{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3.6.7 | packaged by conda-forge | (default, Jul  2 2019, 02:18:42) \n",
      "[GCC 7.3.0]\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Import Python libraries  \n",
    "import json\n",
    "from collections import OrderedDict\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import sys\n",
    "from PIL import Image\n",
    "\n",
    "# Import PyTorch libraries\n",
    "import torch\n",
    "from torch import nn\n",
    "from torch import optim\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets, transforms, models\n",
    "\n",
    "# Pretty display for Jupyter notebooks\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "# Print Python version for future reference\n",
    "print(sys.version)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Stats needed to normalize ImageNet images\n",
    "means = [0.485, 0.456, 0.406]\n",
    "std_devs = [0.229, 0.224, 0.225]\n",
    "input_size = 128\n",
    "\n",
    "# Other transforms parameters \n",
    "down_size = 256\n",
    "rotation = 30\n",
    "\n",
    "# Determine batch size for DataLoaders\n",
    "_batch_size = 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: 'data/train/'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-4-c83b8a389ace>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     19\u001b[0m image_datasets = {\n\u001b[1;32m     20\u001b[0m     'train': datasets.ImageFolder('data/train/', \n\u001b[0;32m---> 21\u001b[0;31m                                   transform=data_transforms['train']),\n\u001b[0m\u001b[1;32m     22\u001b[0m     'valid': datasets.ImageFolder('data/val/', \n\u001b[1;32m     23\u001b[0m                                   transform=data_transforms['valid']),\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torchvision/datasets/folder.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, root, transform, target_transform, loader, is_valid_file)\u001b[0m\n\u001b[1;32m    227\u001b[0m                                           \u001b[0mtransform\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtransform\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    228\u001b[0m                                           \u001b[0mtarget_transform\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtarget_transform\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 229\u001b[0;31m                                           is_valid_file=is_valid_file)\n\u001b[0m\u001b[1;32m    230\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimgs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msamples\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torchvision/datasets/folder.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, root, loader, extensions, transform, target_transform, is_valid_file)\u001b[0m\n\u001b[1;32m    106\u001b[0m         super(DatasetFolder, self).__init__(root, transform=transform,\n\u001b[1;32m    107\u001b[0m                                             target_transform=target_transform)\n\u001b[0;32m--> 108\u001b[0;31m         \u001b[0mclasses\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mclass_to_idx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_find_classes\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mroot\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    109\u001b[0m         \u001b[0msamples\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmake_dataset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mroot\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mclass_to_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mextensions\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mis_valid_file\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    110\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msamples\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torchvision/datasets/folder.py\u001b[0m in \u001b[0;36m_find_classes\u001b[0;34m(self, dir)\u001b[0m\n\u001b[1;32m    135\u001b[0m             \u001b[0mNo\u001b[0m \u001b[0;32mclass\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0ma\u001b[0m \u001b[0msubdirectory\u001b[0m \u001b[0mof\u001b[0m \u001b[0manother\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    136\u001b[0m         \"\"\"\n\u001b[0;32m--> 137\u001b[0;31m         \u001b[0mclasses\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0md\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0md\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mscandir\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdir\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0md\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_dir\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    138\u001b[0m         \u001b[0mclasses\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msort\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    139\u001b[0m         \u001b[0mclass_to_idx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0mcls_name\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcls_name\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mclasses\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'data/train/'"
     ]
    }
   ],
   "source": [
    "# Define transforms for the training, validation, and testing sets\n",
    "data_transforms = {\n",
    "    'train': transforms.Compose([transforms.RandomRotation(rotation),\n",
    "                                 transforms.RandomResizedCrop(input_size),\n",
    "                                 transforms.RandomHorizontalFlip(),\n",
    "                                 transforms.ToTensor(),\n",
    "                                 transforms.Normalize(means,std_devs)]),\n",
    "    'valid': transforms.Compose([transforms.Resize(down_size),\n",
    "                                        transforms.CenterCrop(input_size),\n",
    "                                        transforms.ToTensor(),\n",
    "                                        transforms.Normalize(means,std_devs)]),\n",
    "    'test': transforms.Compose([transforms.Resize(down_size),\n",
    "                                        transforms.CenterCrop(input_size),\n",
    "                                        transforms.ToTensor(),\n",
    "                                        transforms.Normalize(means,std_devs)])\n",
    "    }\n",
    "\n",
    "# Build datasets and apply transforms (with ImageFolder)\n",
    "image_datasets = {\n",
    "    'train': datasets.ImageFolder('data/train/', \n",
    "                                  transform=data_transforms['train']),\n",
    "    'valid': datasets.ImageFolder('data/val/', \n",
    "                                  transform=data_transforms['valid']),\n",
