nb/cvae/CVAE.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conditional Variational Autoencoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "19d0f047",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torchvision import transforms\n",
    "from torch.utils.data import DataLoader, SubsetRandomSampler\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load pallets library from project root\n",
    "import sys, os\n",
    "sys.path.append(os.path.abspath(os.path.join(os.getcwd(), '..', '..')))\n",
    "\n",
    "from pallets import (\n",
    "    images as I,\n",
    "    datasets as DS,\n",
    "    models as M,\n",
    "    logging as L\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "844bbd48",
   "metadata": {},
   "source": [
    "# Settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9b8ec428",
   "metadata": {},
   "outputs": [],
   "source": [
    "SAVE_NAME = 'cvae'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ae1b5c2f",
   "metadata": {},
   "outputs": [],
   "source": [
    "USE_GPU = True\n",
    "LOG_LEVEL = 'INFO'\n",
    "LABELS_FILE = 'pallets_labels.json'\n",
    "\n",
    "TEST_SIZE = 1000\n",
    "EPOCHS = 50\n",
    "LR = 1e-03\n",
    "BATCH_SIZE = 32\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Logger pallets (INFO)>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# To GPU, or not to GPU\n",
    "device = M.get_device(require_gpu=USE_GPU)\n",
    "\n",
    "# Logging\n",
    "L.init_logger(level=LOG_LEVEL, notebook=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "35fef1c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | loading punk images\n",
      "INFO | - image 0\n",
      "INFO | - image 1000\n",
      "INFO | - image 2000\n",
      "INFO | - image 3000\n",
      "INFO | - image 4000\n",
      "INFO | - image 5000\n",
      "INFO | - image 6000\n",
      "INFO | - image 7000\n",
      "INFO | - image 8000\n",
      "INFO | - image 9000\n",
      "INFO | punk images complete\n",
      "INFO | loading punk labels: pallets_labels.json\n"
     ]
    }
   ],
   "source": [
    "all_colors = I.get_punk_colors()\n",
    "mapper = DS.ColorOneHotMapper(all_colors)\n",
    "dataset = DS.OneHotCPunksDataset(\n",
    "    mapper, labels_file=LABELS_FILE, test_size=TEST_SIZE\n",
    ")\n",
    "\n",
    "# Uncomment to put entire dataset in GPU RAM\n",
    "# dataset = DS.FastOneHotCPunksDataset(\n",
    "#     device, mapper, labels_file=LABELS_FILE, test_size=TEST_SIZE\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "8eac2395",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_sampler = SubsetRandomSampler(dataset.train_idx)\n",
    "test_sampler = SubsetRandomSampler(dataset.test_idx)\n",
    "\n",
    "train_loader = DataLoader(dataset, batch_size=BATCH_SIZE, sampler=train_sampler)\n",
    "test_loader = DataLoader(dataset, batch_size=BATCH_SIZE, sampler=test_sampler)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b39e0222",
   "metadata": {},
   "source": [
    "# Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d3fe1530",
   "metadata": {},
   "outputs": [],
   "source": [
    "input_dim = 24 * 24 * 222\n",
    "hidden_dim = 576\n",
    "latent_dim = 32\n",
    "classes_dim = len(dataset._label_keys)  # 92\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "9521e7fe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO | model: pallets.models.cvae.CVAE\n",
      "INFO | criterion: pallets.models.cvae.Loss\n",
