oe-c100.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os \n",
    "import sys\n",
    "\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"MIG-GPU-bb1ccb6e-2bc9-c7a1-b25d-3eef9033e192/3/0\"\n",
    "\n",
    "n_epochs = 25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch \n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sb "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pytorch_ood.model import WideResNet\n",
    "from pytorch_ood.detector import EnergyBased, MaxSoftmax\n",
    "from pytorch_ood.utils import OODMetrics, ToUnknown, ToRGB\n",
    "from pytorch_ood.dataset.img import Textures, LSUNCrop, LSUNResize, TinyImageNetCrop, TinyImageNetResize\n",
    "from torchvision.datasets import CIFAR100\n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "from torchvision.transforms import Compose, ToPILImage, Normalize, ToTensor, Resize, ToPILImage\n",
    "import torch \n",
    "\n",
    "def test(model):\n",
    "    \n",
    "    prep = Compose([\n",
    "        ToRGB(),\n",
    "        Resize([32,32]),\n",
    "        ToTensor(),\n",
    "        Normalize(\n",
    "            std=[0.24705882352941176470, 0.24352941176470588235, 0.26156862745098039215], \n",
    "            mean=[0.49137254901960784313, 0.48235294117647058823, 0.44666666666666666666]), \n",
    "        ])\n",
    "\n",
    "    metrics = [] \n",
    "    \n",
    "    with torch.no_grad():\n",
    "        model.eval()\n",
    "\n",
    "        data_in = CIFAR100(root=\"/data_fast/kirchheim/datasets/\", download=True, transform=prep, train=False)\n",
    "\n",
    "        for Data_clz in [Textures, LSUNCrop, LSUNResize, TinyImageNetCrop, TinyImageNetResize]:\n",
    "            ood_name = Data_clz.__name__\n",
    "\n",
    "            data_out = Data_clz(root=\"/data_fast/kirchheim/datasets/\", transform=prep, download=True, \n",
    "                                target_transform=ToUnknown())\n",
    "\n",
    "            data_loader = DataLoader(data_in + data_out, batch_size=128, num_workers=10)\n",
    "\n",
    "            # create detector\n",
    "            softmax = MaxSoftmax(model)\n",
    "            energy = EnergyBased(model)\n",
    "\n",
    "            # evaluate\n",
    "            metric_softmax = OODMetrics()\n",
    "            metric_energy = OODMetrics()\n",
    "\n",
    "            for x, y in data_loader:\n",
    "                logits = model(x.cuda())\n",
    "                metric_softmax.update(softmax.score(logits), y)\n",
    "                metric_energy.update(energy.score(logits), y)\n",
    "\n",
    "            # print(ood_name, metric_softmax.compute())\n",
    "\n",
    "            m = metric_softmax.compute()\n",
    "            m.update({\n",
    "                \"Dataset\": ood_name,\n",
    "                \"Method\": \"Softmax\"\n",
    "            })\n",
    "            metrics.append(m)\n",
    "            \n",
    "            m = metric_energy.compute()\n",
    "            m.update({\n",
    "                \"Dataset\": ood_name,\n",
    "                \"Method\": \"Energy\"\n",
    "            })\n",
    "            metrics.append(m)\n",
    "    \n",
    "    model.train()\n",
    "    \n",
    "    return metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `AUROC` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `PrecisionRecallCurve` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: `pos_label` automatically set 1.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/functional/classification/precision_recall_curve.py:156: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
      "  thresholds = tensor(reversed(thresholds[sl]))\n"
     ]
    }
   ],
   "source": [
    "# create Neural Network\n",
    "model = WideResNet(pretrained=\"cifar100-pt\", num_classes=100).eval().cuda()\n",
    "metrics = test(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrrr}\n",
      "\\toprule\n",
      "{} &  AUROC &  AUPR-IN &  AUPR-OUT &  ACC95TPR &  FPR95TPR \\\\\n",
      "Method  &        &          &           &           &           \\\\\n",
      "\\midrule\n",
      "Energy  &  84.41 &    79.72 &     86.85 &     71.38 &     48.55 \\\\\n",
      "Softmax &  78.71 &    73.35 &     82.58 &     66.36 &     58.58 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd \n",
    "\n",
    "data = pd.DataFrame(metrics)\n",
    "print((data.groupby([\"Method\"]).mean() * 100).to_latex(float_format=\"%.2f\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `AUROC` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `PrecisionRecallCurve` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: `pos_label` automatically set 1.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/functional/classification/precision_recall_curve.py:156: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
