6. Convolutional Neural Network - build mode.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convolutional Neural Network > Recurral Nerual Network in sequential types of data\n",
    "Convolutional Neural network accepts 2D input, 3D inputs\n",
    "convolution Kernal - windows\n",
    "first layer of convolution filters - tend to find things lik edges or curves \n",
    "    looking at combinations of pixels i.e. why it only finds corner or edge\n",
    "next layer - more complex features that edges and curves build \n",
    "    i.e. circle square\n",
    "next layer find combination of cirlces and squares (a bit more complicated)\n",
    "kernal slides window over the entire image and contensing the image\n",
    "after doing convolution - ne condensed version of image\n",
    "Pooling - pooling has a window like nernal\n",
    "Max pooling - inside the kernal takes the maximum value\n",
    "summary - convolutional network - drastically simplifying your image and looks for features of the image.\n",
    "Generally you will have 2-4 layers of the convolutional layers\n",
    "CNN - reducing your image to basic building blocks and then find patterns of those blocks given how many layers you have\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  1%|          | 131/12501 [00:00<00:10, 1194.34it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "directory:  kagglecatsanddogs_3367a/PetImages/Cat\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 12501/12501 [00:09<00:00, 1270.03it/s]\n",
      "  1%|          | 107/12501 [00:00<00:11, 1069.03it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "directory:  kagglecatsanddogs_3367a/PetImages/Dog\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 12501/12501 [00:10<00:00, 1190.38it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cats: 12476\n",
      "Dogs: 12470\n"
     ]
    }
   ],
   "source": [
    "#https://www.youtube.com/watch?v=1gQR24B3ISE&list=PLQVvvaa0QuDdeMyHEYc0gxFpYwHY2Qfdh&index=6\n",
    "import os\n",
    "import cv2\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "\n",
    "#in your pre-processing dataset you dont want to build data everytime you run your code\n",
    "REBUILD_DATA = False\n",
    "\n",
    "class DogsVSCats():\n",
    "    #first, specify the size of the image i.e. 50 by 50\n",
    "    #input images vary in sizes and shape (some are portrait, some are landscape) - we need them to uniform on input\n",
    "    # normalised to the same size\n",
    "    # resizing the images or resize and pad (adding white / black pixels on smaller images with open cv)\n",
    "    #shift over and crop the image, rotate or flip them\n",
    "    IMG_SIZE = 50\n",
    "    CATS = \"kagglecatsanddogs_3367a/PetImages/Cat\"\n",
    "    DOGS = \"kagglecatsanddogs_3367a/PetImages/Dog\"\n",
    "    #Pre-labelled data\n",
    "    LABELS ={CATS: 0, DOGS: 1}\n",
    "    # massive list of images with their label\n",
    "    training_data = []\n",
    "    cat_count = 0\n",
    "    dog_count = 0\n",
    "        \n",
    "    def make_training_data(self):\n",
    "        for label in self.LABELS:\n",
    "            print(\"directory: \",label)\n",
    "            #looping all the images in the directory\n",
    "            for f in tqdm(os.listdir(label)):\n",
    "                try:\n",
    "                    path = os.path.join(label,f)\n",
    "                    img = cv2.imread(path,cv2.IMREAD_GRAYSCALE)\n",
    "                    img = cv2.resize(img, (self.IMG_SIZE, self.IMG_SIZE))\n",
    "                    self.training_data.append([np.array(img), np.eye(2)[self.LABELS[label]]])\n",
    "                    if label == self.CATS:\n",
    "                        self.cat_count += 1\n",
    "                    elif label == self.DOGS:\n",
    "                        self.dog_count += 1\n",
    "                except Exception as e:\n",
    "                    pass\n",
    "                    #print(str(e))                    \n",
    "        np.random.shuffle(self.training_data)\n",
    "        np.save(\"training_data.npy\", self.training_data)\n",
    "        print(\"Cats:\", self.cat_count)\n",
    "        print(\"Dogs:\", self.dog_count)\n",
    "\n",
    "if REBUILD_DATA:\n",
    "    dogsvcats = DogsVSCats()\n",
    "    dogsvcats.make_training_data()\n",
    "            \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "24946\n"
     ]
    }
   ],
   "source": [
    "training_data = np.load(\"training_data.npy\", allow_pickle=True)\n",
    "print(len(training_data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[array([[ 90,  88,  95, ..., 100, 114, 158],\n",
      "       [ 95,  89,  98, ..., 106, 117, 166],\n",
      "       [ 95,  98,  98, ..., 102, 121, 177],\n",
      "       ...,\n",
      "       [ 49,  50,  51, ..., 106, 174, 177],\n",
      "       [ 39,  52,  50, ..., 104, 175, 177],\n",
      "       [ 74,  76,  69, ..., 108, 180, 177]], dtype=uint8)\n",
      " array([1., 0.])]\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "index = 16\n",
    "plt.imshow(training_data[index][0], cmap= \"gray\")\n",
    "plt.show()\n",
    "print(training_data[index])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class Net(nn.Module):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        # third parameter - kernal - 5 by 5 \n",
    "        #2d convolutional layers used for images\n",
    "        #3d is used on scans or models\n",
    "        #1d - sequential or temporal datasets i.e. attention seems to outperform Recurral NN which is the industry standard for sequential type of data\n",
    "        #input = 1, output = 32 convolutional features \n",
    "        self.conv1 = nn.Conv2d(1,32,5)\n",
    "        self.conv2 = nn.Conv2d(32,64,5)\n",
    "        self.conv3 = nn.Conv2d(64,128,5)\n",
    "        #complicated to convert convolutional layer to linear layer\n",
    "        #problem - don't know the input dimention of the linear layer\n",
    "        # make output 512 dense layer\n",
    "        #self.fc1 = nn.Linear ( ??? , 512)\n",
    "        \n",
    "        #running through random data into the convolution layers to get the shape\n",
    "        #multiply the dimentions i.e. 5 by 5 by 3 = 75 -> flattened shape\n",
    "        #1st parameter - how ever amount of feature sets we have\n",
    "        #2nd parameter - however many amount of images -> related to the first parameter of the first convolution layer\n",
    "        # tensor form 1,50,50\n",
    "        x = torch.randn(50,50).view(-1,1,50,50)\n",
    "        self._to_linear = None\n",
    "        self.convs(x)\n",
    "        \n",
    "        # self.fc1 = nn.Linear ( ??? , 512)\n",
    "        #two classes\n",
    "        self.fc1 = nn.Linear(self._to_linear,512)\n",
    "        self.fc2 = nn.Linear(512,2)\n",
    "    def convs(self, x):\n",
    "        #just like the forward method but only applying to the convolutional layers\n",
    "        # Max pooling - sample-based discretization process. The objective is to down-sample an input representation (image, hidden-layer output matrix, etc.), reducing its dimensionality and allowing for assumptions to be made about features contained in the sub-regions binned.\n",
    "        # This is done to in part to help over-fitting by providing an abstracted form of the representation. As well, it reduces the computational cost by reducing the number of parameters to learn and provides basic translation invariance to the internal representation. Max pooling is done by applying a max filter to (usually) non-overlapping subregions of the initial representation.\n",
    "        # activation function of a node defines the output of that node given an input or set of inputs. \n",
    "        # Recified Linear unit - type of activation function  y = max(0, x)\n",
    "        #https://www.tinymind.com/learn/terms/relu\n",
    "        # (2,2) shape of the pooling\n",
    "        x = F.max_pool2d(F.relu(self.conv1(x)), (2,2))\n",
    "        x = F.max_pool2d(F.relu(self.conv2(x)), (2,2))\n",
    "        x = F.max_pool2d(F.relu(self.conv3(x)), (2,2))\n",
    "        \n",
    "        print(x[0].shape)\n",
    "        if self._to_linear is None:\n",
    "            self._to_linear = x[0].shape[0]*x[0].shape[1]*x[0].shape[2]\n",
    "        return x\n",
    "    def forward (self,x):\n",
    "        #required to know the flattened result after initializing during real forward pass\n",
    "        x = self.convs(x)\n",
    "        x = x.view(-1, self._to_linear)\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = self.fc2(x)\n",
    "        #returning an activation layer rather than just x\n",
    "        return F.softmax(x, dim = 1)\n",
    "    \n",
    "net = Net()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2494\n"
     ]
    }
   ],
   "source": [
    "import torch.optim as optim\n",
    "\n",
    "# All the parameters in the nn is controlled by the optimiser\n",
    "optimizer = optim.Adam(net.parameters(), lr=0.001)\n",
    "loss_function = nn.MSELoss()\n",
    "\n",
    "X = torch.Tensor([i[0] for i in training_data]).view(-1,50,50)\n",
    "#Scaling the imagery - between 0 - 1 from 0 - 255\n",
    "X = X/255.0\n",