      "INFO | learn rate: 0.001\n",
      "INFO | epochs: 50\n",
      "INFO | epoch 1 (  0%) loss: 1422887.375000\n",
      "INFO | epoch 1 ( 35%) loss: 90973.002901\n",
      "INFO | epoch 1 ( 70%) loss: 51420.222360\n",
      "INFO | epoch 1 (100%) loss: 38996.729947\n",
      "INFO | epoch 1 (test) loss: 7079.896622\n",
      "INFO | epoch 2 (  0%) loss: 7252.196289\n",
      "INFO | epoch 2 ( 35%) loss: 6298.391553\n",
      "INFO | epoch 2 ( 70%) loss: 5628.102177\n",
      "INFO | epoch 2 (100%) loss: 5211.681871\n",
      "INFO | epoch 2 (test) loss: 3843.845818\n",
      "INFO | epoch 3 (  0%) loss: 3952.945312\n",
      "INFO | epoch 3 ( 35%) loss: 3591.531513\n",
      "INFO | epoch 3 ( 70%) loss: 3380.693630\n",
      "INFO | epoch 3 (100%) loss: 3216.313632\n",
      "INFO | epoch 3 (test) loss: 2664.408588\n",
      "INFO | epoch 4 (  0%) loss: 2886.278320\n",
      "INFO | epoch 4 ( 35%) loss: 2539.441965\n",
      "INFO | epoch 4 ( 70%) loss: 2422.085775\n",
      "INFO | epoch 4 (100%) loss: 2339.493400\n",
      "INFO | epoch 4 (test) loss: 2127.593605\n",
      "INFO | epoch 5 (  0%) loss: 2065.791260\n",
      "INFO | epoch 5 ( 35%) loss: 1989.315619\n",
      "INFO | epoch 5 ( 70%) loss: 1909.277349\n",
      "INFO | epoch 5 (100%) loss: 1850.515226\n",
      "INFO | epoch 5 (test) loss: 1690.020204\n",
      "INFO | epoch 6 (  0%) loss: 1662.896606\n",
      "INFO | epoch 6 ( 35%) loss: 1615.638401\n",
      "INFO | epoch 6 ( 70%) loss: 1560.787056\n",
      "INFO | epoch 6 (100%) loss: 1517.055904\n",
      "INFO | epoch 6 (test) loss: 1401.694407\n",
      "INFO | epoch 7 (  0%) loss: 1405.105469\n",
      "INFO | epoch 7 ( 35%) loss: 1352.177570\n",
      "INFO | epoch 7 ( 70%) loss: 1316.524452\n",
      "INFO | epoch 7 (100%) loss: 1285.303911\n",
      "INFO | epoch 7 (test) loss: 1202.941669\n",
      "INFO | epoch 8 (  0%) loss: 1204.430908\n",
      "INFO | epoch 8 ( 35%) loss: 1158.342332\n",
      "INFO | epoch 8 ( 70%) loss: 1128.372970\n",
      "INFO | epoch 8 (100%) loss: 1108.598820\n",
      "INFO | epoch 8 (test) loss: 1072.452693\n",
      "INFO | epoch 9 (  0%) loss: 1079.528809\n",
      "INFO | epoch 9 ( 35%) loss: 999.908872\n",
      "INFO | epoch 9 ( 70%) loss: 983.019368\n",
      "INFO | epoch 9 (100%) loss: 969.543716\n",
      "INFO | epoch 9 (test) loss: 964.181181\n",
      "INFO | epoch 10 (  0%) loss: 965.695923\n",
      "INFO | epoch 10 ( 35%) loss: 891.008652\n",
      "INFO | epoch 10 ( 70%) loss: 878.880563\n",
      "INFO | epoch 10 (100%) loss: 860.926684\n",
      "INFO | epoch 10 (test) loss: 838.020078\n",
      "INFO | epoch 11 (  0%) loss: 772.781921\n",
      "INFO | epoch 11 ( 35%) loss: 792.922000\n",
      "INFO | epoch 11 ( 70%) loss: 786.050082\n",
      "INFO | epoch 11 (100%) loss: 775.132509\n",
      "INFO | epoch 11 (test) loss: 767.461267\n",
      "INFO | epoch 12 (  0%) loss: 812.139099\n",
      "INFO | epoch 12 ( 35%) loss: 718.267288\n",
      "INFO | epoch 12 ( 70%) loss: 705.518328\n",
      "INFO | epoch 12 (100%) loss: 696.830317\n",
      "INFO | epoch 12 (test) loss: 692.279434\n",
      "INFO | epoch 13 (  0%) loss: 672.396851\n",
      "INFO | epoch 13 ( 35%) loss: 651.880378\n",
      "INFO | epoch 13 ( 70%) loss: 640.389715\n",
      "INFO | epoch 13 (100%) loss: 627.448913\n",
      "INFO | epoch 13 (test) loss: 627.140863\n",
      "INFO | epoch 14 (  0%) loss: 562.780334\n",