      "  thresholds = tensor(reversed(thresholds[sl]))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.22it/s, loss=1.97]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.23it/s, loss=1.76]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.22it/s, loss=1.58]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.21it/s, loss=1.58]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.23it/s, loss=1.52]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.23it/s, loss=1.52]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.23it/s, loss=1.48]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.23it/s, loss=1.43]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 8\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:01<00:00,  3.23it/s, loss=1.39]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 9\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.45]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.38]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 11\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.38]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.37]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.42]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 14\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.34]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.4]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 16\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.38]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 17\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.41]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.38]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.36]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.34]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.33]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.35]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 23\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.25it/s, loss=1.35]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 24\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████| 391/391 [02:00<00:00,  3.24it/s, loss=1.34]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "from pytorch_ood.model import WideResNet\n",
    "from pytorch_ood.detector import EnergyBased, MaxSoftmax\n",
    "from pytorch_ood.utils import OODMetrics\n",
    "from pytorch_ood.dataset.img import Textures\n",
    "from torchvision.datasets import CIFAR100 \n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "from torchvision.transforms import Compose, ToPILImage, Normalize, ToTensor, Resize, ToPILImage\n",
    "import torch \n",
    "\n",
    "from pytorch_ood.loss import OutlierExposureLoss\n",
    "from torch.optim import SGD\n",
    "from tqdm import tqdm\n",
    "\n",
    "torch.manual_seed(10)\n",
    "\n",
    "prep = Compose([\n",
    "    ToPILImage(),\n",
    "    Resize([32,32]),\n",
    "    ToTensor(),\n",
    "    Normalize(\n",
    "        std=[0.24705882352941176470, 0.24352941176470588235, 0.26156862745098039215], \n",
    "        mean=[0.49137254901960784313, 0.48235294117647058823, 0.44666666666666666666]), \n",
    "    ])\n",
    "\n",
    "mt = Compose([\n",
    "    #ToPILImage(),\n",
    "    Resize([32,32]),\n",
    "    ToTensor(),\n",
    "    Normalize(\n",
    "        std=[0.24705882352941176470, 0.24352941176470588235, 0.26156862745098039215], \n",
    "        mean=[0.49137254901960784313, 0.48235294117647058823, 0.44666666666666666666]), \n",
    "    ])\n",
    "\n",
    "x = torch.load(\"samples-c100.pt\")\n",
    "gan_oe_data = torch.stack([prep(t) for t in x])\n",
    "gan_oe_label = torch.ones(size=(len(x),)).long() * -1\n",
    "\n",
    "data_in_train = CIFAR100(root=\"data/\", download=True, transform=mt, target_transform=torch.tensor)\n",
    "data_out_train = TensorDataset(gan_oe_data,gan_oe_label)\n",
    "train_loader = DataLoader(data_in_train + data_out_train, batch_size=256, shuffle=True, num_workers=20)\n",
    "      \n",
    "# create Neural Network\n",
    "model = WideResNet(pretrained=\"cifar100-pt\", num_classes=100).cuda()\n",
    "opti = SGD(model.parameters(), lr=0.001)\n",
    "\n",
    "criterion = OutlierExposureLoss(alpha=0.5)\n",
    "\n",
    "mav = 0.0\n",
    "mavs = []\n",
    "\n",
    "metrics_goe = []\n",
    "\n",
    "\n",
    "# initial test round \n",
    "r = test(model)\n",
    "for entry in r:\n",
    "    entry.update({\"Epoch\": -1})\n",
    "metrics_goe += r\n",
    "\n",
    "for i in range(n_epochs):\n",
    "    print(f\"Epoch {i}\")\n",
    "    \n",
    "    # train \n",
    "    model.train()\n",
    "    bar = tqdm(train_loader)\n",
    "    for x, y in bar:\n",
    "        opti.zero_grad()\n",
    "        y_hat = model(x.cuda())\n",
    "        loss = criterion(y_hat, y.cuda())\n",
    "        mav = 0.2 * loss.item() + 0.8 * mav\n",
    "        mavs.append(mav)\n",
    "\n",
    "        loss.backward()\n",
    "        opti.step()\n",
    "        bar.set_postfix({\"loss\": mav})\n",
    "        \n",
    "    print(80*\"#\")\n",
    "    \n",
    "    r = test(model)\n",
    "    for entry in r:\n",
    "        entry.update({\"Epoch\": i})\n",
    "    \n",
    "    metrics_goe += r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrr}\n",
      "\\toprule\n",
      "{} &  AUROC &  AUPR-IN &  AUPR-OUT &  FPR95TPR \\\\\n",
      "Method  &        &          &           &           \\\\\n",
      "\\midrule\n",
      "Energy  &  92.68 &    89.72 &     94.48 &     26.26 \\\\\n",
      "Softmax &  91.61 &    89.05 &     93.33 &     33.43 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "data = pd.DataFrame(metrics_goe)\n",
    "tmp = data[data[\"Epoch\"] == n_epochs -1]\n",
    "print((tmp.groupby([\"Method\"]).mean() * 100)[[\"AUROC\", \"AUPR-IN\", \"AUPR-OUT\", \"FPR95TPR\"]]\n",