    "y = torch.Tensor([i[1] for i in training_data])\n",
    "\n",
    "VAL_PCT = 0.1\n",
    "val_size = int(len(X)*VAL_PCT)\n",
    "print(val_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train_X is :  22452 with train_y :  22452\n",
      "test_X is :  2494 with test_y :  2494\n"
     ]
    }
   ],
   "source": [
    "train_X = X[: -val_size]\n",
    "train_y = y[: -val_size]\n",
    "\n",
    "test_X = X[-val_size:]\n",
    "test_y = y[-val_size:]\n",
    "\n",
    "print(\"train_X is : \",len(train_X), \"with train_y : \", len(train_y))\n",
    "print(\"test_X is : \",len(test_X), \"with test_y : \", len(test_y))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 225/225 [00:00<00:00, 37465.50it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 100\n",
      "100 200\n",
      "200 300\n",
      "300 400\n",
      "400 500\n",
      "500 600\n",
      "600 700\n",
      "700 800\n",
      "800 900\n",
      "900 1000\n",
      "1000 1100\n",
      "1100 1200\n",
      "1200 1300\n",
      "1300 1400\n",
      "1400 1500\n",
      "1500 1600\n",
      "1600 1700\n",
      "1700 1800\n",
      "1800 1900\n",
      "1900 2000\n",
      "2000 2100\n",
      "2100 2200\n",
      "2200 2300\n",
      "2300 2400\n",
      "2400 2500\n",
      "2500 2600\n",
      "2600 2700\n",
      "2700 2800\n",
      "2800 2900\n",
      "2900 3000\n",
      "3000 3100\n",
      "3100 3200\n",
      "3200 3300\n",
      "3300 3400\n",
      "3400 3500\n",
      "3500 3600\n",
      "3600 3700\n",
      "3700 3800\n",
      "3800 3900\n",
      "3900 4000\n",
      "4000 4100\n",
      "4100 4200\n",
      "4200 4300\n",
      "4300 4400\n",
      "4400 4500\n",
      "4500 4600\n",
      "4600 4700\n",
      "4700 4800\n",
      "4800 4900\n",
      "4900 5000\n",
      "5000 5100\n",
      "5100 5200\n",
      "5200 5300\n",
      "5300 5400\n",
      "5400 5500\n",
      "5500 5600\n",
      "5600 5700\n",
      "5700 5800\n",
      "5800 5900\n",
      "5900 6000\n",
      "6000 6100\n",
      "6100 6200\n",
      "6200 6300\n",
      "6300 6400\n",
      "6400 6500\n",
      "6500 6600\n",
      "6600 6700\n",
      "6700 6800\n",
      "6800 6900\n",
      "6900 7000\n",
      "7000 7100\n",
      "7100 7200\n",
      "7200 7300\n",
      "7300 7400\n",
      "7400 7500\n",
      "7500 7600\n",
      "7600 7700\n",
      "7700 7800\n",
      "7800 7900\n",
      "7900 8000\n",
      "8000 8100\n",
      "8100 8200\n",
      "8200 8300\n",
      "8300 8400\n",
      "8400 8500\n",
      "8500 8600\n",
      "8600 8700\n",
      "8700 8800\n",
      "8800 8900\n",
      "8900 9000\n",
      "9000 9100\n",
      "9100 9200\n",
      "9200 9300\n",
      "9300 9400\n",
      "9400 9500\n",
      "9500 9600\n",
      "9600 9700\n",
      "9700 9800\n",
      "9800 9900\n",
      "9900 10000\n",
      "10000 10100\n",
      "10100 10200\n",
      "10200 10300\n",
      "10300 10400\n",
      "10400 10500\n",
      "10500 10600\n",
      "10600 10700\n",
      "10700 10800\n",
      "10800 10900\n",
      "10900 11000\n",
      "11000 11100\n",
      "11100 11200\n",
      "11200 11300\n",
      "11300 11400\n",
      "11400 11500\n",
      "11500 11600\n",
      "11600 11700\n",
      "11700 11800\n",
      "11800 11900\n",
      "11900 12000\n",
      "12000 12100\n",
      "12100 12200\n",
      "12200 12300\n",
      "12300 12400\n",
      "12400 12500\n",
      "12500 12600\n",
      "12600 12700\n",
      "12700 12800\n",
      "12800 12900\n",
      "12900 13000\n",
      "13000 13100\n",
      "13100 13200\n",
      "13200 13300\n",
      "13300 13400\n",
      "13400 13500\n",
      "13500 13600\n",
      "13600 13700\n",
      "13700 13800\n",
      "13800 13900\n",
      "13900 14000\n",
      "14000 14100\n",
      "14100 14200\n",
      "14200 14300\n",
      "14300 14400\n",
      "14400 14500\n",
      "14500 14600\n",
      "14600 14700\n",
      "14700 14800\n",
      "14800 14900\n",
      "14900 15000\n",
      "15000 15100\n",
      "15100 15200\n",
      "15200 15300\n",
      "15300 15400\n",
      "15400 15500\n",
      "15500 15600\n",
      "15600 15700\n",
      "15700 15800\n",
      "15800 15900\n",
      "15900 16000\n",
      "16000 16100\n",
      "16100 16200\n",
      "16200 16300\n",
      "16300 16400\n",
      "16400 16500\n",
      "16500 16600\n",
      "16600 16700\n",
      "16700 16800\n",
      "16800 16900\n",
      "16900 17000\n",
      "17000 17100\n",
      "17100 17200\n",
      "17200 17300\n",
      "17300 17400\n",
      "17400 17500\n",
      "17500 17600\n",
      "17600 17700\n",
      "17700 17800\n",
      "17800 17900\n",
      "17900 18000\n",
      "18000 18100\n",
      "18100 18200\n",
      "18200 18300\n",
      "18300 18400\n",
      "18400 18500\n",
      "18500 18600\n",
      "18600 18700\n",
      "18700 18800\n",
      "18800 18900\n",
      "18900 19000\n",
      "19000 19100\n",
      "19100 19200\n",
      "19200 19300\n",
      "19300 19400\n",
      "19400 19500\n",
      "19500 19600\n",
      "19600 19700\n",
      "19700 19800\n",
      "19800 19900\n",
      "19900 20000\n",
      "20000 20100\n",
      "20100 20200\n",
      "20200 20300\n",
      "20300 20400\n",
      "20400 20500\n",
      "20500 20600\n",
      "20600 20700\n",
      "20700 20800\n",
      "20800 20900\n",
      "20900 21000\n",
      "21000 21100\n",
      "21100 21200\n",
      "21200 21300\n",
      "21300 21400\n",
      "21400 21500\n",
      "21500 21600\n",
      "21600 21700\n",
      "21700 21800\n",
      "21800 21900\n",
      "21900 22000\n",
      "22000 22100\n",
      "22100 22200\n",
      "22200 22300\n",
      "22300 22400\n",
      "22400 22500\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "BATCH_SIZE = 100\n",
    "EPOCHS = 1\n",
    "#Slices of our data\n",
    "for epoch in range(EPOCHS):\n",
    "    for i in tqdm(range(0, len(train_X), BATCH_SIZE)):\n",
    "        print(i, i + BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  0%|          | 0/225 [00:00<?, ?it/s]\u001b[A\n",
      "  0%|          | 1/225 [00:00<00:26,  8.48it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  1%|          | 2/225 [00:00<00:26,  8.39it/s]\u001b[A\n",
      "  1%|▏         | 3/225 [00:00<00:26,  8.33it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  2%|▏         | 4/225 [00:00<00:27,  8.14it/s]\u001b[A\n",
      "  2%|▏         | 5/225 [00:00<00:27,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  3%|▎         | 6/225 [00:00<00:27,  8.02it/s]\u001b[A\n",
      "  3%|▎         | 7/225 [00:00<00:27,  8.00it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  4%|▎         | 8/225 [00:00<00:27,  8.00it/s]\u001b[A\n",
      "  4%|▍         | 9/225 [00:01<00:26,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  4%|▍         | 10/225 [00:01<00:26,  8.04it/s]\u001b[A\n",
      "  5%|▍         | 11/225 [00:01<00:26,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  5%|▌         | 12/225 [00:01<00:26,  8.07it/s]\u001b[A\n",
      "  6%|▌         | 13/225 [00:01<00:26,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  6%|▌         | 14/225 [00:01<00:26,  8.04it/s]\u001b[A\n",
      "  7%|▋         | 15/225 [00:01<00:26,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  7%|▋         | 16/225 [00:01<00:25,  8.04it/s]\u001b[A\n",
      "  8%|▊         | 17/225 [00:02<00:25,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  8%|▊         | 18/225 [00:02<00:25,  7.99it/s]\u001b[A\n",
      "  8%|▊         | 19/225 [00:02<00:25,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  9%|▉         | 20/225 [00:02<00:25,  8.07it/s]\u001b[A\n",
      "  9%|▉         | 21/225 [00:02<00:25,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 10%|▉         | 22/225 [00:02<00:25,  8.04it/s]\u001b[A\n",
      " 10%|█         | 23/225 [00:02<00:24,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 11%|█         | 24/225 [00:02<00:24,  8.09it/s]\u001b[A\n",
      " 11%|█         | 25/225 [00:03<00:24,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 12%|█▏        | 26/225 [00:03<00:24,  8.05it/s]\u001b[A\n",
      " 12%|█▏        | 27/225 [00:03<00:24,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 12%|█▏        | 28/225 [00:03<00:24,  8.05it/s]\u001b[A\n",
      " 13%|█▎        | 29/225 [00:03<00:24,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 13%|█▎        | 30/225 [00:03<00:24,  8.10it/s]\u001b[A\n",
      " 14%|█▍        | 31/225 [00:03<00:24,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 14%|█▍        | 32/225 [00:03<00:23,  8.07it/s]\u001b[A\n",
      " 15%|█▍        | 33/225 [00:04<00:23,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 15%|█▌        | 34/225 [00:04<00:23,  8.08it/s]\u001b[A\n",
      " 16%|█▌        | 35/225 [00:04<00:23,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 16%|█▌        | 36/225 [00:04<00:23,  8.06it/s]\u001b[A\n",
      " 16%|█▋        | 37/225 [00:04<00:23,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 17%|█▋        | 38/225 [00:04<00:23,  8.06it/s]\u001b[A\n",
      " 17%|█▋        | 39/225 [00:04<00:23,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 18%|█▊        | 40/225 [00:04<00:22,  8.09it/s]\u001b[A\n",