    "    'test': datasets.ImageFolder('data/test/', \n",
    "                                 transform=data_transforms['test'])\n",
    "    }\n",
    "\n",
    "# Define dataloaders\n",
    "dataloaders = {\n",
    "    'train': torch.utils.data.DataLoader(image_datasets['train'], \n",
    "                                         batch_size=_batch_size, shuffle=True),\n",
    "    'valid': torch.utils.data.DataLoader(image_datasets['valid'], \n",
    "                                         batch_size=_batch_size),\n",
    "    'test': torch.utils.data.DataLoader(image_datasets['test'], \n",
    "                                        batch_size=_batch_size)\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "env: JOBLIB_TEMP_FOLDER=/datalab\n"
     ]
    }
   ],
   "source": [
    "%env JOBLIB_TEMP_FOLDER=/tmp\n",
    "os.environ[\"JOBLIB_TEMP_FOLDER\"]=\"/imp/\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://download.pytorch.org/models/resnet50-19c8e357.pth\" to /home/jmanoj01/.cache/torch/hub/checkpoints/resnet50-19c8e357.pth\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5559a36315254368a34551e8c4944fbd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0.00/97.8M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_conv = models.resnet50(pretrained=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for param in model.parameters():\n",
    "#     param.requires_grad = False\n",
    "ct = 0\n",
    "for name, child in model_conv.named_children():\n",
    "    for name2, params in child.named_parameters():\n",
    "        params.requires_grad = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "in_features = model_conv.fc.in_features\n",
    "out_categories = 94\n",
    "layer_1 = 1024\n",
    "\n",
    "# Build classifier\n",
    "classifier = nn.Sequential(OrderedDict([\n",
    "    ('fc1', nn.Linear(in_features, layer_1)),\n",
    "    ('relu1', nn.ReLU()),\n",
    "    ('fc2', nn.Linear(layer_1, out_categories)),\n",
    "    ('output', nn.LogSoftmax(dim=1))\n",
    "    ]))\n",
    "\n",
    "# Assign new classifier to VGG-11 architecture \n",
    "model_conv.fc = classifier\n",
    "\n",
    "# ('dropout1', nn.Dropout(0.5)),\n",
    "#     ('fc1', nn.Linear(in_features, layer_1)),\n",
    "#     ('relu1', nn.ReLU()),\n",
    "#     ('dropout2', nn.Dropout(0.5)),\n",
    "#     ('fc2', nn.Linear(layer_1, layer_1 // 2)),\n",
    "#     ('relu2', nn.ReLU()),\n",
    "#     ('dropout3', nn.Dropout(0.5)),"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# loss_function = nn.NLLLoss()\n",
    "loss_function = nn.CrossEntropyLoss()\n",
    "\n",
    "# Instantiate optimization algorithm \n",
    "learning_rate = 0.0001\n",
    "optimizer = optim.Adam(model_conv.fc.parameters(), lr=learning_rate)\n",
    "\n",
    "# Enable CUDA: use GPUs for model computation\n",
    "if torch.cuda.is_available():\n",
    "    model_conv.to('cuda')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Epochs: number of iterations over the entire training dataset\n",
    "epochs = 6\n",
    "# Number of iterations between printing loss and accuracy \n",
    "print_steps = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# %%time\n",
    "\n",
    "# # Initialize steps\n",
    "# step = 0\n",
    "\n",
    "# # Iterate over number of epochs\n",
    "# for e in range(epochs):\n",
    "#     running_loss = 0 \n",
    "\n",
    "#     # Iterate over the entire training dataset\n",
    "#     # one batch per iteration \n",
    "#     for input_images, labels in dataloaders['train']:\n",
    "#         step += 1     \n",
    "        \n",
    "#         # Enable CUDA: use GPUs for model computation\n",
    "#         if torch.cuda.is_available():\n",
    "#             input_images, labels = input_images.to('cuda'), labels.to('cuda')   \n",
    "        \n",
    "#         # Clear the gradients of all optimized tensors\n",
    "#         optimizer.zero_grad()\n",
    "        \n",
    "#         # Forward pass\n",
    "#         outputs = model_conv.forward(input_images)\n",
    "#         loss = loss_function(outputs, labels)\n",
    "        \n",
    "#         # Backward pass\n",
    "#         loss.backward()\n",
    "#         optimizer.step()\n",
    "        \n",
    "#         # Calculate and print running training loss\n",
    "#         running_loss += loss.item()       \n",
    "#         if step % print_steps == 0:\n",
    "#             print(\"Epoch: {}/{}... \".format(e+1, epochs),\n",
    "#                   \"Loss: {:.4f}\".format(running_loss/print_steps))\n",
    "#             running_loss = 0\n",
    "        \n",
    "# print('\\nCount of data batches seen:  {} \\n'.format(step))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.vision.all import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('gold_labels_for_test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>ref_caption</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>test_img_0.jpg</td>\n",
       "      <td>troll</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>test_img_1.jpg</td>\n",
       "      <td>troll</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>test_img_2.jpg</td>\n",
       "      <td>troll</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>test_img_3.jpg</td>\n",
       "      <td>troll</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>test_img_4.jpg</td>\n",
       "      <td>troll</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0     ref_caption  label\n",
       "0           0  test_img_0.jpg  troll\n",
       "1           1  test_img_1.jpg  troll\n",
       "2           2  test_img_2.jpg  troll\n",
       "3           3  test_img_3.jpg  troll\n",
       "4           4  test_img_4.jpg  troll"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(['Unnamed: 0'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['new_label'] = (df['label']=='troll').astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ref_caption</th>\n",
       "      <th>label</th>\n",
       "      <th>new_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test_img_0.jpg</td>\n",
       "      <td>troll</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test_img_1.jpg</td>\n",
       "      <td>troll</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>test_img_2.jpg</td>\n",
       "      <td>troll</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>test_img_3.jpg</td>\n",
       "      <td>troll</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>test_img_4.jpg</td>\n",
       "      <td>troll</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      ref_caption  label  new_label\n",
       "0  test_img_0.jpg  troll          1\n",