    "      .to_latex(float_format=\"%.2f\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Epoch', ylabel='AUROC'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sb.lineplot(data=data, y=\"AUROC\", x=\"Epoch\", style=\"Method\", hue=\"Dataset\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `AUROC` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `PrecisionRecallCurve` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: `pos_label` automatically set 1.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/functional/classification/precision_recall_curve.py:156: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
      "  thresholds = tensor(reversed(thresholds[sl]))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:37<00:00,  3.23it/s, loss=1.78]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.64]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.54]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.57]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.55]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.57]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.53]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.54]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 8\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.44]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 9\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.51]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.49]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 11\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.45]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.48]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.5]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 14\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.47]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.48]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 16\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.52]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 17\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.49]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.48]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.51]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:37<00:00,  3.23it/s, loss=1.45]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.45]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:37<00:00,  3.23it/s, loss=1.46]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 23\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.45]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 24\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 508/508 [02:36<00:00,  3.24it/s, loss=1.53]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Do it with uniformely distributed outlier samples \n",
    "\"\"\"\n",
    "from pytorch_ood.model import WideResNet\n",
    "from pytorch_ood.detector import EnergyBased, MaxSoftmax\n",
    "from pytorch_ood.utils import OODMetrics\n",
    "from pytorch_ood.dataset.img import Textures\n",
    "from torchvision.datasets import CIFAR100 \n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "from torchvision.transforms import Compose, ToPILImage, Normalize, ToTensor, Resize, ToPILImage\n",
    "import torch \n",
    "\n",
    "from pytorch_ood.loss import OutlierExposureLoss\n",
    "from torch.optim import SGD\n",
    "from tqdm import tqdm\n",
    "\n",
    "torch.manual_seed(10)\n",
    "\n",
    "\n",
    "# TODO: use 80,000?\n",
    "x = torch.randint(0, 255, size=(80000, 3, 32, 32), dtype=torch.uint8) #  torch.load(\"samples.pt\")[:80000]\n",
    "gan_oe_data = torch.stack([prep(t) for t in x])\n",
    "gan_oe_label = torch.ones(size=(len(x),)).long() * -1\n",
    "\n",
    "data_in_train = CIFAR100(root=\"data\", download=True, transform=mt, target_transform=torch.tensor)\n",
    "data_out_train = TensorDataset(gan_oe_data,gan_oe_label)\n",
    "train_loader = DataLoader(data_in_train + data_out_train, batch_size=256, shuffle=True, num_workers=20)\n",
    "      \n",
    "# create Neural Network\n",
    "model = WideResNet(pretrained=\"cifar100-pt\", num_classes=100).cuda()\n",
    "opti = SGD(model.parameters(), lr=0.001)\n",
    "\n",
    "criterion = OutlierExposureLoss(alpha=0.5)\n",
    "\n",
    "mav = 0.0\n",
    "mavs = []\n",
    "\n",
    "metrics_noise = []\n",
    "\n",
    "\n",
    "# initial test round \n",
    "r = test(model)\n",
    "for entry in r:\n",
    "    entry.update({\"Epoch\": -1})\n",
    "metrics_noise += r\n",
    "\n",
    "\n",
    "for i in range(n_epochs):\n",
    "    print(f\"Epoch {i}\")\n",
    "    \n",
    "    # train \n",
    "    model.train()\n",
    "    bar = tqdm(train_loader)\n",
    "    for x, y in bar:\n",
    "        opti.zero_grad()\n",
    "        y_hat = model(x.cuda())\n",
    "        loss = criterion(y_hat, y.cuda())\n",
    "        mav = 0.2 * loss.item() + 0.8 * mav\n",
    "        mavs.append(mav)\n",
    "\n",
    "        loss.backward()\n",
    "        opti.step()\n",
    "        bar.set_postfix({\"loss\": mav})\n",
    "        \n",
    "    print(80*\"#\")\n",
    "    \n",
    "    r = test(model)\n",
    "    for entry in r:\n",
    "        entry.update({\"Epoch\": i})\n",
    "    \n",
    "    metrics_noise += r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrr}\n",
      "\\toprule\n",
      "{} &  AUROC &  AUPR-IN &  AUPR-OUT &  FPR95TPR \\\\\n",
      "Method  &        &          &           &           \\\\\n",
      "\\midrule\n",
      "Energy  &  91.57 &    89.20 &     93.17 &     33.05 \\\\\n",
      "Softmax &  80.75 &    76.22 &     84.69 &     53.76 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "data = pd.DataFrame(metrics_noise)\n",