      "INFO | epoch 14 ( 35%) loss: 588.328482\n",
      "INFO | epoch 14 ( 70%) loss: 580.599440\n",
      "INFO | epoch 14 (100%) loss: 572.043401\n",
      "INFO | epoch 14 (test) loss: 562.892701\n",
      "INFO | epoch 15 (  0%) loss: 484.623627\n",
      "INFO | epoch 15 ( 35%) loss: 534.646434\n",
      "INFO | epoch 15 ( 70%) loss: 526.533734\n",
      "INFO | epoch 15 (100%) loss: 518.887887\n",
      "INFO | epoch 15 (test) loss: 528.138076\n",
      "INFO | epoch 16 (  0%) loss: 502.095551\n",
      "INFO | epoch 16 ( 35%) loss: 486.437920\n",
      "INFO | epoch 16 ( 70%) loss: 481.555071\n",
      "INFO | epoch 16 (100%) loss: 473.151819\n",
      "INFO | epoch 16 (test) loss: 473.533591\n",
      "INFO | epoch 17 (  0%) loss: 449.519165\n",
      "INFO | epoch 17 ( 35%) loss: 441.484015\n",
      "INFO | epoch 17 ( 70%) loss: 438.798391\n",
      "INFO | epoch 17 (100%) loss: 433.365829\n",
      "INFO | epoch 17 (test) loss: 440.816876\n",
      "INFO | epoch 18 (  0%) loss: 399.495850\n",
      "INFO | epoch 18 ( 35%) loss: 408.530015\n",
      "INFO | epoch 18 ( 70%) loss: 402.149140\n",
      "INFO | epoch 18 (100%) loss: 397.606385\n",
      "INFO | epoch 18 (test) loss: 401.019530\n",
      "INFO | epoch 19 (  0%) loss: 453.003021\n",
      "INFO | epoch 19 ( 35%) loss: 377.895344\n",
      "INFO | epoch 19 ( 70%) loss: 368.990254\n",
      "INFO | epoch 19 (100%) loss: 366.206060\n",
      "INFO | epoch 19 (test) loss: 371.304341\n",
      "INFO | epoch 20 (  0%) loss: 335.460205\n",
      "INFO | epoch 20 ( 35%) loss: 343.095791\n",
      "INFO | epoch 20 ( 70%) loss: 340.002079\n",
      "INFO | epoch 20 (100%) loss: 335.803431\n",
      "INFO | epoch 20 (test) loss: 335.894865\n",
      "INFO | epoch 21 (  0%) loss: 296.802490\n",
      "INFO | epoch 21 ( 35%) loss: 312.164853\n",
      "INFO | epoch 21 ( 70%) loss: 308.875708\n",
      "INFO | epoch 21 (100%) loss: 306.165646\n",
      "INFO | epoch 21 (test) loss: 310.249971\n",
      "INFO | epoch 22 (  0%) loss: 286.341583\n",
      "INFO | epoch 22 ( 35%) loss: 289.051791\n",
      "INFO | epoch 22 ( 70%) loss: 288.072267\n",
      "INFO | epoch 22 (100%) loss: 284.024637\n",
      "INFO | epoch 22 (test) loss: 280.037339\n",
      "INFO | epoch 23 (  0%) loss: 242.638611\n",
      "INFO | epoch 23 ( 35%) loss: 271.147570\n",
      "INFO | epoch 23 ( 70%) loss: 264.327403\n",
      "INFO | epoch 23 (100%) loss: 260.964499\n",
      "INFO | epoch 23 (test) loss: 274.264100\n",
      "INFO | epoch 24 (  0%) loss: 289.305695\n",
      "INFO | epoch 24 ( 35%) loss: 244.254910\n",
      "INFO | epoch 24 ( 70%) loss: 242.638935\n",
      "INFO | epoch 24 (100%) loss: 239.258488\n",
      "INFO | epoch 24 (test) loss: 251.874077\n",
      "INFO | epoch 25 (  0%) loss: 244.558105\n",
      "INFO | epoch 25 ( 35%) loss: 229.502627\n",
      "INFO | epoch 25 ( 70%) loss: 229.493192\n",
      "INFO | epoch 25 (100%) loss: 226.526710\n",
      "INFO | epoch 25 (test) loss: 223.854063\n",
      "INFO | epoch 26 (  0%) loss: 181.573822\n",
      "INFO | epoch 26 ( 35%) loss: 210.189668\n",
      "INFO | epoch 26 ( 70%) loss: 207.527818\n",
      "INFO | epoch 26 (100%) loss: 203.983125\n",
      "INFO | epoch 26 (test) loss: 208.331933\n",
      "INFO | epoch 27 (  0%) loss: 239.993164\n",
      "INFO | epoch 27 ( 35%) loss: 189.985752\n",
      "INFO | epoch 27 ( 70%) loss: 191.200293\n",
      "INFO | epoch 27 (100%) loss: 191.034284\n",