       "1  test_img_1.jpg  troll          1\n",
       "2  test_img_2.jpg  troll          1\n",
       "3  test_img_3.jpg  troll          1\n",
       "4  test_img_4.jpg  troll          1"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5922038980509745"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['new_label'].sum()/df.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (<ipython-input-44-2c9247cbbdda>, line 5)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-44-2c9247cbbdda>\"\u001b[0;36m, line \u001b[0;32m5\u001b[0m\n\u001b[0;31m    dataloader = ImageDataLoaders.from_df(path, ,item_tfms=Resize(128, 128), valid='val', test='test', bs=_batch_size, batch_tfms= tfms)\u001b[0m\n\u001b[0m                                                ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
     ]
    }
   ],
   "source": [
    "# data = DataLoaders(dataloaders['train'], dataloaders['valid']).cuda()\n",
    "path = \"meme-classification/data/test_img\"\n",
    "xtra_tfms = Normalize.from_stats(*imagenet_stats)\n",
    "tfms = aug_transforms(pad_mode='zeros', mult=2, min_scale=0.5, xtra_tfms = xtra_tfms)\n",
    "# dataloader = ImageDataLoaders.from_df(path, ,item_tfms=Resize(128, 128), valid='val', test='test', bs=_batch_size, batch_tfms= tfms)\n",
    "testdata = ImageDataLoaders.from_df(df=df, folder=path, item_tfms=Resize(128, 128), label_col=2, bs=_batch_size, batch_tfms= tfms, valid_pct=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = Learner(testdata, model_conv, loss_func=loss_function, opt_func=Adam, metrics=accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.955720</td>\n",
       "      <td>1.313385</td>\n",
       "      <td>0.673184</td>\n",
       "      <td>00:41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.952725</td>\n",
       "      <td>00:03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/fastai/callback/schedule.py:68: UserWarning: This overload of nonzero is deprecated:\n",
      "\tnonzero()\n",
      "Consider using one of the following signatures instead:\n",
      "\tnonzero(*, bool as_tuple) (Triggered internally at  /pytorch/torch/csrc/utils/python_arg_parser.cpp:882.)\n",
      "  idx = (pos >= pcts).nonzero().max()\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-58-4c98b8d209c5>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_one_cycle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m20\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlearning_rate\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/callback/schedule.py\u001b[0m in \u001b[0;36mfit_one_cycle\u001b[0;34m(self, n_epoch, lr_max, div, div_final, pct_start, wd, moms, cbs, reset_opt)\u001b[0m\n\u001b[1;32m    110\u001b[0m     scheds = {'lr': combined_cos(pct_start, lr_max/div, lr_max, lr_max/div_final),\n\u001b[1;32m    111\u001b[0m               'mom': combined_cos(pct_start, *(self.moms if moms is None else moms))}\n\u001b[0;32m--> 112\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcbs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mParamScheduler\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mscheds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0mL\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcbs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreset_opt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mreset_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mwd\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    113\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    114\u001b[0m \u001b[0;31m# Cell\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, n_epoch, lr, wd, cbs, reset_opt)\u001b[0m\n\u001b[1;32m    203\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_hypers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlr\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mlr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    204\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_epoch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mn_epoch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 205\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_fit\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'fit'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelFitException\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_end_cleanup\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    206\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    207\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_end_cleanup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxb\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    152\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    153\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 154\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    155\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    156\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_fit\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    194\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    195\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 196\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'epoch'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelEpochException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    197\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    198\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcbs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreset_opt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    152\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    153\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 154\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    155\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    156\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_epoch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    188\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    189\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 190\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    191\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    192\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_epoch_train\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    180\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    181\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 182\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mall_batches\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'train'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelTrainException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    183\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    