    "tmp = data[data[\"Epoch\"] == n_epochs - 1]\n",
    "print((tmp.groupby([\"Method\"]).mean() * 100)[[\"AUROC\", \"AUPR-IN\", \"AUPR-OUT\", \"FPR95TPR\"]]\n",
    "      .to_latex(float_format=\"%.2f\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Epoch', ylabel='AUROC'>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sb.lineplot(data=data, y=\"AUROC\", x=\"Epoch\", style=\"Method\", hue=\"Dataset\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `AUROC` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `PrecisionRecallCurve` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/utilities/prints.py:36: UserWarning: `pos_label` automatically set 1.\n",
      "  warnings.warn(*args, **kwargs)\n",
      "/home/kirchheim/anaconda3/envs/myenv/lib/python3.8/site-packages/torchmetrics/functional/classification/precision_recall_curve.py:156: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
      "  thresholds = tensor(reversed(thresholds[sl]))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.85]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.52]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.39]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.35]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.33]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.28]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.29]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:56<00:00,  3.29it/s, loss=2.31]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 8\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.26]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 9\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.30it/s, loss=2.27]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.30it/s, loss=2.27]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 11\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.27]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.30it/s, loss=2.28]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.31]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 14\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.27]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.26]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 16\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:56<00:00,  3.29it/s, loss=2.25]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 17\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:56<00:00,  3.29it/s, loss=2.25]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.30it/s, loss=2.25]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.23]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.24]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.25]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.30it/s, loss=2.23]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 23\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:54<00:00,  3.30it/s, loss=2.22]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 24\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████| 1368/1368 [06:55<00:00,  3.29it/s, loss=2.25]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Do it with the usual (non-synthetic outliers)\n",
    "\"\"\"\n",
    "from pytorch_ood.model import WideResNet\n",
    "from pytorch_ood.detector import EnergyBased, MaxSoftmax\n",
    "from pytorch_ood.utils import OODMetrics\n",
    "from pytorch_ood.dataset.img import Textures, TinyImages300k\n",
    "from torchvision.datasets import CIFAR100 \n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "from torchvision.transforms import Compose, ToPILImage, Normalize, ToTensor, Resize, ToPILImage\n",
    "import torch \n",
    "\n",
    "from pytorch_ood.loss import OutlierExposureLoss\n",
    "from torch.optim import SGD\n",
    "from tqdm import tqdm\n",
    "\n",
    "torch.manual_seed(10)\n",
    "\n",
    "\n",
    "data_out_train = TinyImages300k(root=\"/data_fast/kirchheim/datasets/\", transform=mt, download=True, \n",
    "                                target_transform=ToUnknown())\n",
    "\n",
    "data_in_train = CIFAR100(root=\"data/\", download=True, transform=mt)\n",
    "train_loader = DataLoader(data_in_train + data_out_train, batch_size=256, shuffle=True, num_workers=20)\n",
    "      \n",
    "# create Neural Network\n",
    "model = WideResNet(pretrained=\"cifar100-pt\", num_classes=100).cuda()\n",
    "opti = SGD(model.parameters(), lr=0.001)\n",
    "\n",
    "criterion = OutlierExposureLoss(alpha=0.5)\n",
    "\n",
    "mav = 0.0\n",
    "mavs = []\n",
    "\n",
    "metrics_oe = []\n",
    "\n",
    "\n",
    "# initial test round \n",
    "r = test(model)\n",
    "for entry in r:\n",
    "    entry.update({\"Epoch\": -1})\n",
    "metrics_oe += r\n",
    "\n",
    "\n",
    "for i in range(n_epochs):\n",
    "    print(f\"Epoch {i}\")\n",
    "    \n",
    "    # train \n",
    "    model.train()\n",
    "    bar = tqdm(train_loader)\n",
    "    for x, y in bar:\n",
    "        opti.zero_grad()\n",
    "        y_hat = model(x.cuda())\n",
    "        loss = criterion(y_hat, y.cuda())\n",
    "        mav = 0.2 * loss.item() + 0.8 * mav\n",
    "        mavs.append(mav)\n",
    "\n",
    "        loss.backward()\n",
    "        opti.step()\n",
    "        bar.set_postfix({\"loss\": mav})\n",
    "        \n",
    "    print(80*\"#\")\n",
    "    \n",
    "    r = test(model)\n",
    "    for entry in r:\n",
    "        entry.update({\"Epoch\": i})\n",
    "    \n",