      "INFO | epoch 27 (test) loss: 206.462543\n",
      "INFO | epoch 28 (  0%) loss: 169.903488\n",
      "INFO | epoch 28 ( 35%) loss: 178.847659\n",
      "INFO | epoch 28 ( 70%) loss: 179.523196\n",
      "INFO | epoch 28 (100%) loss: 178.393166\n",
      "INFO | epoch 28 (test) loss: 195.238560\n",
      "INFO | epoch 29 (  0%) loss: 222.561859\n",
      "INFO | epoch 29 ( 35%) loss: 171.454504\n",
      "INFO | epoch 29 ( 70%) loss: 173.221006\n",
      "INFO | epoch 29 (100%) loss: 170.446161\n",
      "INFO | epoch 29 (test) loss: 177.570419\n",
      "INFO | epoch 30 (  0%) loss: 163.041199\n",
      "INFO | epoch 30 ( 35%) loss: 167.432833\n",
      "INFO | epoch 30 ( 70%) loss: 163.355911\n",
      "INFO | epoch 30 (100%) loss: 163.100523\n",
      "INFO | epoch 30 (test) loss: 167.642175\n",
      "INFO | epoch 31 (  0%) loss: 161.902527\n",
      "INFO | epoch 31 ( 35%) loss: 149.450778\n",
      "INFO | epoch 31 ( 70%) loss: 150.585629\n",
      "INFO | epoch 31 (100%) loss: 149.627506\n",
      "INFO | epoch 31 (test) loss: 152.908397\n",
      "INFO | epoch 32 (  0%) loss: 155.796280\n",
      "INFO | epoch 32 ( 35%) loss: 135.990653\n",
      "INFO | epoch 32 ( 70%) loss: 143.098346\n",
      "INFO | epoch 32 (100%) loss: 140.027499\n",
      "INFO | epoch 32 (test) loss: 152.978428\n",
      "INFO | epoch 33 (  0%) loss: 111.170067\n",
      "INFO | epoch 33 ( 35%) loss: 132.036911\n",
      "INFO | epoch 33 ( 70%) loss: 130.919764\n",
      "INFO | epoch 33 (100%) loss: 129.458076\n",
      "INFO | epoch 33 (test) loss: 147.489411\n",
      "INFO | epoch 34 (  0%) loss: 109.607239\n",
      "INFO | epoch 34 ( 35%) loss: 165.543425\n",
      "INFO | epoch 34 ( 70%) loss: 148.458235\n",
      "INFO | epoch 34 (100%) loss: 141.181535\n",
      "INFO | epoch 34 (test) loss: 134.016671\n",
      "INFO | epoch 35 (  0%) loss: 113.099304\n",
      "INFO | epoch 35 ( 35%) loss: 119.846042\n",
      "INFO | epoch 35 ( 70%) loss: 121.221982\n",
      "INFO | epoch 35 (100%) loss: 119.546434\n",
      "INFO | epoch 35 (test) loss: 129.577571\n",
      "INFO | epoch 36 (  0%) loss: 107.482361\n",
      "INFO | epoch 36 ( 35%) loss: 118.259935\n",
      "INFO | epoch 36 ( 70%) loss: 114.418997\n",
      "INFO | epoch 36 (100%) loss: 115.706434\n",
      "INFO | epoch 36 (test) loss: 135.409795\n",
      "INFO | epoch 37 (  0%) loss: 142.347214\n",
      "INFO | epoch 37 ( 35%) loss: 106.547493\n",
      "INFO | epoch 37 ( 70%) loss: 107.405305\n",
      "INFO | epoch 37 (100%) loss: 107.313782\n",
      "INFO | epoch 37 (test) loss: 115.969040\n",
      "INFO | epoch 38 (  0%) loss: 105.558899\n",
      "INFO | epoch 38 ( 35%) loss: 105.762894\n",
      "INFO | epoch 38 ( 70%) loss: 105.087370\n",
      "INFO | epoch 38 (100%) loss: 103.492594\n",
      "INFO | epoch 38 (test) loss: 115.981524\n",
      "INFO | epoch 39 (  0%) loss: 97.054688\n",
      "INFO | epoch 39 ( 35%) loss: 109.959447\n",
      "INFO | epoch 39 ( 70%) loss: 108.561257\n",
      "INFO | epoch 39 (100%) loss: 107.194936\n",
      "INFO | epoch 39 (test) loss: 123.119338\n",
      "INFO | epoch 40 (  0%) loss: 96.722099\n",
      "INFO | epoch 40 ( 35%) loss: 107.525156\n",
      "INFO | epoch 40 ( 70%) loss: 101.609633\n",
      "INFO | epoch 40 (100%) loss: 100.658910\n",
      "INFO | epoch 40 (test) loss: 107.436706\n",
      "INFO | epoch 41 (  0%) loss: 90.962791\n",
