n",
            "(epoch: 69, iters: 196, time: 0.481, data: 0.003) G_GAN: 0.315 D_real: 0.999 D_fake: 0.055 G: 1.379 NCE: 1.217 NCE_Y: 0.912 \n",
            "(epoch: 69, iters: 296, time: 0.481, data: 0.002) G_GAN: 0.648 D_real: 0.038 D_fake: 0.292 G: 1.781 NCE: 1.294 NCE_Y: 0.972 \n",
            "End of epoch 69 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 70, iters: 93, time: 0.482, data: 0.003) G_GAN: 0.675 D_real: 0.189 D_fake: 0.158 G: 1.799 NCE: 1.293 NCE_Y: 0.956 \n",
            "(epoch: 70, iters: 193, time: 0.482, data: 0.003) G_GAN: 0.383 D_real: 0.330 D_fake: 0.153 G: 1.711 NCE: 1.800 NCE_Y: 0.857 \n",
            "(epoch: 70, iters: 293, time: 0.481, data: 0.003) G_GAN: 0.542 D_real: 0.409 D_fake: 0.033 G: 1.732 NCE: 1.447 NCE_Y: 0.933 \n",
            "saving the model at the end of epoch 70, iters 21210\n",
            "End of epoch 70 / 400 \t Time Taken: 149 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 71, iters: 90, time: 0.481, data: 0.002) G_GAN: 0.281 D_real: 0.289 D_fake: 0.145 G: 1.534 NCE: 1.558 NCE_Y: 0.948 \n",
            "(epoch: 71, iters: 190, time: 0.481, data: 0.002) G_GAN: 0.201 D_real: 0.204 D_fake: 0.433 G: 1.404 NCE: 1.400 NCE_Y: 1.005 \n",
            "(epoch: 71, iters: 290, time: 0.481, data: 0.003) G_GAN: 0.365 D_real: 0.478 D_fake: 0.091 G: 1.737 NCE: 1.859 NCE_Y: 0.886 \n",
            "End of epoch 71 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 72, iters: 87, time: 0.481, data: 0.003) G_GAN: 0.381 D_real: 0.231 D_fake: 0.367 G: 1.546 NCE: 1.423 NCE_Y: 0.906 \n",
            "(epoch: 72, iters: 187, time: 0.481, data: 0.002) G_GAN: 0.584 D_real: 0.161 D_fake: 0.205 G: 1.735 NCE: 1.323 NCE_Y: 0.979 \n",
            "(epoch: 72, iters: 287, time: 0.481, data: 0.003) G_GAN: 0.329 D_real: 0.177 D_fake: 0.334 G: 1.430 NCE: 1.272 NCE_Y: 0.930 \n",
            "End of epoch 72 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 73, iters: 84, time: 0.481, data: 0.003) G_GAN: 0.194 D_real: 0.514 D_fake: 0.154 G: 1.389 NCE: 1.393 NCE_Y: 0.997 \n",
            "(epoch: 73, iters: 184, time: 0.481, data: 0.003) G_GAN: 0.688 D_real: 0.055 D_fake: 0.150 G: 2.207 NCE: 1.704 NCE_Y: 1.333 \n",
            "(epoch: 73, iters: 284, time: 0.481, data: 0.002) G_GAN: 0.470 D_real: 0.174 D_fake: 0.174 G: 1.583 NCE: 1.275 NCE_Y: 0.952 \n",
            "End of epoch 73 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 74, iters: 81, time: 0.481, data: 0.003) G_GAN: 0.419 D_real: 0.258 D_fake: 0.064 G: 1.545 NCE: 1.294 NCE_Y: 0.957 \n",
            "(epoch: 74, iters: 181, time: 0.480, data: 0.003) G_GAN: 0.374 D_real: 0.309 D_fake: 0.127 G: 1.646 NCE: 1.625 NCE_Y: 0.918 \n",
            "(epoch: 74, iters: 281, time: 0.481, data: 0.003) G_GAN: 0.519 D_real: 0.196 D_fake: 0.021 G: 1.657 NCE: 1.265 NCE_Y: 1.011 \n",
            "End of epoch 74 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 75, iters: 78, time: 0.481, data: 0.002) G_GAN: 0.662 D_real: 0.013 D_fake: 0.186 G: 1.796 NCE: 1.119 NCE_Y: 1.148 \n",