    "    metrics_oe += r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{lrrrr}\n",
      "\\toprule\n",
      "{} &  AUROC &  AUPR-IN &  AUPR-OUT &  FPR95TPR \\\\\n",
      "Method  &        &          &           &           \\\\\n",
      "\\midrule\n",
      "Energy  &  86.29 &    78.60 &     89.69 &     37.18 \\\\\n",
      "Softmax &  87.13 &    81.49 &     89.97 &     37.91 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "data = pd.DataFrame(metrics_oe)\n",
    "tmp = data[data[\"Epoch\"] == n_epochs -1]\n",
    "print((tmp.groupby([\"Method\"]).mean() * 100)[[\"AUROC\", \"AUPR-IN\", \"AUPR-OUT\", \"FPR95TPR\"]]\n",
    "      .to_latex(float_format=\"%.2f\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Epoch', ylabel='AUROC'>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sb.lineplot(data=data, x=\"Epoch\", y=\"AUROC\", style=\"Method\", hue=\"Dataset\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200.68x360 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import rcParams\n",
    "rcParams['pgf.texsystem'] = \"pdflatex\"\n",
    "\n",
    "sb.set_style(\"whitegrid\")\n",
    "sb.set_context(\"paper\")\n",
    "\n",
    "data_oe = pd.DataFrame(metrics_oe)\n",
    "data_noise = pd.DataFrame(metrics_noise)\n",
    "data_goe = pd.DataFrame(metrics_goe)\n",
    "\n",
    "data_oe[\"Outliers\"] = \"Tiny300k\"\n",
    "data_noise[\"Outliers\"] = \"Noise\"\n",
    "data_goe[\"Outliers\"] = \"Generated\"\n",
    "\n",
    "tmp = pd.concat([data_oe, data_noise, data_goe])\n",
    "tmp[\"Epoch\"] += 1 \n",
    "g = sb.relplot(data=tmp, col=\"Outliers\",  x=\"Epoch\", y=\"AUROC\", style=\"Method\", hue=\"Dataset\", kind=\"line\")\n",
    "# plt.legend()\n",
    "leg = g.legend\n",
    "leg.set_bbox_to_anchor([0.95,0.35]) \n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(f\"img/c100-auroc-over-time.png\", dpi=300)\n",
    "plt.savefig(f\"img/c100-auroc-over-time.pgf\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([500000, 3, 32, 32])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = np.load(f\"samples-c100-large.npz\")\n",
    "x = torch.tensor(data[\"x\"])\n",
    "x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100\n",
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n",
      "Epoch 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.199]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.12it/s, loss=0.158]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.12it/s, loss=0.137]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.139]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.13it/s, loss=0.131]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.134]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.132]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.143]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 8\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.122]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 9\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.16it/s, loss=0.127]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.16it/s, loss=0.123]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 11\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.16it/s, loss=0.122]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.135]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.16it/s, loss=0.119]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 14\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.15it/s, loss=0.131]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.16it/s, loss=0.127]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 16\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.11it/s, loss=0.125]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 17\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.12it/s, loss=0.119]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.12it/s, loss=0.116]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.13it/s, loss=0.111]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.12it/s, loss=0.112]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.13it/s, loss=0.117]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.13it/s, loss=0.137]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 23\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.14it/s, loss=0.121]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 24\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:02<00:00,  3.13it/s, loss=0.109]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "1000\n",
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n",
      "Epoch 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 197/197 [01:01<00:00,  3.19it/s, loss=0.54]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.16it/s, loss=0.693]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.502]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.621]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.512]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.242]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.18it/s, loss=0.395]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 7\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.414]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 8\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.392]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 9\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.368]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.18it/s, loss=0.51]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 