      " 18%|█▊        | 41/225 [00:05<00:22,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 19%|█▊        | 42/225 [00:05<00:22,  8.07it/s]\u001b[A\n",
      " 19%|█▉        | 43/225 [00:05<00:22,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 20%|█▉        | 44/225 [00:05<00:22,  8.10it/s]\u001b[A\n",
      " 20%|██        | 45/225 [00:05<00:22,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 20%|██        | 46/225 [00:05<00:22,  8.08it/s]\u001b[A\n",
      " 21%|██        | 47/225 [00:05<00:22,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 21%|██▏       | 48/225 [00:05<00:21,  8.07it/s]\u001b[A\n",
      " 22%|██▏       | 49/225 [00:06<00:21,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 22%|██▏       | 50/225 [00:06<00:21,  8.09it/s]\u001b[A\n",
      " 23%|██▎       | 51/225 [00:06<00:21,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 23%|██▎       | 52/225 [00:06<00:21,  8.08it/s]\u001b[A\n",
      " 24%|██▎       | 53/225 [00:06<00:21,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 24%|██▍       | 54/225 [00:06<00:21,  8.06it/s]\u001b[A\n",
      " 24%|██▍       | 55/225 [00:06<00:21,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 25%|██▍       | 56/225 [00:06<00:20,  8.05it/s]\u001b[A\n",
      " 25%|██▌       | 57/225 [00:07<00:20,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 26%|██▌       | 58/225 [00:07<00:20,  8.07it/s]\u001b[A\n",
      " 26%|██▌       | 59/225 [00:07<00:20,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 27%|██▋       | 60/225 [00:07<00:20,  8.06it/s]\u001b[A\n",
      " 27%|██▋       | 61/225 [00:07<00:20,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 28%|██▊       | 62/225 [00:07<00:20,  8.04it/s]\u001b[A\n",
      " 28%|██▊       | 63/225 [00:07<00:20,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 28%|██▊       | 64/225 [00:07<00:19,  8.06it/s]\u001b[A\n",
      " 29%|██▉       | 65/225 [00:08<00:19,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 29%|██▉       | 66/225 [00:08<00:19,  8.05it/s]\u001b[A\n",
      " 30%|██▉       | 67/225 [00:08<00:19,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 30%|███       | 68/225 [00:08<00:19,  8.11it/s]\u001b[A\n",
      " 31%|███       | 69/225 [00:08<00:19,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 31%|███       | 70/225 [00:08<00:19,  8.08it/s]\u001b[A\n",
      " 32%|███▏      | 71/225 [00:08<00:19,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 32%|███▏      | 72/225 [00:08<00:18,  8.07it/s]\u001b[A\n",
      " 32%|███▏      | 73/225 [00:09<00:18,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 33%|███▎      | 74/225 [00:09<00:18,  8.06it/s]\u001b[A\n",
      " 33%|███▎      | 75/225 [00:09<00:18,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 34%|███▍      | 76/225 [00:09<00:18,  8.08it/s]\u001b[A\n",
      " 34%|███▍      | 77/225 [00:09<00:18,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 35%|███▍      | 78/225 [00:09<00:18,  8.03it/s]\u001b[A\n",
      " 35%|███▌      | 79/225 [00:09<00:18,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 36%|███▌      | 80/225 [00:09<00:17,  8.07it/s]\u001b[A\n",
      " 36%|███▌      | 81/225 [00:10<00:17,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 36%|███▋      | 82/225 [00:10<00:17,  8.10it/s]\u001b[A\n",
      " 37%|███▋      | 83/225 [00:10<00:17,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 37%|███▋      | 84/225 [00:10<00:17,  8.06it/s]\u001b[A\n",
      " 38%|███▊      | 85/225 [00:10<00:17,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 38%|███▊      | 86/225 [00:10<00:17,  8.06it/s]\u001b[A\n",
      " 39%|███▊      | 87/225 [00:10<00:17,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 39%|███▉      | 88/225 [00:10<00:17,  8.03it/s]\u001b[A\n",
      " 40%|███▉      | 89/225 [00:11<00:16,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 40%|████      | 90/225 [00:11<00:16,  8.03it/s]\u001b[A\n",
      " 40%|████      | 91/225 [00:11<00:16,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 41%|████      | 92/225 [00:11<00:16,  8.06it/s]\u001b[A\n",
      " 41%|████▏     | 93/225 [00:11<00:16,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 42%|████▏     | 94/225 [00:11<00:16,  8.10it/s]\u001b[A\n",
      " 42%|████▏     | 95/225 [00:11<00:16,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 43%|████▎     | 96/225 [00:11<00:15,  8.11it/s]\u001b[A\n",
      " 43%|████▎     | 97/225 [00:12<00:15,  8.13it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 44%|████▎     | 98/225 [00:12<00:15,  8.11it/s]\u001b[A\n",
      " 44%|████▍     | 99/225 [00:12<00:15,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 44%|████▍     | 100/225 [00:12<00:15,  8.04it/s]\u001b[A\n",
      " 45%|████▍     | 101/225 [00:12<00:15,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 45%|████▌     | 102/225 [00:12<00:15,  8.07it/s]\u001b[A\n",
      " 46%|████▌     | 103/225 [00:12<00:15,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 46%|████▌     | 104/225 [00:12<00:15,  8.06it/s]\u001b[A\n",
      " 47%|████▋     | 105/225 [00:13<00:14,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 47%|████▋     | 106/225 [00:13<00:14,  8.07it/s]\u001b[A\n",
      " 48%|████▊     | 107/225 [00:13<00:14,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 48%|████▊     | 108/225 [00:13<00:14,  8.05it/s]\u001b[A\n",
      " 48%|████▊     | 109/225 [00:13<00:14,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 49%|████▉     | 110/225 [00:13<00:14,  8.03it/s]\u001b[A\n",
      " 49%|████▉     | 111/225 [00:13<00:14,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 50%|████▉     | 112/225 [00:13<00:14,  8.05it/s]\u001b[A\n",
      " 50%|█████     | 113/225 [00:14<00:13,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 51%|█████     | 114/225 [00:14<00:13,  8.13it/s]\u001b[A\n",
      " 51%|█████     | 115/225 [00:14<00:13,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 52%|█████▏    | 116/225 [00:14<00:13,  8.13it/s]\u001b[A\n",
      " 52%|█████▏    | 117/225 [00:14<00:13,  8.19it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 52%|█████▏    | 118/225 [00:14<00:13,  8.21it/s]\u001b[A\n",
      " 53%|█████▎    | 119/225 [00:14<00:12,  8.18it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 53%|█████▎    | 120/225 [00:14<00:12,  8.18it/s]\u001b[A\n",
      " 54%|█████▍    | 121/225 [00:14<00:12,  8.13it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 54%|█████▍    | 122/225 [00:15<00:12,  8.09it/s]\u001b[A\n",
      " 55%|█████▍    | 123/225 [00:15<00:12,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 55%|█████▌    | 124/225 [00:15<00:12,  8.04it/s]\u001b[A\n",
      " 56%|█████▌    | 125/225 [00:15<00:12,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 56%|█████▌    | 126/225 [00:15<00:12,  8.06it/s]\u001b[A\n",
      " 56%|█████▋    | 127/225 [00:15<00:12,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 57%|█████▋    | 128/225 [00:15<00:11,  8.09it/s]\u001b[A\n",
      " 57%|█████▋    | 129/225 [00:15<00:11,  8.00it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 58%|█████▊    | 130/225 [00:16<00:11,  8.02it/s]\u001b[A\n",
      " 58%|█████▊    | 131/225 [00:16<00:11,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 59%|█████▊    | 132/225 [00:16<00:11,  8.02it/s]\u001b[A\n",
      " 59%|█████▉    | 133/225 [00:16<00:11,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 60%|█████▉    | 134/225 [00:16<00:11,  7.99it/s]\u001b[A\n",
      " 60%|██████    | 135/225 [00:16<00:11,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 60%|██████    | 136/225 [00:16<00:11,  8.04it/s]\u001b[A\n",
      " 61%|██████    | 137/225 [00:16<00:10,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 61%|██████▏   | 138/225 [00:17<00:10,  8.03it/s]\u001b[A\n",
      " 62%|██████▏   | 139/225 [00:17<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 62%|██████▏   | 140/225 [00:17<00:10,  8.08it/s]\u001b[A\n",
      " 63%|██████▎   | 141/225 [00:17<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 63%|██████▎   | 142/225 [00:17<00:10,  8.06it/s]\u001b[A\n",
      " 64%|██████▎   | 143/225 [00:17<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 64%|██████▍   | 144/225 [00:17<00:10,  8.07it/s]\u001b[A\n",
      " 64%|██████▍   | 145/225 [00:17<00:09,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 65%|██████▍   | 146/225 [00:18<00:09,  8.10it/s]\u001b[A\n",
      " 65%|██████▌   | 147/225 [00:18<00:09,  8.13it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 66%|██████▌   | 148/225 [00:18<00:09,  8.09it/s]\u001b[A\n",
      " 66%|██████▌   | 149/225 [00:18<00:09,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 67%|██████▋   | 150/225 [00:18<00:09,  8.07it/s]\u001b[A\n",
      " 67%|██████▋   | 151/225 [00:18<00:09,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 68%|██████▊   | 152/225 [00:18<00:09,  8.07it/s]\u001b[A\n",
      " 68%|██████▊   | 153/225 [00:18<00:08,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 68%|██████▊   | 154/225 [00:19<00:08,  8.09it/s]\u001b[A\n",
      " 69%|██████▉   | 155/225 [00:19<00:08,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 69%|██████▉   | 156/225 [00:19<00:08,  8.05it/s]\u001b[A\n",
      " 70%|██████▉   | 157/225 [00:19<00:08,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 70%|███████   | 158/225 [00:19<00:08,  8.01it/s]\u001b[A\n",
