barlow_twins.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from torch import nn\n",
    "from torch import optim\n",
    "import torch.nn.functional as F\n",
    "from torch.utils.data import DataLoader\n",
    "import torchvision.models as models\n",
    "\n",
    "\n",
    "from torchvision import models\n",
    "import torchvision.transforms as T\n",
    "from torchvision.datasets import CIFAR10\n",
    "\n",
    "import torchutils as tu\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from torch.utils.tensorboard import SummaryWriter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "DEVICE = 'cpu'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "Ya_transforms = T.Compose(\n",
    "    [\n",
    "        T.Resize((32, 32)),\n",
    "        T.ToTensor(),\n",
    "        T.RandomChoice(\n",
    "            [\n",
    "                T.ColorJitter(),\n",
    "                T.RandomHorizontalFlip(),\n",
    "                T.GaussianBlur(3),\n",
    "                T.RandomSolarize(.6)\n",
    "            ]\n",
    "        )\n",
    "    ]\n",
    ")\n",
    "\n",
    "Yb_transforms = T.Compose(\n",
    "    [\n",
    "        T.Resize((32, 32)),\n",
    "        T.ToTensor(),\n",
    "        T.RandomChoice(\n",
    "            [\n",
    "                T.ColorJitter(),\n",
    "                T.RandomHorizontalFlip(),\n",
    "                T.GaussianBlur(9),\n",
    "                T.RandomSolarize(.3)\n",
    "            ]\n",
    "        )\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "A_dataset = CIFAR10(\n",
    "    'data',\n",
    "    train = 'True',\n",
    "    download = True,\n",
    "    transform=Ya_transforms\n",
    ")\n",
    "\n",
    "\n",
    "B_dataset = CIFAR10(\n",
    "    'data',\n",
    "    train = 'True',\n",
    "    download = True,\n",
    "    transform=Yb_transforms\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "BATCH_SIZE=256\n",
    "A_loader = DataLoader(A_dataset, batch_size=BATCH_SIZE)\n",
    "B_loader = DataLoader(B_dataset, batch_size=BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x14e225070>"
      ]
     },
     "execution_count": 7,
     "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": [
    "n = 8\n",
    "a_example = next(iter(A_loader))[0][:n].view(3, 32*n, 32).permute(1, 2, 0)\n",
    "b_example = next(iter(B_loader))[0][:n].view(3, 32*n, 32).permute(1, 2, 0)\n",
    "\n",
    "plt.imshow(torch.cat([a_example, b_example], 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# class Encoder(nn.Module):\n",
    "#     def __init__(self) -> None:\n",
    "#         super().__init__()\n",
    "\n",
    "#         self.conv1 = nn.Sequential(\n",
    "#             nn.Conv2d(1, 32, 4, 1),\n",
    "#             nn.Dropout(),\n",
    "#             nn.ReLU(),\n",
    "#             nn.BatchNorm2d(32)\n",
    "#         )\n",
    "\n",
    "#         self.conv2 = nn.Sequential(\n",
    "#             nn.Conv2d(32, 16, 6, 2),\n",
    "#             nn.Dropout(),\n",
    "#             nn.ReLU(),\n",
    "#             nn.BatchNorm2d(16)\n",
    "#         )\n",
    "\n",
    "#         self.linear = nn.Sequential(\n",
    "#             nn.LazyLinear(16)\n",
    "#         )\n",
    "\n",
    "#     def forward(self, x):\n",
    "#         x = self.conv1(x)\n",
    "#         x = self.conv2(x)\n",
    "#         x = x.view(x.size(0), -1)\n",
    "#         out = self.linear(x)\n",
    "#         return out\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/paveldumin/miniforge3/envs/pt/lib/python3.9/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "resnet = models.resnet18(weights=models.ResNet18_Weights)\n",
    "encoder = torch.nn.Sequential(*(list(resnet.children())[:-1]))\n",
    "for i in encoder.parameters():\n",
    "    i.requires_grad = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Projector(nn.Module):\n",
    "    def __init__(self) -> None:\n",
    "        super().__init__()\n",
    "\n",
    "        self.proj = nn.Sequential(\n",
    "            nn.Linear(in_features=512, out_features=32),\n",