      "INFO | epoch 41 ( 35%) loss: 102.348862\n",
      "INFO | epoch 41 ( 70%) loss: 100.096526\n",
      "INFO | epoch 41 (100%) loss: 102.870939\n",
      "INFO | epoch 41 (test) loss: 111.233327\n",
      "INFO | epoch 42 (  0%) loss: 102.811951\n",
      "INFO | epoch 42 ( 35%) loss: 101.603687\n",
      "INFO | epoch 42 ( 70%) loss: 105.370946\n",
      "INFO | epoch 42 (100%) loss: 103.570248\n",
      "INFO | epoch 42 (test) loss: 121.109555\n",
      "INFO | epoch 43 (  0%) loss: 90.410202\n",
      "INFO | epoch 43 ( 35%) loss: 102.753502\n",
      "INFO | epoch 43 ( 70%) loss: 102.078284\n",
      "INFO | epoch 43 (100%) loss: 100.079712\n",
      "INFO | epoch 43 (test) loss: 128.211039\n",
      "INFO | epoch 44 (  0%) loss: 95.855476\n",
      "INFO | epoch 44 ( 35%) loss: 105.181576\n",
      "INFO | epoch 44 ( 70%) loss: 102.598658\n",
      "INFO | epoch 44 (100%) loss: 106.844522\n",
      "INFO | epoch 44 (test) loss: 114.335600\n",
      "INFO | epoch 45 (  0%) loss: 94.986794\n",
      "INFO | epoch 45 ( 35%) loss: 100.097543\n",
      "INFO | epoch 45 ( 70%) loss: 97.160113\n",
      "INFO | epoch 45 (100%) loss: 97.059401\n",
      "INFO | epoch 45 (test) loss: 103.243093\n",
      "INFO | epoch 46 (  0%) loss: 81.093018\n",
      "INFO | epoch 46 ( 35%) loss: 103.479680\n",
      "INFO | epoch 46 ( 70%) loss: 112.794185\n",
      "INFO | epoch 46 (100%) loss: 110.376595\n",
      "INFO | epoch 46 (test) loss: 120.758259\n",
      "INFO | epoch 47 (  0%) loss: 112.856918\n",
      "INFO | epoch 47 ( 35%) loss: 98.317698\n",
      "INFO | epoch 47 ( 70%) loss: 99.967633\n",
      "INFO | epoch 47 (100%) loss: 101.208050\n",
      "INFO | epoch 47 (test) loss: 147.431703\n",
      "INFO | epoch 48 (  0%) loss: 154.021576\n",
      "INFO | epoch 48 ( 35%) loss: 112.425756\n",
      "INFO | epoch 48 ( 70%) loss: 114.979028\n",
      "INFO | epoch 48 (100%) loss: 115.083640\n",
      "INFO | epoch 48 (test) loss: 146.712965\n",
      "INFO | epoch 49 (  0%) loss: 116.538826\n",
      "INFO | epoch 49 ( 35%) loss: 126.980562\n",
      "INFO | epoch 49 ( 70%) loss: 129.232897\n",
      "INFO | epoch 49 (100%) loss: 128.622187\n",
      "INFO | epoch 49 (test) loss: 135.030421\n",
      "INFO | epoch 50 (  0%) loss: 103.008781\n",
      "INFO | epoch 50 ( 35%) loss: 119.959253\n",
      "INFO | epoch 50 ( 70%) loss: 121.489050\n",
      "INFO | epoch 50 (100%) loss: 121.852806\n",
      "INFO | epoch 50 (test) loss: 137.946773\n",
      "INFO | model blob saved to /home/jmsdnns/ML/pallets/saved/cvae.pkl\n",
      "INFO | training losses saved to /home/jmsdnns/ML/pallets/saved/cvae.json\n"
     ]
    }
   ],
   "source": [
    "model = M.cvae.CVAE(input_dim, hidden_dim, latent_dim, classes_dim)\n",
    "criterion = M.cvae.Loss()\n",
    "\n",
    "train_losses, test_losses = M.cvae.train(\n",
    "    device, model, criterion, train_loader, test_loader,\n",
    "    learn_rate=LR, epochs=EPOCHS\n",
    ")\n",
    "\n",
    "M.save(SAVE_NAME, model, train_losses, test_losses)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "77fb32ad",
   "metadata": {},
   "source": [
    "# Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46b6dd3b",
   "metadata": {},
   "source": [
    "## Random Inputs w/ Real Labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f2b9d212",
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "\n",
    "def rand_label():\n",
    "    label_idx = int(random.random() * len(dataset._labels))\n",