            "(epoch: 75, iters: 178, time: 0.481, data: 0.003) G_GAN: 0.366 D_real: 0.230 D_fake: 0.178 G: 1.533 NCE: 1.443 NCE_Y: 0.890 \n",
            "(epoch: 75, iters: 278, time: 0.481, data: 0.003) G_GAN: 0.445 D_real: 0.217 D_fake: 0.177 G: 1.524 NCE: 1.262 NCE_Y: 0.895 \n",
            "saving the model at the end of epoch 75, iters 22725\n",
            "End of epoch 75 / 400 \t Time Taken: 147 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 76, iters: 75, time: 0.481, data: 0.002) G_GAN: 0.482 D_real: 0.143 D_fake: 0.278 G: 1.554 NCE: 1.233 NCE_Y: 0.910 \n",
            "(epoch: 76, iters: 175, time: 0.481, data: 0.002) G_GAN: 0.441 D_real: 0.354 D_fake: 0.075 G: 1.698 NCE: 1.322 NCE_Y: 1.190 \n",
            "(epoch: 76, iters: 275, time: 0.481, data: 0.003) G_GAN: 0.397 D_real: 0.101 D_fake: 0.284 G: 1.416 NCE: 1.093 NCE_Y: 0.945 \n",
            "End of epoch 76 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 77, iters: 72, time: 0.481, data: 0.003) G_GAN: 0.308 D_real: 0.474 D_fake: 0.129 G: 1.384 NCE: 1.143 NCE_Y: 1.009 \n",
            "(epoch: 77, iters: 172, time: 0.481, data: 0.003) G_GAN: 0.286 D_real: 0.251 D_fake: 0.243 G: 1.561 NCE: 1.689 NCE_Y: 0.859 \n",
            "(epoch: 77, iters: 272, time: 0.480, data: 0.002) G_GAN: 0.105 D_real: 0.260 D_fake: 0.246 G: 1.241 NCE: 1.313 NCE_Y: 0.958 \n",
            "End of epoch 77 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 78, iters: 69, time: 0.480, data: 0.002) G_GAN: 0.374 D_real: 0.066 D_fake: 0.227 G: 1.420 NCE: 1.127 NCE_Y: 0.966 \n",
            "(epoch: 78, iters: 169, time: 0.481, data: 0.003) G_GAN: 0.572 D_real: 0.045 D_fake: 0.251 G: 2.013 NCE: 1.828 NCE_Y: 1.054 \n",
            "(epoch: 78, iters: 269, time: 0.480, data: 0.002) G_GAN: 0.492 D_real: 0.250 D_fake: 0.050 G: 1.687 NCE: 1.391 NCE_Y: 0.999 \n",
            "End of epoch 78 / 400 \t Time Taken: 148 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 79, iters: 66, time: 0.480, data: 0.002) G_GAN: 0.439 D_real: 0.332 D_fake: 0.083 G: 1.675 NCE: 1.632 NCE_Y: 0.842 \n",
            "(epoch: 79, iters: 166, time: 0.480, data: 0.002) G_GAN: 0.840 D_real: 0.141 D_fake: 0.068 G: 2.077 NCE: 1.518 NCE_Y: 0.957 \n",
            "(epoch: 79, iters: 266, time: 0.480, data: 0.002) G_GAN: 0.293 D_real: 0.406 D_fake: 0.063 G: 1.622 NCE: 1.640 NCE_Y: 1.018 \n",
            "End of epoch 79 / 400 \t Time Taken: 146 sec\n",
            "learning rate = 0.0002000\n",
            "(epoch: 80, iters: 63, time: 0.481, data: 0.002) G_GAN: 0.413 D_real: 0.240 D_fake: 0.163 G: 1.614 NCE: 1.243 NCE_Y: 1.159 \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AbBSlniqPEiK"
      },
      "source": [
        "##Test"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7vgnZhxLOEHf"
      },
      "source": [
        "python test.py --dataroot ./datasets/grumpifycat --name grumpycat_CUT --CUT_mode CUT --phase train"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}