      " 71%|███████   | 159/225 [00:19<00:08,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 71%|███████   | 160/225 [00:19<00:08,  8.05it/s]\u001b[A\n",
      " 72%|███████▏  | 161/225 [00:19<00:07,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 72%|███████▏  | 162/225 [00:20<00:07,  8.07it/s]\u001b[A\n",
      " 72%|███████▏  | 163/225 [00:20<00:07,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 73%|███████▎  | 164/225 [00:20<00:07,  8.07it/s]\u001b[A\n",
      " 73%|███████▎  | 165/225 [00:20<00:07,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 74%|███████▍  | 166/225 [00:20<00:07,  8.04it/s]\u001b[A\n",
      " 74%|███████▍  | 167/225 [00:20<00:07,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 75%|███████▍  | 168/225 [00:20<00:07,  8.06it/s]\u001b[A\n",
      " 75%|███████▌  | 169/225 [00:20<00:06,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 76%|███████▌  | 170/225 [00:21<00:06,  8.07it/s]\u001b[A\n",
      " 76%|███████▌  | 171/225 [00:21<00:06,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 76%|███████▋  | 172/225 [00:21<00:06,  8.09it/s]\u001b[A\n",
      " 77%|███████▋  | 173/225 [00:21<00:06,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 77%|███████▋  | 174/225 [00:21<00:06,  8.09it/s]\u001b[A\n",
      " 78%|███████▊  | 175/225 [00:21<00:06,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 78%|███████▊  | 176/225 [00:21<00:06,  8.07it/s]\u001b[A\n",
      " 79%|███████▊  | 177/225 [00:21<00:05,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 79%|███████▉  | 178/225 [00:22<00:05,  8.09it/s]\u001b[A\n",
      " 80%|███████▉  | 179/225 [00:22<00:05,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 80%|████████  | 180/225 [00:22<00:05,  8.07it/s]\u001b[A\n",
      " 80%|████████  | 181/225 [00:22<00:05,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 81%|████████  | 182/225 [00:22<00:05,  8.09it/s]\u001b[A\n",
      " 81%|████████▏ | 183/225 [00:22<00:05,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 82%|████████▏ | 184/225 [00:22<00:05,  8.04it/s]\u001b[A\n",
      " 82%|████████▏ | 185/225 [00:22<00:04,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 83%|████████▎ | 186/225 [00:23<00:04,  8.07it/s]\u001b[A\n",
      " 83%|████████▎ | 187/225 [00:23<00:04,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 84%|████████▎ | 188/225 [00:23<00:04,  8.12it/s]\u001b[A\n",
      " 84%|████████▍ | 189/225 [00:23<00:04,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 84%|████████▍ | 190/225 [00:23<00:04,  8.10it/s]\u001b[A\n",
      " 85%|████████▍ | 191/225 [00:23<00:04,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 85%|████████▌ | 192/225 [00:23<00:04,  8.10it/s]\u001b[A\n",
      " 86%|████████▌ | 193/225 [00:23<00:03,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 86%|████████▌ | 194/225 [00:24<00:03,  8.06it/s]\u001b[A\n",
      " 87%|████████▋ | 195/225 [00:24<00:03,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 87%|████████▋ | 196/225 [00:24<00:03,  8.09it/s]\u001b[A\n",
      " 88%|████████▊ | 197/225 [00:24<00:03,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 88%|████████▊ | 198/225 [00:24<00:03,  8.15it/s]\u001b[A\n",
      " 88%|████████▊ | 199/225 [00:24<00:03,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 89%|████████▉ | 200/225 [00:24<00:03,  8.12it/s]\u001b[A\n",
      " 89%|████████▉ | 201/225 [00:24<00:02,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 90%|████████▉ | 202/225 [00:25<00:02,  8.12it/s]\u001b[A\n",
      " 90%|█████████ | 203/225 [00:25<00:02,  8.16it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 91%|█████████ | 204/225 [00:25<00:02,  8.14it/s]\u001b[A\n",
      " 91%|█████████ | 205/225 [00:25<00:02,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 92%|█████████▏| 206/225 [00:25<00:02,  8.08it/s]\u001b[A\n",
      " 92%|█████████▏| 207/225 [00:25<00:02,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 92%|█████████▏| 208/225 [00:25<00:02,  8.13it/s]\u001b[A\n",
      " 93%|█████████▎| 209/225 [00:25<00:01,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 93%|█████████▎| 210/225 [00:26<00:01,  8.02it/s]\u001b[A\n",
      " 94%|█████████▍| 211/225 [00:26<00:01,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 94%|█████████▍| 212/225 [00:26<00:01,  8.10it/s]\u001b[A\n",
      " 95%|█████████▍| 213/225 [00:26<00:01,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 95%|█████████▌| 214/225 [00:26<00:01,  8.09it/s]\u001b[A\n",
      " 96%|█████████▌| 215/225 [00:26<00:01,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 96%|█████████▌| 216/225 [00:26<00:01,  8.06it/s]\u001b[A\n",
      " 96%|█████████▋| 217/225 [00:26<00:00,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 97%|█████████▋| 218/225 [00:27<00:00,  8.07it/s]\u001b[A\n",
      " 97%|█████████▋| 219/225 [00:27<00:00,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 98%|█████████▊| 220/225 [00:27<00:00,  8.10it/s]\u001b[A\n",
      " 98%|█████████▊| 221/225 [00:27<00:00,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 99%|█████████▊| 222/225 [00:27<00:00,  8.06it/s]\u001b[A\n",
      " 99%|█████████▉| 223/225 [00:27<00:00,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 225/225 [00:27<00:00,  8.09it/s]\u001b[A\n",
      "\n",
      "  0%|          | 0/225 [00:00<?, ?it/s]\u001b[A\n",
      "  0%|          | 1/225 [00:00<00:27,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  1%|          | 2/225 [00:00<00:27,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  1%|▏         | 3/225 [00:00<00:27,  8.08it/s]\u001b[A\n",
      "  2%|▏         | 4/225 [00:00<00:27,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  2%|▏         | 5/225 [00:00<00:27,  8.10it/s]\u001b[A\n",
      "  3%|▎         | 6/225 [00:00<00:27,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  3%|▎         | 7/225 [00:00<00:27,  8.06it/s]\u001b[A\n",
      "  4%|▎         | 8/225 [00:00<00:26,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  4%|▍         | 9/225 [00:01<00:26,  8.07it/s]\u001b[A\n",
      "  4%|▍         | 10/225 [00:01<00:26,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  5%|▍         | 11/225 [00:01<00:26,  8.07it/s]\u001b[A\n",
      "  5%|▌         | 12/225 [00:01<00:26,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  6%|▌         | 13/225 [00:01<00:26,  8.06it/s]\u001b[A\n",
      "  6%|▌         | 14/225 [00:01<00:26,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  7%|▋         | 15/225 [00:01<00:26,  8.05it/s]\u001b[A\n",
      "  7%|▋         | 16/225 [00:01<00:25,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  8%|▊         | 17/225 [00:02<00:25,  8.04it/s]\u001b[A\n",
      "  8%|▊         | 18/225 [00:02<00:25,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  8%|▊         | 19/225 [00:02<00:25,  8.04it/s]\u001b[A\n",
      "  9%|▉         | 20/225 [00:02<00:25,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  9%|▉         | 21/225 [00:02<00:25,  8.07it/s]\u001b[A\n",
      " 10%|▉         | 22/225 [00:02<00:25,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 10%|█         | 23/225 [00:02<00:24,  8.12it/s]\u001b[A\n",
      " 11%|█         | 24/225 [00:02<00:24,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 11%|█         | 25/225 [00:03<00:24,  8.06it/s]\u001b[A\n",
      " 12%|█▏        | 26/225 [00:03<00:24,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 12%|█▏        | 27/225 [00:03<00:24,  8.08it/s]\u001b[A\n",
      " 12%|█▏        | 28/225 [00:03<00:24,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 13%|█▎        | 29/225 [00:03<00:24,  8.07it/s]\u001b[A\n",
      " 13%|█▎        | 30/225 [00:03<00:24,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 14%|█▍        | 31/225 [00:03<00:23,  8.11it/s]\u001b[A\n",
      " 14%|█▍        | 32/225 [00:03<00:23,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 15%|█▍        | 33/225 [00:04<00:23,  8.10it/s]\u001b[A\n",
      " 15%|█▌        | 34/225 [00:04<00:23,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 16%|█▌        | 35/225 [00:04<00:23,  8.12it/s]\u001b[A\n",
      " 16%|█▌        | 36/225 [00:04<00:23,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 16%|█▋        | 37/225 [00:04<00:23,  8.04it/s]\u001b[A\n",
      " 17%|█▋        | 38/225 [00:04<00:23,  7.93it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 17%|█▋        | 39/225 [00:04<00:23,  8.01it/s]\u001b[A\n",
      " 18%|█▊        | 40/225 [00:04<00:23,  7.98it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 18%|█▊        | 41/225 [00:05<00:23,  7.91it/s]\u001b[A\n",
      " 19%|█▊        | 42/225 [00:05<00:23,  7.86it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 19%|█▉        | 43/225 [00:05<00:22,  7.96it/s]\u001b[A\n",
      " 20%|█▉        | 44/225 [00:05<00:22,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 20%|██        | 45/225 [00:05<00:22,  8.00it/s]\u001b[A\n",
      " 20%|██        | 46/225 [00:05<00:22,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 21%|██        | 47/225 [00:05<00:22,  8.03it/s]\u001b[A\n",
      " 21%|██▏       | 48/225 [00:05<00:22,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 22%|██▏       | 49/225 [00:06<00:21,  8.10it/s]\u001b[A\n",