    "            nn.BatchNorm1d(32),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(32, 16)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.proj(x.view(x.size(0), -1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "class F_theta(nn.Module):\n",
    "    def __init__(self, encoder: torch.nn.Module) -> None:\n",
    "        super().__init__()\n",
    "\n",
    "        self.encoder = encoder\n",
    "        self.projector = Projector()\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.projector(self.encoder(x))\n",
    "model = F_theta(encoder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([256, 512])\n"
     ]
    }
   ],
   "source": [
    "out = encoder(torch.randn(BATCH_SIZE, 3, 32, 32))\n",
    "print(out.view(out.shape[0], -1).shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "========================================================================================================\n",
      "Layer                                          Kernel             Output          Params           FLOPs\n",
      "========================================================================================================\n",
      "0_encoder.Conv2d_0                           [3, 64, 7, 7]   [256, 64, 16, 16]       9,408   616,562,688\n",
      "1_encoder.BatchNorm2d_1                               [64]   [256, 64, 16, 16]         128    16,777,216\n",
      "2_encoder.ReLU_2                                         -   [256, 64, 16, 16]           0             0\n",
      "3_encoder.MaxPool2d_3                                    -     [256, 64, 8, 8]           0             0\n",
      "4_encoder.4.0.Conv2d_conv1                  [64, 64, 3, 3]     [256, 64, 8, 8]      36,864   603,979,776\n",
      "5_encoder.4.0.BatchNorm2d_bn1                         [64]     [256, 64, 8, 8]         128     4,194,304\n",
      "6_encoder.4.0.ReLU_relu                                  -     [256, 64, 8, 8]           0             0\n",
      "7_encoder.4.0.Conv2d_conv2                  [64, 64, 3, 3]     [256, 64, 8, 8]      36,864   603,979,776\n",
      "8_encoder.4.0.BatchNorm2d_bn2                         [64]     [256, 64, 8, 8]         128     4,194,304\n",
      "9_encoder.4.0.ReLU_relu                                  -     [256, 64, 8, 8]           0             0\n",
      "10_encoder.4.1.Conv2d_conv1                 [64, 64, 3, 3]     [256, 64, 8, 8]      36,864   603,979,776\n",
      "11_encoder.4.1.BatchNorm2d_bn1                        [64]     [256, 64, 8, 8]         128     4,194,304\n",
      "12_encoder.4.1.ReLU_relu                                 -     [256, 64, 8, 8]           0             0\n",
      "13_encoder.4.1.Conv2d_conv2                 [64, 64, 3, 3]     [256, 64, 8, 8]      36,864   603,979,776\n",
      "14_encoder.4.1.BatchNorm2d_bn2                        [64]     [256, 64, 8, 8]         128     4,194,304\n",
      "15_encoder.4.1.ReLU_relu                                 -     [256, 64, 8, 8]           0             0\n",
      "16_encoder.5.0.Conv2d_conv1                [64, 128, 3, 3]    [256, 128, 4, 4]      73,728   301,989,888\n",
      "17_encoder.5.0.BatchNorm2d_bn1                       [128]    [256, 128, 4, 4]         256     2,097,152\n",
      "18_encoder.5.0.ReLU_relu                                 -    [256, 128, 4, 4]           0             0\n",
      "19_encoder.5.0.Conv2d_conv2               [128, 128, 3, 3]    [256, 128, 4, 4]     147,456   603,979,776\n",
      "20_encoder.5.0.BatchNorm2d_bn2                       [128]    [256, 128, 4, 4]         256     2,097,152\n",
      "21_encoder.5.0.downsample.Conv2d_0         [64, 128, 1, 1]    [256, 128, 4, 4]       8,192    33,554,432\n",
      "22_encoder.5.0.downsample.BatchNorm2d_1              [128]    [256, 128, 4, 4]         256     2,097,152\n",
      "23_encoder.5.0.ReLU_relu                                 -    [256, 128, 4, 4]           0             0\n",
      "24_encoder.5.1.Conv2d_conv1               [128, 128, 3, 3]    [256, 128, 4, 4]     147,456   603,979,776\n",
      "25_encoder.5.1.BatchNorm2d_bn1                       [128]    [256, 128, 4, 4]         256     2,097,152\n",