184\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mds_idx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdl\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    152\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    153\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 154\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    155\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    156\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mall_batches\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    158\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mall_batches\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    159\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 160\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mo\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mone_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    161\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    162\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mone_batch\u001b[0;34m(self, i, b)\u001b[0m\n\u001b[1;32m    176\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    177\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_split\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 178\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_one_batch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'batch'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelBatchException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    179\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    180\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    152\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    153\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 154\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    155\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    156\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_one_batch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    161\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    162\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 163\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    164\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'after_pred'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    165\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torchvision/models/resnet.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    218\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    219\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 220\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_forward_impl\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    221\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    222\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torchvision/models/resnet.py\u001b[0m in \u001b[0;36m_forward_impl\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    207\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    208\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 209\u001b[0;31m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    210\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer3\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    211\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer4\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    115\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    116\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 117\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    118\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    119\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torchvision/models/resnet.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    106\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrelu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    107\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 108\u001b[0;31m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconv2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    109\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbn2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    110\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrelu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    725\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    726\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 727\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    728\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    729\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/conv.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    421\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    422\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 423\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_conv_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mweight\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    424\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    425\u001b[0m \u001b[0;32mclass\u001b[0m \u001b[0mConv3d\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_ConvNd\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/conv.py\u001b[0m in \u001b[0;36m_conv_forward\u001b[0;34m(self, input, weight)\u001b[0m\n\u001b[1;32m    418\u001b[0m                             _pair(0), self.dilation, self.groups)\n\u001b[1;32m    419\u001b[0m         return F.conv2d(input, weight, self.bias, self.stride,\n\u001b[0;32m--> 420\u001b[0;31m                         self.padding, self.dilation, self.groups)\n\u001b[0m\u001b[1;32m    421\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    422\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/torch_core.py\u001b[0m in \u001b[0;36m__torch_function__\u001b[0;34m(self, func, types, args, kwargs)\u001b[0m\n\u001b[1;32m    315\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    316\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__torch_function__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 317\u001b[0;31m         \u001b[0;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_C\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDisableTorchFunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_convert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkwargs\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__class__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    318\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mret\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mTensorBase\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mret\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_meta\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mas_copy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    319\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mret\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "learn.fit_one_cycle(20, learning_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# learn.show_results()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# learn.show_training_loop()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# learn.recorder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "ename": "RuntimeError",