    "    features = dataset._labels[label_idx]\n",
    "    enabled_names = [k for k,v in zip(dataset._label_keys, features) if v.item() == 1]\n",
    "    return features.to(device), enabled_names\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "38c47d81",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate new image\n",
    "\n",
    "def rand_punk():\n",
    "    z = torch.randn(1, latent_dim).to(device)\n",
    "    # print(z.shape)\n",
    "    features, names = rand_label()\n",
    "    print(f\"Features: {', '.join(names)}\")\n",
    "\n",
    "    with torch.no_grad():\n",
    "        model.eval()\n",
    "        generated_image = model.decoder(z, features.unsqueeze(0))\n",
    "        # print(generated_image.shape)\n",
    "\n",
    "    decoded_one_hot = generated_image[0]\n",
    "    decoded_one_hot = decoded_one_hot.view((222, 24, 24))\n",
    "    # print(decoded_one_hot.shape)\n",
    "    decoded = DS.one_hot_to_rgba(decoded_one_hot, dataset.mapper)\n",
    "    print(f\"Shape: {decoded.shape}\")\n",
    "    return decoded\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31e74ceb",
   "metadata": {},
   "source": [
    "### 5 Randoms w/ Real Labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: male, clownEyesBlue, crazyHair, cigarette\n",
      "Shape: torch.Size([4, 24, 24])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "decoded = rand_punk()\n",
    "\n",
    "plt.imshow(transforms.functional.to_pil_image(decoded))\n",
    "plt.axis('off')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6744444a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: male, peakSpike, eyePatch\n",
      "Shape: torch.Size([4, 24, 24])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "decoded = rand_punk()\n",
    "\n",
    "plt.imshow(transforms.functional.to_pil_image(decoded))\n",
    "plt.axis('off')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e190c583",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: male, mohawkThin, mole\n",
      "Shape: torch.Size([4, 24, 24])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "decoded = rand_punk()\n",
    "\n",
    "plt.imshow(transforms.functional.to_pil_image(decoded))\n",
    "plt.axis('off')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "be357c27",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: female, purpleEyeShadow, cap, cigarette\n",
      "Shape: torch.Size([4, 24, 24])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "decoded = rand_punk()\n",
    "\n",
    "plt.imshow(transforms.functional.to_pil_image(decoded))\n",
    "plt.axis('off')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d8f339ae",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: male, stringyHair\n",
      "Shape: torch.Size([4, 24, 24])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "decoded = rand_punk()\n",
    "\n",
    "plt.imshow(transforms.functional.to_pil_image(decoded))\n",
    "plt.axis('off')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e1bd761",
   "metadata": {},
   "source": [
    "## Reconstruction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a150efab",
   "metadata": {},
   "outputs": [],
   "source": [
    "def reconstruct_punk(idx):\n",
    "    punk = I.get_punk_tensor(idx)\n",
    "    p = DS.rgba_to_one_hot(punk, dataset.mapper)\n",
    "    p = p.unsqueeze(0)\n",
    "    p = p.to(device)\n",
    "\n",
    "    _, labels = dataset[idx]\n",