      "26_encoder.5.1.ReLU_relu                                 -    [256, 128, 4, 4]           0             0\n",
      "27_encoder.5.1.Conv2d_conv2               [128, 128, 3, 3]    [256, 128, 4, 4]     147,456   603,979,776\n",
      "28_encoder.5.1.BatchNorm2d_bn2                       [128]    [256, 128, 4, 4]         256     2,097,152\n",
      "29_encoder.5.1.ReLU_relu                                 -    [256, 128, 4, 4]           0             0\n",
      "30_encoder.6.0.Conv2d_conv1               [128, 256, 3, 3]    [256, 256, 2, 2]     294,912   301,989,888\n",
      "31_encoder.6.0.BatchNorm2d_bn1                       [256]    [256, 256, 2, 2]         512     1,048,576\n",
      "32_encoder.6.0.ReLU_relu                                 -    [256, 256, 2, 2]           0             0\n",
      "33_encoder.6.0.Conv2d_conv2               [256, 256, 3, 3]    [256, 256, 2, 2]     589,824   603,979,776\n",
      "34_encoder.6.0.BatchNorm2d_bn2                       [256]    [256, 256, 2, 2]         512     1,048,576\n",
      "35_encoder.6.0.downsample.Conv2d_0        [128, 256, 1, 1]    [256, 256, 2, 2]      32,768    33,554,432\n",
      "36_encoder.6.0.downsample.BatchNorm2d_1              [256]    [256, 256, 2, 2]         512     1,048,576\n",
      "37_encoder.6.0.ReLU_relu                                 -    [256, 256, 2, 2]           0             0\n",
      "38_encoder.6.1.Conv2d_conv1               [256, 256, 3, 3]    [256, 256, 2, 2]     589,824   603,979,776\n",
      "39_encoder.6.1.BatchNorm2d_bn1                       [256]    [256, 256, 2, 2]         512     1,048,576\n",
      "40_encoder.6.1.ReLU_relu                                 -    [256, 256, 2, 2]           0             0\n",
      "41_encoder.6.1.Conv2d_conv2               [256, 256, 3, 3]    [256, 256, 2, 2]     589,824   603,979,776\n",
      "42_encoder.6.1.BatchNorm2d_bn2                       [256]    [256, 256, 2, 2]         512     1,048,576\n",
      "43_encoder.6.1.ReLU_relu                                 -    [256, 256, 2, 2]           0             0\n",
      "44_encoder.7.0.Conv2d_conv1               [256, 512, 3, 3]    [256, 512, 1, 1]   1,179,648   301,989,888\n",
      "45_encoder.7.0.BatchNorm2d_bn1                       [512]    [256, 512, 1, 1]       1,024       524,288\n",
      "46_encoder.7.0.ReLU_relu                                 -    [256, 512, 1, 1]           0             0\n",
      "47_encoder.7.0.Conv2d_conv2               [512, 512, 3, 3]    [256, 512, 1, 1]   2,359,296   603,979,776\n",
      "48_encoder.7.0.BatchNorm2d_bn2                       [512]    [256, 512, 1, 1]       1,024       524,288\n",
      "49_encoder.7.0.downsample.Conv2d_0        [256, 512, 1, 1]    [256, 512, 1, 1]     131,072    33,554,432\n",
      "50_encoder.7.0.downsample.BatchNorm2d_1              [512]    [256, 512, 1, 1]       1,024       524,288\n",
      "51_encoder.7.0.ReLU_relu                                 -    [256, 512, 1, 1]           0             0\n",
      "52_encoder.7.1.Conv2d_conv1               [512, 512, 3, 3]    [256, 512, 1, 1]   2,359,296   603,979,776\n",
      "53_encoder.7.1.BatchNorm2d_bn1                       [512]    [256, 512, 1, 1]       1,024       524,288\n",
      "54_encoder.7.1.ReLU_relu                                 -    [256, 512, 1, 1]           0             0\n",
      "55_encoder.7.1.Conv2d_conv2               [512, 512, 3, 3]    [256, 512, 1, 1]   2,359,296   603,979,776\n",
      "56_encoder.7.1.BatchNorm2d_bn2                       [512]    [256, 512, 1, 1]       1,024       524,288\n",
      "57_encoder.7.1.ReLU_relu                                 -    [256, 512, 1, 1]           0             0\n",
      "58_encoder.AdaptiveAvgPool2d_8                           -    [256, 512, 1, 1]           0       262,144\n",
      "59_projector.proj.Linear_0                       [512, 32]           [256, 32]      16,416     8,380,416\n",
      "60_projector.proj.BatchNorm1d_1                       [32]           [256, 32]          64        32,768\n",