     "evalue": "Error(s) in loading state_dict for ResNet:\n\tMissing key(s) in state_dict: \"conv1.weight\", \"bn1.weight\", \"bn1.bias\", \"bn1.running_mean\", \"bn1.running_var\", \"layer1.0.conv1.weight\", \"layer1.0.bn1.weight\", \"layer1.0.bn1.bias\", \"layer1.0.bn1.running_mean\", \"layer1.0.bn1.running_var\", \"layer1.0.conv2.weight\", \"layer1.0.bn2.weight\", \"layer1.0.bn2.bias\", \"layer1.0.bn2.running_mean\", \"layer1.0.bn2.running_var\", \"layer1.0.conv3.weight\", \"layer1.0.bn3.weight\", \"layer1.0.bn3.bias\", \"layer1.0.bn3.running_mean\", \"layer1.0.bn3.running_var\", \"layer1.0.downsample.0.weight\", \"layer1.0.downsample.1.weight\", \"layer1.0.downsample.1.bias\", \"layer1.0.downsample.1.running_mean\", \"layer1.0.downsample.1.running_var\", \"layer1.1.conv1.weight\", \"layer1.1.bn1.weight\", \"layer1.1.bn1.bias\", \"layer1.1.bn1.running_mean\", \"layer1.1.bn1.running_var\", \"layer1.1.conv2.weight\", \"layer1.1.bn2.weight\", \"layer1.1.bn2.bias\", \"layer1.1.bn2.running_mean\", \"layer1.1.bn2.running_var\", \"layer1.1.conv3.weight\", \"layer1.1.bn3.weight\", \"layer1.1.bn3.bias\", \"layer1.1.bn3.running_mean\", \"layer1.1.bn3.running_var\", \"layer1.2.conv1.weight\", \"layer1.2.bn1.weight\", \"layer1.2.bn1.bias\", \"layer1.2.bn1.running_mean\", \"layer1.2.bn1.running_var\", \"layer1.2.conv2.weight\", \"layer1.2.bn2.weight\", \"layer1.2.bn2.bias\", \"layer1.2.bn2.running_mean\", \"layer1.2.bn2.running_var\", \"layer1.2.conv3.weight\", \"layer1.2.bn3.weight\", \"layer1.2.bn3.bias\", \"layer1.2.bn3.running_mean\", \"layer1.2.bn3.running_var\", \"layer2.0.conv1.weight\", \"layer2.0.bn1.weight\", \"layer2.0.bn1.bias\", \"layer2.0.bn1.running_mean\", \"layer2.0.bn1.running_var\", \"layer2.0.conv2.weight\", \"layer2.0.bn2.weight\", \"layer2.0.bn2.bias\", \"layer2.0.bn2.running_mean\", \"layer2.0.bn2.running_var\", \"layer2.0.conv3.weight\", \"layer2.0.bn3.weight\", \"layer2.0.bn3.bias\", \"layer2.0.bn3.running_mean\", \"layer2.0.bn3.running_var\", \"layer2.0.downsample.0.weight\", \"layer2.0.downsample.1.weight\", \"layer2.0.downsample.1.bias\", \"layer2.0.downsample.1.running_mean\", \"layer2.0.downsample.1.running_var\", \"layer2.1.conv1.weight\", \"layer2.1.bn1.weight\", \"layer2.1.bn1.bias\", \"layer2.1.bn1.running_mean\", \"layer2.1.bn1.running_var\", \"layer2.1.conv2.weight\", \"layer2.1.bn2.weight\", \"layer2.1.bn2.bias\", \"layer2.1.bn2.running_mean\", \"layer2.1.bn2.running_var\", \"layer2.1.conv3.weight\", \"layer2.1.bn3.weight\", \"layer2.1.bn3.bias\", \"layer2.1.bn3.running_mean\", \"layer2.1.bn3.running_var\", \"layer2.2.conv1.weight\", \"layer2.2.bn1.weight\", \"layer2.2.bn1.bias\", \"layer2.2.bn1.running_mean\", \"layer2.2.bn1.running_var\", \"layer2.2.conv2.weight\", \"layer2.2.bn2.weight\", \"layer2.2.bn2.bias\", \"layer2.2.bn2.running_mean\", \"layer2.2.bn2.running_var\", \"layer2.2.conv3.weight\", \"layer2.2.bn3.weight\", \"layer2.2.bn3.bias\", \"layer2.2.bn3.running_mean\", \"layer2.2.bn3.running_var\", \"layer2.3.conv1.weight\", \"layer2.3.bn1.weight\", \"layer2.3.bn1.bias\", \"layer2.3.bn1.running_mean\", \"layer2.3.bn1.running_var\", \"layer2.3.conv2.weight\", \"layer2.3.bn2.weight\", \"layer2.3.bn2.bias\", \"layer2.3.bn2.running_mean\", \"layer2.3.bn2.running_var\", \"layer2.3.conv3.weight\", \"layer2.3.bn3.weight\", \"layer2.3.bn3.bias\", \"layer2.3.bn3.running_mean\", \"layer2.3.bn3.running_var\", \"layer3.0.conv1.weight\", \"layer3.0.bn1.weight\", \"layer3.0.bn1.bias\", \"layer3.0.bn1.running_mean\", \"layer3.0.bn1.running_var\", \"layer3.0.conv2.weight\", \"layer3.0.bn2.weight\", \"layer3.0.bn2.bias\", \"layer3.0.bn2.running_mean\", \"layer3.0.bn2.running_var\", \"layer3.0.conv3.weight\", \"layer3.0.bn3.weight\", \"layer3.0.bn3.bias\", \"layer3.0.bn3.running_mean\", \"layer3.0.bn3.running_var\", \"layer3.0.downsample.0.weight\", \"layer3.0.downsample.1.weight\", \"layer3.0.downsample.1.bias\", \"layer3.0.downsample.1.running_mean\", \"layer3.0.downsample.1.running_var\", \"layer3.1.conv1.weight\", \"layer3.1.bn1.weight\", \"layer3.1.bn1.bias\", \"layer3.1.bn1.running_mean\", \"layer3.1.bn1.running_var\", \"layer3.1.conv2.weight\", \"layer3.1.bn2.weight\", \"layer3.1.bn2.bias\", \"layer3.1.bn2.running_mean\", \"layer3.1.bn2.running_var\", \"layer3.1.conv3.weight\", \"layer3.1.bn3.weight\", \"layer3.1.bn3.bias\", \"layer3.1.bn3.running_mean\", \"layer3.1.bn3.running_var\", \"layer3.2.conv1.weight\", \"layer3.2.bn1.weight\", \"layer3.2.bn1.bias\", \"layer3.2.bn1.running_mean\", \"layer3.2.bn1.running_var\", \"layer3.2.conv2.weight\", \"layer3.2.bn2.weight\", \"layer3.2.bn2.bias\", \"layer3.2.bn2.running_mean\", \"layer3.2.bn2.running_var\", \"layer3.2.conv3.weight\", \"layer3.2.bn3.weight\", \"layer3.2.bn3.bias\", \"layer3.2.bn3.running_mean\", \"layer3.2.bn3.running_var\", \"layer3.3.conv1.weight\", \"layer3.3.bn1.weight\", \"layer3.3.bn1.bias\", \"layer3.3.bn1.running_mean\", \"layer3.3.bn1.running_var\", \"layer3.3.conv2.weight\", \"layer3.3.bn2.weight\", \"layer3.3.bn2.bias\", \"layer3.3.bn2.running_mean\", \"layer3.3.bn2.running_var\", \"layer3.3.conv3.weight\", \"layer3.3.bn3.weight\", \"layer3.3.bn3.bias\", \"layer3.3.bn3.running_mean\", \"layer3.3.bn3.running_var\", \"layer3.4.conv1.weight\", \"layer3.4.bn1.weight\", \"layer3.4.bn1.bias\", \"layer3.4.bn1.running_mean\", \"layer3.4.bn1.running_var\", \"layer3.4.conv2.weight\", \"layer3.4.bn2.weight\", \"layer3.4.bn2.bias\", \"layer3.4.bn2.running_mean\", \"layer3.4.bn2.running_var\", \"layer3.4.conv3.weight\", \"layer3.4.bn3.weight\", \"layer3.4.bn3.bias\", \"layer3.4.bn3.running_mean\", \"layer3.4.bn3.running_var\", \"layer3.5.conv1.weight\", \"layer3.5.bn1.weight\", \"layer3.5.bn1.bias\", \"layer3.5.bn1.running_mean\", \"layer3.5.bn1.running_var\", \"layer3.5.conv2.weight\", \"layer3.5.bn2.weight\", \"layer3.5.bn2.bias\", \"layer3.5.bn2.running_mean\", \"layer3.5.bn2.running_var\", \"layer3.5.conv3.weight\", \"layer3.5.bn3.weight\", \"layer3.5.bn3.bias\", \"layer3.5.bn3.running_mean\", \"layer3.5.bn3.running_var\", \"layer4.0.conv1.weight\", \"layer4.0.bn1.weight\", \"layer4.0.bn1.bias\", \"layer4.0.bn1.running_mean\", \"layer4.0.bn1.running_var\", \"layer4.0.conv2.weight\", \"layer4.0.bn2.weight\", \"layer4.0.bn2.bias\", \"layer4.0.bn2.running_mean\", \"layer4.0.bn2.running_var\", \"layer4.0.conv3.weight\", \"layer4.0.bn3.weight\", \"layer4.0.bn3.bias\", \"layer4.0.bn3.running_mean\", \"layer4.0.bn3.running_var\", \"layer4.0.downsample.0.weight\", \"layer4.0.downsample.1.weight\", \"layer4.0.downsample.1.bias\", \"layer4.0.downsample.1.running_mean\", \"layer4.0.downsample.1.running_var\", \"layer4.1.conv1.weight\", \"layer4.1.bn1.weight\", \"layer4.1.bn1.bias\", \"layer4.1.bn1.running_mean\", \"layer4.1.bn1.running_var\", \"layer4.1.conv2.weight\", \"layer4.1.bn2.weight\", \"layer4.1.bn2.bias\", \"layer4.1.bn2.running_mean\", \"layer4.1.bn2.running_var\", \"layer4.1.conv3.weight\", \"layer4.1.bn3.weight\", \"layer4.1.bn3.bias\", \"layer4.1.bn3.running_mean\", \"layer4.1.bn3.running_var\", \"layer4.2.conv1.weight\", \"layer4.2.bn1.weight\", \"layer4.2.bn1.bias\", \"layer4.2.bn1.running_mean\", \"layer4.2.bn1.running_var\", \"layer4.2.conv2.weight\", \"layer4.2.bn2.weight\", \"layer4.2.bn2.bias\", \"layer4.2.bn2.running_mean\", \"layer4.2.bn2.running_var\", \"layer4.2.conv3.weight\", \"layer4.2.bn3.weight\", \"layer4.2.bn3.bias\", \"layer4.2.bn3.running_mean\", \"layer4.2.bn3.running_var\", \"fc.fc1.weight\", \"fc.fc1.bias\", \"fc.fc2.weight\", \"fc.fc2.bias\". \n\tUnexpected key(s) in state_dict: \"0.0.weight\", \"0.1.weight\", \"0.1.bias\", \"0.1.running_mean\", \"0.1.running_var\", \"0.1.num_batches_tracked\", \"0.4.0.conv1.weight\", \"0.4.0.bn1.weight\", \"0.4.0.bn1.bias\", \"0.4.0.bn1.running_mean\", \"0.4.0.bn1.running_var\", \"0.4.0.bn1.num_batches_tracked\", \"0.4.0.conv2.weight\", \"0.4.0.bn2.weight\", \"0.4.0.bn2.bias\", \"0.4.0.bn2.running_mean\", \"0.4.0.bn2.running_var\", \"0.4.0.bn2.num_batches_tracked\", \"0.4.0.conv3.weight\", \"0.4.0.bn3.weight\", \"0.4.0.bn3.bias\", \"0.4.0.bn3.running_mean\", \"0.4.0.bn3.running_var\", \"0.4.0.bn3.num_batches_tracked\", \"0.4.0.downsample.0.weight\", \"0.4.0.downsample.1.weight\", \"0.4.0.downsample.1.bias\", \"0.4.0.downsample.1.running_mean\", \"0.4.0.downsample.1.running_var\", \"0.4.0.downsample.1.num_batches_tracked\", \"0.4.1.conv1.weight\", \"0.4.1.bn1.weight\", \"0.4.1.bn1.bias\", \"0.4.1.bn1.running_mean\", \"0.4.1.bn1.running_var\", \"0.4.1.bn1.num_batches_tracked\", \"0.4.1.conv2.weight\", \"0.4.1.bn2.weight\", \"0.4.1.bn2.bias\", \"0.4.1.bn2.running_mean\", \"0.4.1.bn2.running_var\", \"0.4.1.bn2.num_batches_tracked\", \"0.4.1.conv3.weight\", \"0.4.1.bn3.weight\", \"0.4.1.bn3.bias\", \"0.4.1.bn3.running_mean\", \"0.4.1.bn3.running_var\", \"0.4.1.bn3.num_batches_tracked\", \"0.4.2.conv1.weight\", \"0.4.2.bn1.weight\", \"0.4.2.bn1.bias\", \"0.4.2.bn1.running_mean\", \"0.4.2.bn1.running_var\", \"0.4.2.bn1.num_batches_tracked\", \"0.4.2.conv2.weight\", \"0.4.2.bn2.weight\", \"0.4.2.bn2.bias\", \"0.4.2.bn2.running_mean\", \"0.4.2.bn2.running_var\", \"0.4.2.bn2.num_batches_tracked\", \"0.4.2.conv3.weight\", \"0.4.2.bn3.weight\", \"0.4.2.bn3.bias\", \"0.4.2.bn3.running_mean\", \"0.4.2.bn3.running_var\", \"0.4.2.bn3.num_batches_tracked\", \"0.5.0.conv1.weight\", \"0.5.0.bn1.weight\", \"0.5.0.bn1.bias\", \"0.5.0.bn1.running_mean\", \"0.5.0.bn1.running_var\", \"0.5.0.bn1.num_batches_tracked\", \"0.5.0.conv2.weight\", \"0.5.0.bn2.weight\", \"0.5.0.bn2.bias\", \"0.5.0.bn2.running_mean\", \"0.5.0.bn2.running_var\", \"0.5.0.bn2.num_batches_tracked\", \"0.5.0.conv3.weight\", \"0.5.0.bn3.weight\", \"0.5.0.bn3.bias\", \"0.5.0.bn3.running_mean\", \"0.5.0.bn3.running_var\", \"0.5.0.bn3.num_batches_tracked\", \"0.5.0.downsample.0.weight\", \"0.5.0.downsample.1.weight\", \"0.5.0.downsample.1.bias\", \"0.5.0.downsample.1.running_mean\", \"0.5.0.downsample.1.running_var\", \"0.5.0.downsample.1.num_batches_tracked\", \"0.5.1.conv1.weight\", \"0.5.1.bn1.weight\", \"0.5.1.bn1.bias\", \"0.5.1.bn1.running_mean\", \"0.5.1.bn1.running_var\", \"0.5.1.bn1.num_batches_tracked\", \"0.5.1.conv2.weight\", \"0.5.1.bn2.weight\", \"0.5.1.bn2.bias\", \"0.5.1.bn2.running_mean\", \"0.5.1.bn2.running_var\", \"0.5.1.bn2.num_batches_tracked\", \"0.5.1.conv3.weight\", \"0.5.1.bn3.weight\", \"0.5.1.bn3.bias\", \"0.5.1.bn3.running_mean\", \"0.5.1.bn3.running_var\", \"0.5.1.bn3.num_batches_tracked\", \"0.5.2.conv1.weight\", \"0.5.2.bn1.weight\", \"0.5.2.bn1.bias\", \"0.5.2.bn1.running_mean\", \"0.5.2.bn1.running_var\", \"0.5.2.bn1.num_batches_tracked\", \"0.5.2.conv2.weight\", \"0.5.2.bn2.weight\", \"0.5.2.bn2.bias\", \"0.5.2.bn2.running_mean\", \"0.5.2.bn2.running_var\", \"0.5.2.bn2.num_batches_tracked\", \"0.5.2.conv3.weight\", \"0.5.2.bn3.weight\", \"0.5.2.bn3.bias\", \"0.5.2.bn3.running_mean\", \"0.5.2.bn3.running_var\", \"0.5.2.bn3.num_batches_tracked\", \"0.5.3.conv1.weight\", \"0.5.3.bn1.weight\", \"0.5.3.bn1.bias\", \"0.5.3.bn1.running_mean\", \"0.5.3.bn1.running_var\", \"0.5.3.bn1.num_batches_tracked\", \"0.5.3.conv2.weight\", \"0.5.3.bn2.weight\", \"0.5.3.bn2.bias\", \"0.5.3.bn2.running_mean\", \"0.5.3.bn2.running_var\", \"0.5.3.bn2.num_batches_tracked\", \"0.5.3.conv3.weight\", \"0.5.3.bn3.weight\", \"0.5.3.bn3.bias\", \"0.5.3.bn3.running_mean\", \"0.5.3.bn3.running_var\", \"0.5.3.bn3.num_batches_tracked\", \"0.6.0.conv1.weight\", \"0.6.0.bn1.weight\", \"0.6.0.bn1.bias\", \"0.6.0.bn1.running_mean\", \"0.6.0.bn1.running_var\", \"0.6.0.bn1.num_batches_tracked\", \"0.6.0.conv2.weight\", \"0.6.0.bn2.weight\", \"0.6.0.bn2.bias\", \"0.6.0.bn2.running_mean\", \"0.6.0.bn2.running_var\", \"0.6.0.bn2.num_batches_tracked\", \"0.6.0.conv3.weight\", \"0.6.0.bn3.weight\", \"0.6.0.bn3.bias\", \"0.6.0.bn3.running_mean\", \"0.6.0.bn3.running_var\", \"0.6.0.bn3.num_batches_tracked\", \"0.6.0.downsample.0.weight\", \"0.6.0.downsample.1.weight\", \"0.6.0.downsample.1.bias\", \"0.6.0.downsample.1.running_mean\", \"0.6.0.downsample.1.running_var\", \"0.6.0.downsample.1.num_batches_tracked\", \"0.6.1.conv1.weight\", \"0.6.1.bn1.weight\", \"0.6.1.bn1.bias\", \"0.6.1.bn1.running_mean\", \"0.6.1.bn1.running_var\", \"0.6.1.bn1.num_batches_tracked\", \"0.6.1.conv2.weight\", \"0.6.1.bn2.weight\", \"0.6.1.bn2.bias\", \"0.6.1.bn2.running_mean\", \"0.6.1.bn2.running_var\", \"0.6.1.bn2.num_batches_tracked\", \"0.6.1.conv3.weight\", \"0.6.1.bn3.weight\", \"0.6.1.bn3.bias\", \"0.6.1.bn3.running_mean\", \"0.6.1.bn3.running_var\", \"0.6.1.bn3.num_batches_tracked\", \"0.6.2.conv1.weight\", \"0.6.2.bn1.weight\", \"0.6.2.bn1.bias\", \"0.6.2.bn1.running_mean\", \"0.6.2.bn1.running_var\", \"0.6.2.bn1.num_batches_tracked\", \"0.6.2.conv2.weight\", \"0.6.2.bn2.weight\", \"0.6.2.bn2.bias\", \"0.6.2.bn2.running_mean\", \"0.6.2.bn2.running_var\", \"0.6.2.bn2.num_batches_tracked\", \"0.6.2.conv3.weight\", \"0.6.2.bn3.weight\", \"0.6.2.bn3.bias\", \"0.6.2.bn3.running_mean\", \"0.6.2.bn3.running_var\", \"0.6.2.bn3.num_batches_tracked\", \"0.6.3.conv1.weight\", \"0.6.3.bn1.weight\", \"0.6.3.bn1.bias\", \"0.6.3.bn1.running_mean\", \"0.6.3.bn1.running_var\", \"0.6.3.bn1.num_batches_tracked\", \"0.6.3.conv2.weight\", \"0.6.3.bn2.weight\", \"0.6.3.bn2.bias\", \"0.6.3.bn2.running_mean\", \"0.6.3.bn2.running_var\", \"0.6.3.bn2.num_batches_tracked\", \"0.6.3.conv3.weight\", \"0.6.3.bn3.weight\", \"0.6.3.bn3.bias\", \"0.6.3.bn3.running_mean\", \"0.6.3.bn3.running_var\", \"0.6.3.bn3.num_batches_tracked\", \"0.6.4.conv1.weight\", \"0.6.4.bn1.weight\", \"0.6.4.bn1.bias\", \"0.6.4.bn1.running_mean\", \"0.6.4.bn1.running_var\", \"0.6.4.bn1.num_batches_tracked\", \"0.6.4.conv2.weight\", \"0.6.4.bn2.weight\", \"0.6.4.bn2.bias\", \"0.6.4.bn2.running_mean\", \"0.6.4.bn2.running_var\", \"0.6.4.bn2.num_batches_tracked\", \"0.6.4.conv3.weight\", \"0.6.4.bn3.weight\", \"0.6.4.bn3.bias\", \"0.6.4.bn3.running_mean\", \"0.6.4.bn3.running_var\", \"0.6.4.bn3.num_batches_tracked\", \"0.6.5.conv1.weight\", \"0.6.5.bn1.weight\", \"0.6.5.bn1.bias\", \"0.6.5.bn1.running_mean\", \"0.6.5.bn1.running_var\", \"0.6.5.bn1.num_batches_tracked\", \"0.6.5.conv2.weight\", \"0.6.5.bn2.weight\", \"0.6.5.bn2.bias\", \"0.6.5.bn2.running_mean\", \"0.6.5.bn2.running_var\", \"0.6.5.bn2.num_batches_tracked\", \"0.6.5.conv3.weight\", \"0.6.5.bn3.weight\", \"0.6.5.bn3.bias\", \"0.6.5.bn3.running_mean\", \"0.6.5.bn3.running_var\", \"0.6.5.bn3.num_batches_tracked\", \"0.7.0.conv1.weight\", \"0.7.0.bn1.weight\", \"0.7.0.bn1.bias\", \"0.7.0.bn1.running_mean\", \"0.7.0.bn1.running_var\", \"0.7.0.bn1.num_batches_tracked\", \"0.7.0.conv2.weight\", \"0.7.0.bn2.weight\", \"0.7.0.bn2.bias\", \"0.7.0.bn2.running_mean\", \"0.7.0.bn2.running_var\", \"0.7.0.bn2.num_batches_tracked\", \"0.7.0.conv3.weight\", \"0.7.0.bn3.weight\", \"0.7.0.bn3.bias\", \"0.7.0.bn3.running_mean\", \"0.7.0.bn3.running_var\", \"0.7.0.bn3.num_batches_tracked\", \"0.7.0.downsample.0.weight\", \"0.7.0.downsample.1.weight\", \"0.7.0.downsample.1.bias\", \"0.7.0.downsample.1.running_mean\", \"0.7.0.downsample.1.running_var\", \"0.7.0.downsample.1.num_batches_tracked\", \"0.7.1.conv1.weight\", \"0.7.1.bn1.weight\", \"0.7.1.bn1.bias\", \"0.7.1.bn1.running_mean\", \"0.7.1.bn1.running_var\", \"0.7.1.bn1.num_batches_tracked\", \"0.7.1.conv2.weight\", \"0.7.1.bn2.weight\", \"0.7.1.bn2.bias\", \"0.7.1.bn2.running_mean\", \"0.7.1.bn2.running_var\", \"0.7.1.bn2.num_batches_tracked\", \"0.7.1.conv3.weight\", \"0.7.1.bn3.weight\", \"0.7.1.bn3.bias\", \"0.7.1.bn3.running_mean\", \"0.7.1.bn3.running_var\", \"0.7.1.bn3.num_batches_tracked\", \"0.7.2.conv1.weight\", \"0.7.2.bn1.weight\", \"0.7.2.bn1.bias\", \"0.7.2.bn1.running_mean\", \"0.7.2.bn1.running_var\", \"0.7.2.bn1.num_batches_tracked\", \"0.7.2.conv2.weight\", \"0.7.2.bn2.weight\", \"0.7.2.bn2.bias\", \"0.7.2.bn2.running_mean\", \"0.7.2.bn2.running_var\", \"0.7.2.bn2.num_batches_tracked\", \"0.7.2.conv3.weight\", \"0.7.2.bn3.weight\", \"0.7.2.bn3.bias\", \"0.7.2.bn3.running_mean\", \"0.7.2.bn3.running_var\", \"0.7.2.bn3.num_batches_tracked\", \"1.2.weight\", \"1.2.bias\", \"1.2.running_mean\", \"1.2.running_var\", \"1.2.num_batches_tracked\", \"1.4.weight\", \"1.6.weight\", \"1.6.bias\", \"1.6.running_mean\", \"1.6.running_var\", \"1.6.num_batches_tracked\", \"1.8.weight\". ",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-43-f629e28d57bf>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# learn.save('dragonfly-netV2-1')\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'memeclassifier-1'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mload\u001b[0;34m(self, file, device, **kwargs)\u001b[0m\n\u001b[1;32m    347\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_opt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    348\u001b[0m     \u001b[0mfile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjoin_path_file\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mext\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'.pth'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 349\u001b[0;31m     \u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    350\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    351\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(file, model, opt, with_opt, device, strict)\u001b[0m\n\u001b[1;32m     49\u001b[0m     \u001b[0mhasopt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstate\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m==\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'model'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'opt'\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     50\u001b[0m     \u001b[0mmodel_state\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstate\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'model'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mhasopt\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mstate\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 51\u001b[0;31m     \u001b[0mget_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_state_dict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_state\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstrict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstrict\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     52\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhasopt\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mwith_opt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload_state_dict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstate\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'opt'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36mload_state_dict\u001b[0;34m(self, state_dict, strict)\u001b[0m\n\u001b[1;32m   1050\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0merror_msgs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1051\u001b[0m             raise RuntimeError('Error(s) in loading state_dict for {}:\\n\\t{}'.format(\n\u001b[0;32m-> 1052\u001b[0;31m                                self.__class__.