    "    l = labels.unsqueeze(0)\n",
    "    l = l.to(device)\n",
    "    enabled_features = [k for k,v in zip(dataset._label_keys, labels) if v.item() == 1]\n",
    "\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        reconstructed, mu, logvar = model.forward(p, l)\n",
    "\n",
    "    recon_punk = reconstructed[0].cpu()\n",
    "    recon_punk = DS.one_hot_to_rgba(recon_punk, dataset.mapper)\n",
    "\n",
    "    return punk, recon_punk, enabled_features\n",
    "\n",
    "\n",
    "def draw_two(img1, img2):\n",
    "    page_size = 2\n",
    "    view_x, view_y = 4*page_size, 2*page_size\n",
    "    fig = plt.figure(figsize=(view_x, view_y))\n",
    "    fig.add_subplot(1, 2, 1)\n",
    "    plt.imshow(transforms.functional.to_pil_image(img1))\n",
    "    plt.axis('off')\n",
    "    fig.add_subplot(1, 2, 2)\n",
    "    plt.imshow(transforms.functional.to_pil_image(img2))\n",
    "    plt.axis('off')\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a12b220",
   "metadata": {},
   "source": [
    "### 5 Reconstructions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "142e642d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: female, rosyCheeks, darkHair, cigarette\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "punk, recon_punk, features = reconstruct_punk(1000)\n",
    "\n",
    "print(f\"Features: {', '.join(features)}\")\n",
    "draw_two(punk, recon_punk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: female, stringyHair, purpleLipstick, earring\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "punk, recon_punk, features = reconstruct_punk(2001)\n",
    "\n",
    "print(f\"Features: {', '.join(features)}\")\n",
    "draw_two(punk, recon_punk)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "8e401eca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: female, wildWhiteHair, greenEyeShadow, hotLipstick\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "punk, recon_punk, features = reconstruct_punk(5000)\n",
    "\n",
    "print(f\"Features: {', '.join(features)}\")\n",
    "draw_two(punk, recon_punk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "99f3ac30",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: male, rosyCheeks, mustache, vr, cap, cigarette\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "punk, recon_punk, features = reconstruct_punk(8000)\n",
    "\n",
    "print(f\"Features: {', '.join(features)}\")\n",
    "draw_two(punk, recon_punk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "eb402159",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Features: male, handlebars, bandana\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "punk, recon_punk, features = reconstruct_punk(1337)\n",
    "\n",
    "print(f\"Features: {', '.join(features)}\")\n",
    "draw_two(punk, recon_punk)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76e07a58",
   "metadata": {},
   "source": [
    "## Losses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function matplotlib.pyplot.show(close=None, block=None)>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(\"Train & Test loss\")\n",
    "plt.plot(train_losses, label='train loss')\n",
    "plt.plot(test_losses, label='test loss')\n",
    "plt.legend()\n",
    "plt.show\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c1261252",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[77993.45989340924,\n",
       " 10423.36374182735,\n",