      " 22%|██▏       | 50/225 [00:06<00:21,  8.13it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 23%|██▎       | 51/225 [00:06<00:21,  8.11it/s]\u001b[A\n",
      " 23%|██▎       | 52/225 [00:06<00:21,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 24%|██▎       | 53/225 [00:06<00:21,  8.09it/s]\u001b[A\n",
      " 24%|██▍       | 54/225 [00:06<00:21,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 24%|██▍       | 55/225 [00:06<00:21,  8.07it/s]\u001b[A\n",
      " 25%|██▍       | 56/225 [00:06<00:20,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 25%|██▌       | 57/225 [00:07<00:20,  8.02it/s]\u001b[A\n",
      " 26%|██▌       | 58/225 [00:07<00:20,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 26%|██▌       | 59/225 [00:07<00:20,  8.05it/s]\u001b[A\n",
      " 27%|██▋       | 60/225 [00:07<00:20,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 27%|██▋       | 61/225 [00:07<00:20,  8.07it/s]\u001b[A\n",
      " 28%|██▊       | 62/225 [00:07<00:20,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 28%|██▊       | 63/225 [00:07<00:20,  8.02it/s]\u001b[A\n",
      " 28%|██▊       | 64/225 [00:07<00:20,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 29%|██▉       | 65/225 [00:08<00:19,  8.08it/s]\u001b[A\n",
      " 29%|██▉       | 66/225 [00:08<00:19,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 30%|██▉       | 67/225 [00:08<00:19,  8.14it/s]\u001b[A\n",
      " 30%|███       | 68/225 [00:08<00:19,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 31%|███       | 69/225 [00:08<00:19,  8.12it/s]\u001b[A\n",
      " 31%|███       | 70/225 [00:08<00:19,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 32%|███▏      | 71/225 [00:08<00:19,  8.08it/s]\u001b[A\n",
      " 32%|███▏      | 72/225 [00:08<00:19,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 32%|███▏      | 73/225 [00:09<00:18,  8.04it/s]\u001b[A\n",
      " 33%|███▎      | 74/225 [00:09<00:18,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 33%|███▎      | 75/225 [00:09<00:18,  8.07it/s]\u001b[A\n",
      " 34%|███▍      | 76/225 [00:09<00:18,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 34%|███▍      | 77/225 [00:09<00:18,  8.14it/s]\u001b[A\n",
      " 35%|███▍      | 78/225 [00:09<00:18,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 35%|███▌      | 79/225 [00:09<00:17,  8.12it/s]\u001b[A\n",
      " 36%|███▌      | 80/225 [00:09<00:17,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 36%|███▌      | 81/225 [00:10<00:17,  8.12it/s]\u001b[A\n",
      " 36%|███▋      | 82/225 [00:10<00:17,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 37%|███▋      | 83/225 [00:10<00:17,  8.08it/s]\u001b[A\n",
      " 37%|███▋      | 84/225 [00:10<00:17,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 38%|███▊      | 85/225 [00:10<00:17,  8.11it/s]\u001b[A\n",
      " 38%|███▊      | 86/225 [00:10<00:17,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 39%|███▊      | 87/225 [00:10<00:17,  8.10it/s]\u001b[A\n",
      " 39%|███▉      | 88/225 [00:10<00:16,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 40%|███▉      | 89/225 [00:11<00:16,  8.08it/s]\u001b[A\n",
      " 40%|████      | 90/225 [00:11<00:16,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 40%|████      | 91/225 [00:11<00:16,  8.16it/s]\u001b[A\n",
      " 41%|████      | 92/225 [00:11<00:16,  8.19it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 41%|████▏     | 93/225 [00:11<00:16,  8.19it/s]\u001b[A\n",
      " 42%|████▏     | 94/225 [00:11<00:16,  8.17it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 42%|████▏     | 95/225 [00:11<00:15,  8.22it/s]\u001b[A\n",
      " 43%|████▎     | 96/225 [00:11<00:15,  8.21it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 43%|████▎     | 97/225 [00:12<00:15,  8.14it/s]\u001b[A\n",
      " 44%|████▎     | 98/225 [00:12<00:15,  8.14it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 44%|████▍     | 99/225 [00:12<00:15,  8.13it/s]\u001b[A\n",
      " 44%|████▍     | 100/225 [00:12<00:15,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 45%|████▍     | 101/225 [00:12<00:15,  8.06it/s]\u001b[A\n",
      " 45%|████▌     | 102/225 [00:12<00:15,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 46%|████▌     | 103/225 [00:12<00:15,  8.06it/s]\u001b[A\n",
      " 46%|████▌     | 104/225 [00:12<00:15,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 47%|████▋     | 105/225 [00:13<00:14,  8.10it/s]\u001b[A\n",
      " 47%|████▋     | 106/225 [00:13<00:14,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 48%|████▊     | 107/225 [00:13<00:14,  8.07it/s]\u001b[A\n",
      " 48%|████▊     | 108/225 [00:13<00:14,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 48%|████▊     | 109/225 [00:13<00:14,  8.13it/s]\u001b[A\n",
      " 49%|████▉     | 110/225 [00:13<00:14,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 49%|████▉     | 111/225 [00:13<00:14,  8.09it/s]\u001b[A\n",
      " 50%|████▉     | 112/225 [00:13<00:14,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 50%|█████     | 113/225 [00:13<00:13,  8.08it/s]\u001b[A\n",
      " 51%|█████     | 114/225 [00:14<00:13,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 51%|█████     | 115/225 [00:14<00:13,  7.97it/s]\u001b[A\n",
      " 52%|█████▏    | 116/225 [00:14<00:13,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 52%|█████▏    | 117/225 [00:14<00:13,  8.05it/s]\u001b[A\n",
      " 52%|█████▏    | 118/225 [00:14<00:13,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 53%|█████▎    | 119/225 [00:14<00:13,  8.07it/s]\u001b[A\n",
      " 53%|█████▎    | 120/225 [00:14<00:12,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 54%|█████▍    | 121/225 [00:14<00:12,  8.08it/s]\u001b[A\n",
      " 54%|█████▍    | 122/225 [00:15<00:12,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 55%|█████▍    | 123/225 [00:15<00:12,  8.09it/s]\u001b[A\n",
      " 55%|█████▌    | 124/225 [00:15<00:12,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 56%|█████▌    | 125/225 [00:15<00:12,  8.09it/s]\u001b[A\n",
      " 56%|█████▌    | 126/225 [00:15<00:12,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 56%|█████▋    | 127/225 [00:15<00:12,  8.11it/s]\u001b[A\n",
      " 57%|█████▋    | 128/225 [00:15<00:12,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 57%|█████▋    | 129/225 [00:15<00:11,  8.09it/s]\u001b[A\n",
      " 58%|█████▊    | 130/225 [00:16<00:11,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 58%|█████▊    | 131/225 [00:16<00:11,  8.06it/s]\u001b[A\n",
      " 59%|█████▊    | 132/225 [00:16<00:11,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 59%|█████▉    | 133/225 [00:16<00:11,  8.04it/s]\u001b[A\n",
      " 60%|█████▉    | 134/225 [00:16<00:11,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 60%|██████    | 135/225 [00:16<00:11,  8.10it/s]\u001b[A\n",
      " 60%|██████    | 136/225 [00:16<00:10,  8.15it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 61%|██████    | 137/225 [00:16<00:10,  8.06it/s]\u001b[A\n",
      " 61%|██████▏   | 138/225 [00:17<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 62%|██████▏   | 139/225 [00:17<00:10,  8.08it/s]\u001b[A\n",
      " 62%|██████▏   | 140/225 [00:17<00:10,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 63%|██████▎   | 141/225 [00:17<00:10,  8.08it/s]\u001b[A\n",
      " 63%|██████▎   | 142/225 [00:17<00:10,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 64%|██████▎   | 143/225 [00:17<00:10,  8.06it/s]\u001b[A\n",
      " 64%|██████▍   | 144/225 [00:17<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 64%|██████▍   | 145/225 [00:17<00:09,  8.06it/s]\u001b[A\n",
      " 65%|██████▍   | 146/225 [00:18<00:09,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 65%|██████▌   | 147/225 [00:18<00:09,  8.09it/s]\u001b[A\n",
      " 66%|██████▌   | 148/225 [00:18<00:09,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 66%|██████▌   | 149/225 [00:18<00:09,  8.11it/s]\u001b[A\n",
      " 67%|██████▋   | 150/225 [00:18<00:09,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 67%|██████▋   | 151/225 [00:18<00:09,  8.09it/s]\u001b[A\n",
      " 68%|██████▊   | 152/225 [00:18<00:09,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 68%|██████▊   | 153/225 [00:18<00:08,  8.05it/s]\u001b[A\n",
      " 68%|██████▊   | 154/225 [00:19<00:08,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 69%|██████▉   | 155/225 [00:19<00:08,  8.09it/s]\u001b[A\n",
      " 69%|██████▉   | 156/225 [00:19<00:08,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 70%|██████▉   | 157/225 [00:19<00:08,  8.06it/s]\u001b[A\n",
      " 70%|███████   | 158/225 [00:19<00:08,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 71%|███████   | 159/225 [00:19<00:08,  8.08it/s]\u001b[A\n",
      " 71%|███████   | 160/225 [00:19<00:08,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 72%|███████▏  | 161/225 [00:19<00:07,  8.11it/s]\u001b[A\n",
      " 72%|███████▏  | 162/225 [00:20<00:07,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 72%|███████▏  | 163/225 [00:20<00:07,  8.07it/s]\u001b[A\n",
      " 73%|███████▎  | 164/225 [00:20<00:07,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 73%|███████▎  | 165/225 [00:20<00:07,  8.09it/s]\u001b[A\n",