      "61_projector.proj.ReLU_2                                 -           [256, 32]           0             0\n",
      "62_projector.proj.Linear_3                        [32, 16]           [256, 16]         528       258,048\n",
      "========================================================================================================\n",
      "Total params: 11,193,520\n",
      "Trainable params: 17,008\n",
      "Non-trainable params: 11,176,512\n",
      "Total FLOPs: 9,535,770,624 / 9.54 GFLOPs\n",
      "--------------------------------------------------------------------------------------------------------\n",
      "Input size (MB): 0.00\n",
      "Forward/backward pass size (MB): 299.22\n",
      "Params size (MB): 42.70\n",
      "Estimated Total Size (MB): 341.92\n",
      "========================================================================================================\n"
     ]
    }
   ],
   "source": [
    "tu.get_model_summary(\n",
    "    model.to(DEVICE), \n",
    "    x=torch.randn(BATCH_SIZE, 3, 32, 32, device=DEVICE)\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalize(x):\n",
    "    return (x - x.mean(0)) / x.std(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = torch.optim.Adam(model.parameters())\n",
    "writer = SummaryWriter()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([256, 16])\n",
      "torch.Size([256, 16])\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[32], line 10\u001b[0m\n\u001b[1;32m      8\u001b[0m out_a \u001b[39m=\u001b[39m model(x_a[\u001b[39m0\u001b[39m]\u001b[39m.\u001b[39mto(DEVICE))\u001b[39m.\u001b[39mview(x_a[\u001b[39m0\u001b[39m]\u001b[39m.\u001b[39msize(\u001b[39m0\u001b[39m), \u001b[39m-\u001b[39m\u001b[39m1\u001b[39m)\n\u001b[1;32m      9\u001b[0m \u001b[39mprint\u001b[39m(out_a\u001b[39m.\u001b[39mshape)\n\u001b[0;32m---> 10\u001b[0m out_b \u001b[39m=\u001b[39m model(x_b[\u001b[39m0\u001b[39;49m]\u001b[39m.\u001b[39;49mto(DEVICE))\u001b[39m.\u001b[39mview(x_b[\u001b[39m0\u001b[39m]\u001b[39m.\u001b[39msize(\u001b[39m0\u001b[39m), \u001b[39m-\u001b[39m\u001b[39m1\u001b[39m)\n\u001b[1;32m     12\u001b[0m out_a_normed \u001b[39m=\u001b[39m normalize(out_a)\n\u001b[1;32m     13\u001b[0m out_b_normed \u001b[39m=\u001b[39m normalize(out_b)\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1502\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
      "Cell \u001b[0;32mIn[26], line 9\u001b[0m, in \u001b[0;36mF_theta.forward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, x):\n\u001b[0;32m----> 9\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mprojector(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mencoder(x))\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1502\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/container.py:217\u001b[0m, in \u001b[0;36mSequential.forward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    215\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39minput\u001b[39m):\n\u001b[1;32m    216\u001b[0m     \u001b[39mfor\u001b[39;00m module \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m:\n\u001b[0;32m--> 217\u001b[0m         \u001b[39minput\u001b[39m \u001b[39m=\u001b[39m module(\u001b[39minput\u001b[39;49m)\n\u001b[1;32m    218\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39minput\u001b[39m\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1502\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/container.py:217\u001b[0m, in \u001b[0;36mSequential.forward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    215\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39minput\u001b[39m):\n\u001b[1;32m    216\u001b[0m     \u001b[39mfor\u001b[39;00m module \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m:\n\u001b[0;32m--> 217\u001b[0m         \u001b[39minput\u001b[39m \u001b[39m=\u001b[39m module(\u001b[39minput\u001b[39;49m)\n\u001b[1;32m    218\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39minput\u001b[39m\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1502\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torchvision/models/resnet.py:92\u001b[0m, in \u001b[0;36mBasicBlock.forward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m     