       " 6432.627264793883,\n",
       " 4678.986799524186,\n",
       " 3701.0304517813606,\n",
       " 3034.1118073159078,\n",
       " 2570.6078216336296,\n",
       " 2217.197640060533,\n",
       " 1939.0874317791445,\n",
       " 1721.8533670411887,\n",
       " 1550.2650185442985,\n",
       " 1393.6606340340688,\n",
       " 1254.897825849817,\n",
       " 1144.0868023811502,\n",
       " 1037.7757749625132,\n",
       " 946.3036388640708,\n",
       " 866.7316576369266,\n",
       " 795.2127706108364,\n",
       " 732.4121194393077,\n",
       " 671.6068627107228,\n",
       " 612.3312921726957,\n",
       " 568.0492743634163,\n",
       " 521.9289981490331,\n",
       " 478.5169764309065,\n",
       " 453.05342015476094,\n",
       " 407.96625039932576,\n",
       " 382.0685673977466,\n",
       " 356.7863326918149,\n",
       " 340.89232271613804,\n",
       " 326.20104653108206,\n",
       " 299.2550123769341,\n",
       " 280.05499803259016,\n",
       " 258.91615262944646,\n",
       " 282.36307017008465,\n",
       " 239.09286791213015,\n",
       " 231.4128674716814,\n",
       " 214.62756449110964,\n",
       " 206.9851871788079,\n",
       " 214.38987203016347,\n",
       " 201.31782082632077,\n",
       " 205.74187740028327,\n",
       " 207.1404960984034,\n",
       " 200.1594230164873,\n",
       " 213.6890443842462,\n",
       " 194.11880233439993,\n",
       " 220.75319003382472,\n",
       " 202.41609981888575,\n",
       " 230.167280034816,\n",
       " 257.24437497186324,\n",
       " 243.70561177679832]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_losses\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "15d6f86b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[14159.793243408203,\n",
       " 7687.691635131836,\n",
       " 5328.817176818848,\n",
       " 4255.187210083008,\n",
       " 3380.040407180786,\n",
       " 2803.3888149261475,\n",
       " 2405.8833389282227,\n",
       " 2144.9053859710693,\n",
       " 1928.362361907959,\n",
       " 1676.0401554107666,\n",
       " 1534.922534942627,\n",
       " 1384.5588684082031,\n",
       " 1254.2817268371582,\n",
       " 1125.7854022979736,\n",
       " 1056.2761521339417,\n",
       " 947.0671820640564,\n",
       " 881.6337523460388,\n",
       " 802.0390605926514,\n",
       " 742.608681678772,\n",
       " 671.7897305488586,\n",
       " 620.4999413490295,\n",
       " 560.0746772289276,\n",
       " 548.5281991958618,\n",
       " 503.7481541633606,\n",
       " 447.7081253528595,\n",
       " 416.66386556625366,\n",
       " 412.9250867366791,\n",
       " 390.4771201610565,\n",
       " 355.1408386230469,\n",
       " 335.2843506336212,\n",
       " 305.8167943954468,\n",
       " 305.95685625076294,\n",
       " 294.9788212776184,\n",
       " 268.0333411693573,\n",
       " 259.15514290332794,\n",
       " 270.81958961486816,\n",
       " 231.93807983398438,\n",
       " 231.9630470275879,\n",
       " 246.23867630958557,\n",
       " 214.87341284751892,\n",
       " 222.46665477752686,\n",
       " 242.21910989284515,\n",
       " 256.4220778942108,\n",
       " 228.67119932174683,\n",
       " 206.48618698120117,\n",
       " 241.51651895046234,\n",
       " 294.86340618133545,\n",
       " 293.42593026161194,\n",
       " 270.0608425140381,\n",
       " 275.8935457468033]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_losses\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}