11\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.186]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 12\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.328]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 13\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.498]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 14\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.535]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 15\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.344]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 16\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.394]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 17\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.43]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 18\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.364]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 19\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.13it/s, loss=0.503]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.349]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 21\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.16it/s, loss=0.472]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 22\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.438]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 23\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.42]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 24\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 197/197 [01:02<00:00,  3.17it/s, loss=0.61]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "10000\n",
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n",
      "Epoch 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:01<00:00,  3.17it/s, loss=0.208]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:01<00:00,  3.17it/s, loss=0.172]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████| 196/196 [01:01<00:00,  3.18it/s, loss=0.15]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 3\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:01<00:00,  3.18it/s, loss=0.121]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:01<00:00,  3.17it/s, loss=0.128]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n",
      "Files already downloaded and verified\n",
      "Epoch 5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████| 196/196 [01:01<00:00,  3.18it/s, loss=0.136]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "################################################################################\n"
     ]
    }
   ],
   "source": [
    "torch.manual_seed(10)\n",
    "\n",
    "# TODO: use 80,000?\n",
    "# data = np.load(f\"samples-c100-large.npz\")\n",
    "\n",
    "metrics_l = []\n",
    "    \n",
    "for l in [100, 1000, 10000, 20000, 50000, 100000]:\n",
    "    print(l)\n",
    "    xn = x[:l]\n",
    "    gan_oe_data = torch.stack([prep(t) for t in xn])\n",
    "    gan_oe_label = torch.ones(size=(len(xn),)).long() * -1\n",
    "\n",
    "    data_in_train = CIFAR100(root=\"data/\", download=True, transform=mt, target_transform=torch.tensor)\n",
    "    data_out_train = TensorDataset(gan_oe_data, gan_oe_label)\n",
    "    train_loader = DataLoader(data_in_train + data_out_train, batch_size=256, shuffle=True, num_workers=20)\n",
    "\n",
    "    # create Neural Network\n",
    "    model = WideResNet(pretrained=\"cifar100-pt\", num_classes=100).cuda()\n",
    "    opti = SGD(model.parameters(), lr=0.001)\n",
    "\n",
    "    criterion = OutlierExposureLoss(alpha=0.5)\n",
    "\n",
    "    mav = 0.0\n",
    "    mavs = []\n",
    "\n",
    "    r = test(model)\n",
    "    for entry in r:\n",
    "        entry.update({\n",
    "            \"Epoch\": -1,\n",
    "            \"Length\": l\n",
    "        })\n",
    "\n",
    "    metrics_l += r\n",
    "        \n",
    "\n",
    "    for i in range(n_epochs):\n",
    "        print(f\"Epoch {i}\")\n",
    "\n",
    "        # train \n",
    "        model.train()\n",
    "        bar = tqdm(train_loader)\n",
    "        for x, y in bar:\n",
    "            opti.zero_grad()\n",
    "            y_hat = model(x.cuda())\n",
    "            loss = criterion(y_hat, y.cuda())\n",
    "            mav = 0.2 * loss.item() + 0.8 * mav\n",
    "            mavs.append(mav)\n",
    "\n",
    "            loss.backward()\n",
    "            opti.step()\n",
    "            bar.set_postfix({\"loss\": mav})\n",
    "\n",
    "        print(80*\"#\")\n",
    "\n",
    "        r = test(model)\n",
    "        for entry in r:\n",
    "            entry.update({\n",
    "                \"Epoch\": i,\n",
    "                \"Length\": l\n",
    "            })\n",
    "\n",
    "        metrics_l += r"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp = pd.DataFrame(metrics_l)\n",
    "tmp[\"Samples\"] = tmp[\"Length\"]\n",
    "tmp[\"Epoch\"] = tmp[\"Epoch\"] + 1 \n",
    "tmp = tmp.groupby([\"Epoch\", \"Samples\", \"Method\"]).mean()\n",
    "\n",
    "sb.lineplot(data=tmp,  x=\"Epoch\", y=\"AUROC\", hue=\"Samples\", style=\"Method\") \n",
    "plt.tight_layout()\n",
    "plt.savefig(\"img/size-vs-auroc-c100.png\", dpi=300)\n",
    "plt.savefig(\"img/size-vs-auroc-c100.pgf\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "myenv",
   "language": "python",
   "name": "myenv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}