      " 74%|███████▍  | 166/225 [00:20<00:07,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 74%|███████▍  | 167/225 [00:20<00:07,  8.05it/s]\u001b[A\n",
      " 75%|███████▍  | 168/225 [00:20<00:07,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 75%|███████▌  | 169/225 [00:20<00:06,  8.01it/s]\u001b[A\n",
      " 76%|███████▌  | 170/225 [00:21<00:06,  7.99it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 76%|███████▌  | 171/225 [00:21<00:06,  7.99it/s]\u001b[A\n",
      " 76%|███████▋  | 172/225 [00:21<00:06,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 77%|███████▋  | 173/225 [00:21<00:06,  8.00it/s]\u001b[A\n",
      " 77%|███████▋  | 174/225 [00:21<00:06,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 78%|███████▊  | 175/225 [00:21<00:06,  8.03it/s]\u001b[A\n",
      " 78%|███████▊  | 176/225 [00:21<00:06,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 79%|███████▊  | 177/225 [00:21<00:05,  8.06it/s]\u001b[A\n",
      " 79%|███████▉  | 178/225 [00:22<00:05,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 80%|███████▉  | 179/225 [00:22<00:05,  8.08it/s]\u001b[A\n",
      " 80%|████████  | 180/225 [00:22<00:05,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 80%|████████  | 181/225 [00:22<00:05,  8.07it/s]\u001b[A\n",
      " 81%|████████  | 182/225 [00:22<00:05,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 81%|████████▏ | 183/225 [00:22<00:05,  8.06it/s]\u001b[A\n",
      " 82%|████████▏ | 184/225 [00:22<00:05,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 82%|████████▏ | 185/225 [00:22<00:04,  8.05it/s]\u001b[A\n",
      " 83%|████████▎ | 186/225 [00:23<00:04,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 83%|████████▎ | 187/225 [00:23<00:04,  8.07it/s]\u001b[A\n",
      " 84%|████████▎ | 188/225 [00:23<00:04,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 84%|████████▍ | 189/225 [00:23<00:04,  8.06it/s]\u001b[A\n",
      " 84%|████████▍ | 190/225 [00:23<00:04,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 85%|████████▍ | 191/225 [00:23<00:04,  8.07it/s]\u001b[A\n",
      " 85%|████████▌ | 192/225 [00:23<00:04,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 86%|████████▌ | 193/225 [00:23<00:03,  8.09it/s]\u001b[A\n",
      " 86%|████████▌ | 194/225 [00:24<00:03,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 87%|████████▋ | 195/225 [00:24<00:03,  8.08it/s]\u001b[A\n",
      " 87%|████████▋ | 196/225 [00:24<00:03,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 88%|████████▊ | 197/225 [00:24<00:03,  8.10it/s]\u001b[A\n",
      " 88%|████████▊ | 198/225 [00:24<00:03,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 88%|████████▊ | 199/225 [00:24<00:03,  8.11it/s]\u001b[A\n",
      " 89%|████████▉ | 200/225 [00:24<00:03,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 89%|████████▉ | 201/225 [00:24<00:02,  8.10it/s]\u001b[A\n",
      " 90%|████████▉ | 202/225 [00:25<00:02,  8.13it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 90%|█████████ | 203/225 [00:25<00:02,  8.11it/s]\u001b[A\n",
      " 91%|█████████ | 204/225 [00:25<00:02,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 91%|█████████ | 205/225 [00:25<00:02,  8.08it/s]\u001b[A\n",
      " 92%|█████████▏| 206/225 [00:25<00:02,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 92%|█████████▏| 207/225 [00:25<00:02,  8.06it/s]\u001b[A\n",
      " 92%|█████████▏| 208/225 [00:25<00:02,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 93%|█████████▎| 209/225 [00:25<00:01,  8.06it/s]\u001b[A\n",
      " 93%|█████████▎| 210/225 [00:26<00:01,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 94%|█████████▍| 211/225 [00:26<00:01,  8.07it/s]\u001b[A\n",
      " 94%|█████████▍| 212/225 [00:26<00:01,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 95%|█████████▍| 213/225 [00:26<00:01,  8.13it/s]\u001b[A\n",
      " 95%|█████████▌| 214/225 [00:26<00:01,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 96%|█████████▌| 215/225 [00:26<00:01,  8.09it/s]\u001b[A\n",
      " 96%|█████████▌| 216/225 [00:26<00:01,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 96%|█████████▋| 217/225 [00:26<00:00,  8.06it/s]\u001b[A\n",
      " 97%|█████████▋| 218/225 [00:27<00:00,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 97%|█████████▋| 219/225 [00:27<00:00,  8.09it/s]\u001b[A\n",
      " 98%|█████████▊| 220/225 [00:27<00:00,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 98%|█████████▊| 221/225 [00:27<00:00,  8.10it/s]\u001b[A\n",
      " 99%|█████████▊| 222/225 [00:27<00:00,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 99%|█████████▉| 223/225 [00:27<00:00,  8.10it/s]\u001b[A\n",
      "100%|██████████| 225/225 [00:27<00:00,  8.09it/s]\u001b[A\n",
      "\n",
      "  0%|          | 0/225 [00:00<?, ?it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  0%|          | 1/225 [00:00<00:27,  8.26it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  1%|          | 2/225 [00:00<00:27,  8.22it/s]\u001b[A\n",
      "  1%|▏         | 3/225 [00:00<00:27,  8.19it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  2%|▏         | 4/225 [00:00<00:27,  8.15it/s]\u001b[A\n",
      "  2%|▏         | 5/225 [00:00<00:26,  8.18it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  3%|▎         | 6/225 [00:00<00:26,  8.16it/s]\u001b[A\n",
      "  3%|▎         | 7/225 [00:00<00:26,  8.13it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  4%|▎         | 8/225 [00:00<00:26,  8.11it/s]\u001b[A\n",
      "  4%|▍         | 9/225 [00:01<00:26,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  4%|▍         | 10/225 [00:01<00:26,  8.06it/s]\u001b[A\n",
      "  5%|▍         | 11/225 [00:01<00:26,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  5%|▌         | 12/225 [00:01<00:26,  8.09it/s]\u001b[A\n",
      "  6%|▌         | 13/225 [00:01<00:26,  8.14it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  6%|▌         | 14/225 [00:01<00:26,  8.10it/s]\u001b[A\n",
      "  7%|▋         | 15/225 [00:01<00:26,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  7%|▋         | 16/225 [00:01<00:25,  8.05it/s]\u001b[A\n",
      "  8%|▊         | 17/225 [00:02<00:25,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  8%|▊         | 18/225 [00:02<00:25,  8.11it/s]\u001b[A\n",
      "  8%|▊         | 19/225 [00:02<00:25,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  9%|▉         | 20/225 [00:02<00:25,  8.02it/s]\u001b[A\n",
      "  9%|▉         | 21/225 [00:02<00:25,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 10%|▉         | 22/225 [00:02<00:25,  8.04it/s]\u001b[A\n",
      " 10%|█         | 23/225 [00:02<00:25,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 11%|█         | 24/225 [00:02<00:25,  8.03it/s]\u001b[A\n",
      " 11%|█         | 25/225 [00:03<00:24,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 12%|█▏        | 26/225 [00:03<00:24,  8.07it/s]\u001b[A\n",
      " 12%|█▏        | 27/225 [00:03<00:24,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 12%|█▏        | 28/225 [00:03<00:24,  8.06it/s]\u001b[A\n",
      " 13%|█▎        | 29/225 [00:03<00:24,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 13%|█▎        | 30/225 [00:03<00:24,  8.07it/s]\u001b[A\n",
      " 14%|█▍        | 31/225 [00:03<00:24,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 14%|█▍        | 32/225 [00:03<00:23,  8.07it/s]\u001b[A\n",
      " 15%|█▍        | 33/225 [00:04<00:23,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 15%|█▌        | 34/225 [00:04<00:23,  8.08it/s]\u001b[A\n",
      " 16%|█▌        | 35/225 [00:04<00:23,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 16%|█▌        | 36/225 [00:04<00:23,  8.08it/s]\u001b[A\n",
      " 16%|█▋        | 37/225 [00:04<00:23,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 17%|█▋        | 38/225 [00:04<00:23,  8.06it/s]\u001b[A\n",
      " 17%|█▋        | 39/225 [00:04<00:23,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 18%|█▊        | 40/225 [00:04<00:22,  8.09it/s]\u001b[A\n",
      " 18%|█▊        | 41/225 [00:05<00:22,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 19%|█▊        | 42/225 [00:05<00:22,  8.09it/s]\u001b[A\n",
      " 19%|█▉        | 43/225 [00:05<00:22,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 20%|█▉        | 44/225 [00:05<00:22,  8.09it/s]\u001b[A\n",
      " 20%|██        | 45/225 [00:05<00:22,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 20%|██        | 46/225 [00:05<00:22,  8.10it/s]\u001b[A\n",
      " 21%|██        | 47/225 [00:05<00:21,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 21%|██▏       | 48/225 [00:05<00:22,  8.03it/s]\u001b[A\n",
      " 22%|██▏       | 49/225 [00:06<00:21,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 22%|██▏       | 50/225 [00:06<00:21,  8.08it/s]\u001b[A\n",
      " 23%|██▎       | 51/225 [00:06<00:21,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 23%|██▎       | 52/225 [00:06<00:21,  8.07it/s]\u001b[A\n",
      " 24%|██▎       | 53/225 [00:06<00:21,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 24%|██▍       | 54/225 [00:06<00:21,  8.09it/s]\u001b[A\n",
      " 24%|██▍       | 55/225 [00:06<00:21,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 25%|██▍       | 56/225 [00:06<00:21,  8.02it/s]\u001b[A\n",