89\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, x: Tensor) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m Tensor:\n\u001b[1;32m     90\u001b[0m     identity \u001b[39m=\u001b[39m x\n\u001b[0;32m---> 92\u001b[0m     out \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mconv1(x)\n\u001b[1;32m     93\u001b[0m     out \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mbn1(out)\n\u001b[1;32m     94\u001b[0m     out \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mrelu(out)\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m   1502\u001b[0m \u001b[39m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[39m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/conv.py:463\u001b[0m, in \u001b[0;36mConv2d.forward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    462\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39minput\u001b[39m: Tensor) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m Tensor:\n\u001b[0;32m--> 463\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_conv_forward(\u001b[39minput\u001b[39;49m, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mweight, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mbias)\n",
      "File \u001b[0;32m~/miniforge3/envs/pt/lib/python3.9/site-packages/torch/nn/modules/conv.py:459\u001b[0m, in \u001b[0;36mConv2d._conv_forward\u001b[0;34m(self, input, weight, bias)\u001b[0m\n\u001b[1;32m    455\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpadding_mode \u001b[39m!=\u001b[39m \u001b[39m'\u001b[39m\u001b[39mzeros\u001b[39m\u001b[39m'\u001b[39m:\n\u001b[1;32m    456\u001b[0m     \u001b[39mreturn\u001b[39;00m F\u001b[39m.\u001b[39mconv2d(F\u001b[39m.\u001b[39mpad(\u001b[39minput\u001b[39m, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_reversed_padding_repeated_twice, mode\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpadding_mode),\n\u001b[1;32m    457\u001b[0m                     weight, bias, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mstride,\n\u001b[1;32m    458\u001b[0m                     _pair(\u001b[39m0\u001b[39m), \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdilation, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mgroups)\n\u001b[0;32m--> 459\u001b[0m \u001b[39mreturn\u001b[39;00m F\u001b[39m.\u001b[39;49mconv2d(\u001b[39minput\u001b[39;49m, weight, bias, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mstride,\n\u001b[1;32m    460\u001b[0m                 \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mpadding, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdilation, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mgroups)\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "model.train()\n",
    "\n",
    "losses = []\n",
    "\n",
    "for epoch in range(10):\n",
    "    batch_losses = []\n",
    "    for batch_idx, (x_a, x_b) in enumerate(zip(A_loader, B_loader)):\n",
    "        out_a = model(x_a[0].to(DEVICE)).view(x_a[0].size(0), -1)\n",
    "        # print(out_a.shape)\n",
    "        out_b = model(x_b[0].to(DEVICE)).view(x_b[0].size(0), -1)\n",
    "\n",
    "        out_a_normed = normalize(out_a)\n",
    "        out_b_normed = normalize(out_b)\n",
    "\n",
    "        c = torch.matmul(out_a_normed.T, out_b_normed) / BATCH_SIZE\n",
    "\n",
    "        c_diff = (c - torch.eye(c.shape[0]).to(DEVICE)).pow(2)\n",
    "\n",
    "        c_diff.flatten()[:-1].view(c_diff.size(0) - 1, c_diff.size(0)+1)[:, 1:].mul_(.5)\n",
    "\n",
    "        loss = c_diff.sum()\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        batch_losses.append(loss.item())\n",
    "        writer.add_scalar('Loss/train', batch_losses[-1], batch_idx)\n",
    "    losses.append(np.mean(batch_losses))\n",
    "\n",
    "    print(f'Epoch n: {epoch}, loss: {losses[-1]}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pt",
   "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": "3.9.15"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "a7df815f77bf8edb0a6a656480e7ee3c360a4b494ba862a12bd6032824eda046"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}