      " 25%|██▌       | 57/225 [00:07<00:20,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 26%|██▌       | 58/225 [00:07<00:20,  8.06it/s]\u001b[A\n",
      " 26%|██▌       | 59/225 [00:07<00:20,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 27%|██▋       | 60/225 [00:07<00:20,  8.05it/s]\u001b[A\n",
      " 27%|██▋       | 61/225 [00:07<00:20,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 28%|██▊       | 62/225 [00:07<00:20,  8.09it/s]\u001b[A\n",
      " 28%|██▊       | 63/225 [00:07<00:20,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 28%|██▊       | 64/225 [00:07<00:19,  8.13it/s]\u001b[A\n",
      " 29%|██▉       | 65/225 [00:08<00:19,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 29%|██▉       | 66/225 [00:08<00:19,  8.12it/s]\u001b[A\n",
      " 30%|██▉       | 67/225 [00:08<00:19,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 30%|███       | 68/225 [00:08<00:19,  8.13it/s]\u001b[A\n",
      " 31%|███       | 69/225 [00:08<00:19,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 31%|███       | 70/225 [00:08<00:19,  8.11it/s]\u001b[A\n",
      " 32%|███▏      | 71/225 [00:08<00:19,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 32%|███▏      | 72/225 [00:08<00:19,  8.05it/s]\u001b[A\n",
      " 32%|███▏      | 73/225 [00:09<00:18,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 33%|███▎      | 74/225 [00:09<00:18,  8.05it/s]\u001b[A\n",
      " 33%|███▎      | 75/225 [00:09<00:18,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 34%|███▍      | 76/225 [00:09<00:18,  7.94it/s]\u001b[A\n",
      " 34%|███▍      | 77/225 [00:09<00:18,  7.97it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 35%|███▍      | 78/225 [00:09<00:18,  7.96it/s]\u001b[A\n",
      " 35%|███▌      | 79/225 [00:09<00:18,  7.97it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 36%|███▌      | 80/225 [00:09<00:18,  7.92it/s]\u001b[A\n",
      " 36%|███▌      | 81/225 [00:10<00:18,  7.94it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 36%|███▋      | 82/225 [00:10<00:17,  7.98it/s]\u001b[A\n",
      " 37%|███▋      | 83/225 [00:10<00:17,  7.98it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 37%|███▋      | 84/225 [00:10<00:17,  7.99it/s]\u001b[A\n",
      " 38%|███▊      | 85/225 [00:10<00:17,  7.92it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 38%|███▊      | 86/225 [00:10<00:17,  7.97it/s]\u001b[A\n",
      " 39%|███▊      | 87/225 [00:10<00:17,  7.98it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 39%|███▉      | 88/225 [00:10<00:17,  7.98it/s]\u001b[A\n",
      " 40%|███▉      | 89/225 [00:11<00:16,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 40%|████      | 90/225 [00:11<00:16,  8.05it/s]\u001b[A\n",
      " 40%|████      | 91/225 [00:11<00:16,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 41%|████      | 92/225 [00:11<00:16,  8.08it/s]\u001b[A\n",
      " 41%|████▏     | 93/225 [00:11<00:16,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 42%|████▏     | 94/225 [00:11<00:16,  8.08it/s]\u001b[A\n",
      " 42%|████▏     | 95/225 [00:11<00:16,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 43%|████▎     | 96/225 [00:11<00:15,  8.08it/s]\u001b[A\n",
      " 43%|████▎     | 97/225 [00:12<00:15,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 44%|████▎     | 98/225 [00:12<00:15,  8.10it/s]\u001b[A\n",
      " 44%|████▍     | 99/225 [00:12<00:15,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 44%|████▍     | 100/225 [00:12<00:15,  8.09it/s]\u001b[A\n",
      " 45%|████▍     | 101/225 [00:12<00:15,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 45%|████▌     | 102/225 [00:12<00:15,  8.05it/s]\u001b[A\n",
      " 46%|████▌     | 103/225 [00:12<00:15,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 46%|████▌     | 104/225 [00:12<00:14,  8.09it/s]\u001b[A\n",
      " 47%|████▋     | 105/225 [00:13<00:14,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 47%|████▋     | 106/225 [00:13<00:14,  8.08it/s]\u001b[A\n",
      " 48%|████▊     | 107/225 [00:13<00:14,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 48%|████▊     | 108/225 [00:13<00:14,  8.09it/s]\u001b[A\n",
      " 48%|████▊     | 109/225 [00:13<00:14,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 49%|████▉     | 110/225 [00:13<00:14,  8.09it/s]\u001b[A\n",
      " 49%|████▉     | 111/225 [00:13<00:14,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 50%|████▉     | 112/225 [00:13<00:13,  8.09it/s]\u001b[A\n",
      " 50%|█████     | 113/225 [00:14<00:13,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 51%|█████     | 114/225 [00:14<00:13,  7.99it/s]\u001b[A\n",
      " 51%|█████     | 115/225 [00:14<00:13,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 52%|█████▏    | 116/225 [00:14<00:13,  8.02it/s]\u001b[A\n",
      " 52%|█████▏    | 117/225 [00:14<00:13,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 52%|█████▏    | 118/225 [00:14<00:13,  8.05it/s]\u001b[A\n",
      " 53%|█████▎    | 119/225 [00:14<00:13,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 53%|█████▎    | 120/225 [00:14<00:12,  8.11it/s]\u001b[A\n",
      " 54%|█████▍    | 121/225 [00:15<00:12,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 54%|█████▍    | 122/225 [00:15<00:12,  8.12it/s]\u001b[A\n",
      " 55%|█████▍    | 123/225 [00:15<00:12,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 55%|█████▌    | 124/225 [00:15<00:12,  8.07it/s]\u001b[A\n",
      " 56%|█████▌    | 125/225 [00:15<00:12,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 56%|█████▌    | 126/225 [00:15<00:12,  8.10it/s]\u001b[A\n",
      " 56%|█████▋    | 127/225 [00:15<00:12,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 57%|█████▋    | 128/225 [00:15<00:12,  8.08it/s]\u001b[A\n",
      " 57%|█████▋    | 129/225 [00:15<00:11,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 58%|█████▊    | 130/225 [00:16<00:11,  8.07it/s]\u001b[A\n",
      " 58%|█████▊    | 131/225 [00:16<00:11,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 59%|█████▊    | 132/225 [00:16<00:11,  8.04it/s]\u001b[A\n",
      " 59%|█████▉    | 133/225 [00:16<00:11,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 60%|█████▉    | 134/225 [00:16<00:11,  8.05it/s]\u001b[A\n",
      " 60%|██████    | 135/225 [00:16<00:11,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 60%|██████    | 136/225 [00:16<00:11,  8.07it/s]\u001b[A\n",
      " 61%|██████    | 137/225 [00:16<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 61%|██████▏   | 138/225 [00:17<00:10,  8.08it/s]\u001b[A\n",
      " 62%|██████▏   | 139/225 [00:17<00:10,  8.09it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 62%|██████▏   | 140/225 [00:17<00:10,  8.06it/s]\u001b[A\n",
      " 63%|██████▎   | 141/225 [00:17<00:10,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 63%|██████▎   | 142/225 [00:17<00:10,  8.00it/s]\u001b[A\n",
      " 64%|██████▎   | 143/225 [00:17<00:10,  7.94it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 64%|██████▍   | 144/225 [00:17<00:10,  8.00it/s]\u001b[A\n",
      " 64%|██████▍   | 145/225 [00:17<00:09,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 65%|██████▍   | 146/225 [00:18<00:09,  8.05it/s]\u001b[A\n",
      " 65%|██████▌   | 147/225 [00:18<00:09,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 66%|██████▌   | 148/225 [00:18<00:09,  8.02it/s]\u001b[A\n",
      " 66%|██████▌   | 149/225 [00:18<00:09,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 67%|██████▋   | 150/225 [00:18<00:09,  8.09it/s]\u001b[A\n",
      " 67%|██████▋   | 151/225 [00:18<00:09,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 68%|██████▊   | 152/225 [00:18<00:09,  8.09it/s]\u001b[A\n",
      " 68%|██████▊   | 153/225 [00:18<00:08,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 68%|██████▊   | 154/225 [00:19<00:08,  8.04it/s]\u001b[A\n",
      " 69%|██████▉   | 155/225 [00:19<00:08,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 69%|██████▉   | 156/225 [00:19<00:08,  8.09it/s]\u001b[A\n",
      " 70%|██████▉   | 157/225 [00:19<00:08,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 70%|███████   | 158/225 [00:19<00:08,  8.08it/s]\u001b[A\n",
      " 71%|███████   | 159/225 [00:19<00:08,  8.11it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 71%|███████   | 160/225 [00:19<00:08,  8.10it/s]\u001b[A\n",
      " 72%|███████▏  | 161/225 [00:19<00:07,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 72%|███████▏  | 162/225 [00:20<00:07,  8.10it/s]\u001b[A\n",
      " 72%|███████▏  | 163/225 [00:20<00:07,  8.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 73%|███████▎  | 164/225 [00:20<00:07,  8.06it/s]\u001b[A\n",
      " 73%|███████▎  | 165/225 [00:20<00:07,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 74%|███████▍  | 166/225 [00:20<00:07,  8.04it/s]\u001b[A\n",
      " 74%|███████▍  | 167/225 [00:20<00:07,  8.05it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 75%|███████▍  | 168/225 [00:20<00:07,  8.07it/s]\u001b[A\n",
      " 75%|███████▌  | 169/225 [00:20<00:06,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 76%|███████▌  | 170/225 [00:21<00:06,  8.06it/s]\u001b[A\n",
      " 76%|███████▌  | 171/225 [00:21<00:06,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 76%|███████▋  | 172/225 [00:21<00:06,  8.05it/s]\u001b[A\n",
      " 77%|███████▋  | 173/225 [00:21<00:06,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 77%|███████▋  | 174/225 [00:21<00:06,  8.06it/s]\u001b[A\n",
      " 78%|███████▊  | 175/225 [00:21<00:06,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 78%|███████▊  | 176/225 [00:21<00:06,  8.06it/s]\u001b[A\n",
      " 79%|███████▊  | 177/225 [00:21<00:05,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 79%|███████▉  | 178/225 [00:22<00:05,  8.06it/s]\u001b[A\n",
      " 80%|███████▉  | 179/225 [00:22<00:05,  8.00it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 80%|████████  | 180/225 [00:22<00:05,  8.04it/s]\u001b[A\n",
      " 80%|████████  | 181/225 [00:22<00:05,  8.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 81%|████████  | 182/225 [00:22<00:05,  8.01it/s]\u001b[A\n",
      " 81%|████████▏ | 183/225 [00:22<00:05,  8.01it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 82%|████████▏ | 184/225 [00:22<00:05,  8.08it/s]\u001b[A\n",
      " 82%|████████▏ | 185/225 [00:22<00:05,  7.99it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 83%|████████▎ | 186/225 [00:23<00:04,  7.99it/s]\u001b[A\n",
      " 83%|████████▎ | 187/225 [00:23<00:04,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 84%|████████▎ | 188/225 [00:23<00:04,  8.12it/s]\u001b[A\n",
      " 84%|████████▍ | 189/225 [00:23<00:04,  8.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 84%|████████▍ | 190/225 [00:23<00:04,  8.09it/s]\u001b[A\n",
      " 85%|████████▍ | 191/225 [00:23<00:04,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 85%|████████▌ | 192/225 [00:23<00:04,  8.10it/s]\u001b[A\n",
      " 86%|████████▌ | 193/225 [00:23<00:03,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 86%|████████▌ | 194/225 [00:24<00:03,  8.04it/s]\u001b[A\n",
      " 87%|████████▋ | 195/225 [00:24<00:03,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 87%|████████▋ | 196/225 [00:24<00:03,  8.02it/s]\u001b[A\n",
      " 88%|████████▊ | 197/225 [00:24<00:03,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 88%|████████▊ | 198/225 [00:24<00:03,  8.06it/s]\u001b[A\n",
      " 88%|████████▊ | 199/225 [00:24<00:03,  7.96it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 89%|████████▉ | 200/225 [00:24<00:03,  7.97it/s]\u001b[A\n",
      " 89%|████████▉ | 201/225 [00:24<00:03,  7.86it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 90%|████████▉ | 202/225 [00:25<00:02,  7.90it/s]\u001b[A\n",
      " 90%|█████████ | 203/225 [00:25<00:02,  7.76it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 91%|█████████ | 204/225 [00:25<00:02,  7.85it/s]\u001b[A\n",
      " 91%|█████████ | 205/225 [00:25<00:02,  7.83it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 92%|█████████▏| 206/225 [00:25<00:02,  7.90it/s]\u001b[A\n",
      " 92%|█████████▏| 207/225 [00:25<00:02,  7.93it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 92%|█████████▏| 208/225 [00:25<00:02,  7.99it/s]\u001b[A\n",
      " 93%|█████████▎| 209/225 [00:25<00:01,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 93%|█████████▎| 210/225 [00:26<00:01,  8.02it/s]\u001b[A\n",
      " 94%|█████████▍| 211/225 [00:26<00:01,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 94%|█████████▍| 212/225 [00:26<00:01,  8.02it/s]\u001b[A\n",
      " 95%|█████████▍| 213/225 [00:26<00:01,  8.03it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 95%|█████████▌| 214/225 [00:26<00:01,  8.06it/s]\u001b[A\n",
      " 96%|█████████▌| 215/225 [00:26<00:01,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 96%|█████████▌| 216/225 [00:26<00:01,  8.11it/s]\u001b[A\n",
      " 96%|█████████▋| 217/225 [00:26<00:00,  8.12it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 97%|█████████▋| 218/225 [00:27<00:00,  8.12it/s]\u001b[A\n",
      " 97%|█████████▋| 219/225 [00:27<00:00,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 98%|█████████▊| 220/225 [00:27<00:00,  8.06it/s]\u001b[A\n",
      " 98%|█████████▊| 221/225 [00:27<00:00,  8.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 99%|█████████▊| 222/225 [00:27<00:00,  8.08it/s]\u001b[A\n",
      " 99%|█████████▉| 223/225 [00:27<00:00,  8.04it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "100%|██████████| 225/225 [00:27<00:00,  8.07it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "tensor(0.0990, grad_fn=<MseLossBackward>)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "BATCH_SIZE = 100\n",
    "EPOCHS = 3\n",
    "#Slices of our data\n",
    "for epoch in range(EPOCHS):\n",
    "    for i in tqdm(range(0, len(train_X), BATCH_SIZE)):\n",
    "        batch_X = train_X[i:i+BATCH_SIZE].view(-1,1,50,50)\n",
    "        batch_y = train_y[i:i+BATCH_SIZE]\n",
    "        \n",
    "        net.zero_grad()\n",
    "        outputs = net(batch_X)\n",
    "        loss = loss_function(outputs, batch_y)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "print(loss)\n",
    "        \n",
    "        # Do the fitment optimisation - zero the gradients"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  0%|          | 0/2494 [00:00<?, ?it/s]\u001b[A\n",
      "  2%|▏         | 62/2494 [00:00<00:03, 619.36it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "  5%|▍         | 123/2494 [00:00<00:03, 616.35it/s]\u001b[A\n",
      "  7%|▋         | 186/2494 [00:00<00:03, 619.29it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 10%|▉         | 247/2494 [00:00<00:03, 616.30it/s]\u001b[A\n",
      " 12%|█▏        | 311/2494 [00:00<00:03, 623.06it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 15%|█▍        | 371/2494 [00:00<00:03, 615.79it/s]\u001b[A\n",
      " 17%|█▋        | 433/2494 [00:00<00:03, 616.88it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 20%|█▉        | 496/2494 [00:00<00:03, 618.76it/s]\u001b[A\n",
      " 22%|██▏       | 559/2494 [00:00<00:03, 620.08it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 25%|██▍       | 620/2494 [00:01<00:03, 614.99it/s]\u001b[A\n",
      " 27%|██▋       | 682/2494 [00:01<00:02, 616.31it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 30%|██▉       | 743/2494 [00:01<00:02, 606.87it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 32%|███▏      | 805/2494 [00:01<00:02, 608.79it/s]\u001b[A\n",
      " 35%|███▍      | 868/2494 [00:01<00:02, 613.02it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 37%|███▋      | 930/2494 [00:01<00:02, 614.93it/s]\u001b[A\n",
      " 40%|███▉      | 993/2494 [00:01<00:02, 619.21it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 42%|████▏     | 1055/2494 [00:01<00:02, 615.58it/s]\u001b[A\n",
      " 45%|████▍     | 1117/2494 [00:01<00:02, 614.90it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 47%|████▋     | 1179/2494 [00:01<00:02, 614.41it/s]\u001b[A\n",
      " 50%|████▉     | 1242/2494 [00:02<00:02, 618.85it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 52%|█████▏    | 1304/2494 [00:02<00:01, 619.02it/s]\u001b[A\n",
      " 55%|█████▍    | 1366/2494 [00:02<00:01, 617.30it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 57%|█████▋    | 1428/2494 [00:02<00:01, 612.44it/s]\u001b[A\n",
      " 60%|█████▉    | 1491/2494 [00:02<00:01, 615.62it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 62%|██████▏   | 1553/2494 [00:02<00:01, 611.29it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 65%|██████▍   | 1615/2494 [00:02<00:01, 613.71it/s]\u001b[A\n",
      " 67%|██████▋   | 1679/2494 [00:02<00:01, 620.26it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 70%|██████▉   | 1742/2494 [00:02<00:01, 613.87it/s]\u001b[A\n",
      " 72%|███████▏  | 1804/2494 [00:02<00:01, 613.70it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 75%|███████▍  | 1868/2494 [00:03<00:01, 619.39it/s]\u001b[A\n",
      " 77%|███████▋  | 1930/2494 [00:03<00:00, 619.40it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 80%|███████▉  | 1992/2494 [00:03<00:00, 612.95it/s]\u001b[A\n",
      " 82%|████████▏ | 2054/2494 [00:03<00:00, 607.64it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 85%|████████▍ | 2115/2494 [00:03<00:00, 606.35it/s]\u001b[A\n",
      " 87%|████████▋ | 2178/2494 [00:03<00:00, 613.10it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 90%|████████▉ | 2240/2494 [00:03<00:00, 612.94it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 92%|█████████▏| 2302/2494 [00:03<00:00, 600.15it/s]\u001b[A\n",
      " 95%|█████████▍| 2363/2494 [00:03<00:00, 600.88it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      " 97%|█████████▋| 2424/2494 [00:03<00:00, 596.53it/s]\u001b[A\n",
      "100%|██████████| 2494/2494 [00:04<00:00, 612.51it/s]\u001b[A"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "torch.Size([128, 2, 2])\n",
      "Accuracy:  0.753\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "correct = 0\n",
    "total = 0\n",
    "\n",
    "with torch.no_grad():\n",
    "    for i in tqdm(range(len(test_X))):\n",
    "        real_class = torch.argmax(test_y[i])\n",
    "        net_out = net(test_X[i].view(-1,1,50,50))[0]\n",
    "        \n",
    "        #print(real_class, net_out)\n",
    "        predicted_class = torch.argmax(net_out)\n",
    "        if predicted_class == real_class:\n",
    "            correct += 1\n",
    "        total += 1\n",
    "print(\"Accuracy: \", round (correct/total,3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}