__name__, \"\\n\\t\".join(error_msgs)))\n\u001b[0m\u001b[1;32m   1053\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0m_IncompatibleKeys\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmissing_keys\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0munexpected_keys\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1054\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mRuntimeError\u001b[0m: Error(s) in loading state_dict for ResNet:\n\tMissing key(s) in state_dict: \"conv1.weight\", \"bn1.weight\", \"bn1.bias\", \"bn1.running_mean\", \"bn1.running_var\", \"layer1.0.conv1.weight\", \"layer1.0.bn1.weight\", \"layer1.0.bn1.bias\", \"layer1.0.bn1.running_mean\", \"layer1.0.bn1.running_var\", \"layer1.0.conv2.weight\", \"layer1.0.bn2.weight\", \"layer1.0.bn2.bias\", \"layer1.0.bn2.running_mean\", \"layer1.0.bn2.running_var\", \"layer1.0.conv3.weight\", \"layer1.0.bn3.weight\", \"layer1.0.bn3.bias\", \"layer1.0.bn3.running_mean\", \"layer1.0.bn3.running_var\", \"layer1.0.downsample.0.weight\", \"layer1.0.downsample.1.weight\", \"layer1.0.downsample.1.bias\", \"layer1.0.downsample.1.running_mean\", \"layer1.0.downsample.1.running_var\", \"layer1.1.conv1.weight\", \"layer1.1.bn1.weight\", \"layer1.1.bn1.bias\", \"layer1.1.bn1.running_mean\", \"layer1.1.bn1.running_var\", \"layer1.1.conv2.weight\", \"layer1.1.bn2.weight\", \"layer1.1.bn2.bias\", \"layer1.1.bn2.running_mean\", \"layer1.1.bn2.running_var\", \"layer1.1.conv3.weight\", \"layer1.1.bn3.weight\", \"layer1.1.bn3.bias\", \"layer1.1.bn3.running_mean\", \"layer1.1.bn3.running_var\", \"layer1.2.conv1.weight\", \"layer1.2.bn1.weight\", \"layer1.2.bn1.bias\", \"layer1.2.bn1.running_mean\", \"layer1.2.bn1.running_var\", \"layer1.2.conv2.weight\", \"layer1.2.bn2.weight\", \"layer1.2.bn2.bias\", \"layer1.2.bn2.running_mean\", \"layer1.2.bn2.running_var\", \"layer1.2.conv3.weight\", \"layer1.2.bn3.weight\", \"layer1.2.bn3.bias\", \"layer1.2.bn3.running_mean\", \"layer1.2.bn3.running_var\", \"layer2.0.conv1.weight\", \"layer2.0.bn1.weight\", \"layer2.0.bn1.bias\", \"layer2.0.bn1.running_mean\", \"layer2.0.bn1.running_var\", \"layer2.0.conv2.weight\", \"layer2.0.bn2.weight\", \"layer2.0.bn2.bias\", \"layer2.0.bn2.running_mean\", \"layer2.0.bn2.running_var\", \"layer2.0.conv3.weight\", \"layer2.0.bn3.weight\", \"layer2.0.bn3.bias\", \"layer2.0.bn3.running_mean\", \"layer2.0.bn3.running_var\", \"layer2.0.downsample.0.weight\", \"layer2.0.downsample.1.weight\", \"layer2.0.downsample.1.bias\", \"layer2.0.downsample.1.running_mean\", \"layer2.0.downsample.1.running_var\", \"layer2.1.conv1.weight\", \"layer2.1.bn1.weight\", \"layer2.1.bn1.bias\", \"layer2.1.bn1.running_mean\", \"layer2.1.bn1.running_var\", \"layer2.1.conv2.weight\", \"layer2.1.bn2.weight\", \"layer2.1.bn2.bias\", \"layer2.1.bn2.running_mean\", \"layer2.1.bn2.running_var\", \"layer2.1.conv3.weight\", \"layer2.1.bn3.weight\", \"layer2.1.bn3.bias\", \"layer2.1.bn3.running_mean\", \"layer2.1.bn3.running_var\", \"layer2.2.conv1.weight\", \"layer2.2.bn1.weight\", \"layer2.2.bn1.bias\", \"layer2.2.bn1.running_mean\", \"layer2.2.bn1.running_var\", \"layer2.2.conv2.weight\", \"layer2.2.bn2.weight\", \"layer2.2.bn2.bias\", \"layer2.2.bn2.running_mean\", \"layer2.2.bn2.running_var\", \"layer2.2.conv3.weight\", \"layer2.2.bn3.weight\", \"layer2.2.bn3.bias\", \"layer2.2.bn3.running_mean\", \"layer2.2.bn3.running_var\", \"layer2.3.conv1.weight\", \"layer2.3.bn1.weight\", \"layer2.3.bn1.bias\", \"layer2.3.bn1.running_mean\", \"layer2.3.bn1.running_var\", \"layer2.3.conv2.weight\", \"layer2.3.bn2.weight\", \"layer2.3.bn2.bias\", \"layer2.3.bn2.running_mean\", \"layer2.3.bn2.running_var\", \"layer2.3.conv3.weight\", \"layer2.3.bn3.weight\", \"layer2.3.bn3.bias\", \"layer2.3.bn3.running_mean\", \"layer2.3.bn3.running_var\", \"layer3.0.conv1.weight\", \"layer3.0.bn1.weight\", \"layer3.0.bn1.bias\", \"layer3.0.bn1.running_mean\", \"layer3.0.bn1.running_var\", \"layer3.0.conv2.weight\", \"layer3.0.bn2.weight\", \"layer3.0.bn2.bias\", \"layer3.0.bn2.running_mean\", \"layer3.0.bn2.running_var\", \"layer3.0.conv3.weight\", \"layer3.0.bn3.weight\", \"layer3.0.bn3.bias\", \"layer3.0.bn3.running_mean\", \"layer3.0.bn3.running_var\", \"layer3.0.downsample.0.weight\", \"layer3.0.downsample.1.weight\", \"layer3.0.downsample.1.bias\", \"layer3.0.downsample.1.running_mean\", \"layer3.0.downsample.1.running_var\", \"layer3.1.conv1.weight\", \"layer3.1.bn1.weight\", \"layer3.1.bn1.bias\", \"layer3.1.bn1.running_mean\", \"layer3.1.bn1.running_var\", \"layer3.1.conv2.weight\", \"layer3.1.bn2.weight\", \"layer3.1.bn2.bias\", \"layer3.1.bn2.running_mean\", \"layer3.1.bn2.running_var\", \"layer3.1.conv3.weight\", \"layer3.1.bn3.weight\", \"layer3.1.bn3.bias\", \"layer3.1.bn3.running_mean\", \"layer3.1.bn3.running_var\", \"layer3.2.conv1.weight\", \"layer3.2.bn1.weight\", \"layer3.2.bn1.bias\", \"layer3.2.bn1.running_mean\", \"layer3.2.bn1.running_var\", \"layer3.2.conv2.weight\", \"layer3.2.bn2.weight\", \"layer3.2.bn2.bias\", \"layer3.2.bn2.running_mean\", \"layer3.2.bn2.running_var\", \"layer3.2.conv3.weight\", \"layer3.2.bn3.weight\", \"layer3.2.bn3.bias\", \"layer3.2.bn3.running_mean\", \"layer3.2.bn3.running_var\", \"layer3.3.conv1.weight\", \"layer3.3.bn1.weight\", \"layer3.3.bn1.bias\", \"layer3.3.bn1.running_mean\", \"layer3.3.bn1.running_var\", \"layer3.3.conv2.weight\", \"layer3.3.bn2.weight\", \"layer3.3.bn2.bias\", \"layer3.3.bn2.running_mean\", \"layer3.3.bn2.running_var\", \"layer3.3.conv3.weight\", \"layer3.3.bn3.weight\", \"layer3.3.bn3.bias\", \"layer3.3.bn3.running_mean\", \"layer3.3.bn3.running_var\", \"layer3.4.conv1.weight\", \"layer3.4.bn1.weight\", \"layer3.4.bn1.bias\", \"layer3.4.bn1.running_mean\", \"layer3.4.bn1.running_var\", \"layer3.4.conv2.weight\", \"layer3.4.bn2.weight\", \"layer3.4.bn2.bias\", \"layer3.4.bn2.running_mean\", \"layer3.4.bn2.running_var\", \"layer3.4.conv3.weight\", \"layer3.4.bn3.weight\", \"layer3.4.bn3.bias\", \"layer3.4.bn3.running_mean\", \"layer3.4.bn3.running_var\", \"layer3.5.conv1.weight\", \"layer3.5.bn1.weight\", \"layer3.5.bn1.bias\", \"layer3.5.bn1.running_mean\", \"layer3.5.bn1.running_var\", \"layer3.5.conv2.weight\", \"layer3.5.bn2.weight\", \"layer3.5.bn2.bias\", \"layer3.5.bn2.running_mean\", \"layer3.5.bn2.running_var\", \"layer3.5.conv3.weight\", \"layer3.5.bn3.weight\", \"layer3.5.bn3.bias\", \"layer3.5.bn3.running_mean\", \"layer3.5.bn3.running_var\", \"layer4.0.conv1.weight\", \"layer4.0.bn1.weight\", \"layer4.0.bn1.bias\", \"layer4.0.bn1.running_mean\", \"layer4.0.bn1.running_var\", \"layer4.0.conv2.weight\", \"layer4.0.bn2.weight\", \"layer4.0.bn2.bias\", \"layer4.0.bn2.running_mean\", \"layer4.0.bn2.running_var\", \"layer4.0.conv3.weight\", \"layer4.0.bn3.weight\", \"layer4.0.bn3.bias\", \"layer4.0.bn3.running_mean\", \"layer4.0.bn3.running_var\", \"layer4.0.downsample.0.weight\", \"layer4.0.downsample.1.weight\", \"layer4.0.downsample.1.bias\", \"layer4.0.downsample.1.running_mean\", \"layer4.0.downsample.1.running_var\", \"layer4.1.conv1.weight\", \"layer4.1.bn1.weight\", \"layer4.1.bn1.bias\", \"layer4.1.bn1.running_mean\", \"layer4.1.bn1.running_var\", \"layer4.1.conv2.weight\", \"layer4.1.bn2.weight\", \"layer4.1.bn2.bias\", \"layer4.1.bn2.running_mean\", \"layer4.1.bn2.running_var\", \"layer4.1.conv3.weight\", \"layer4.1.bn3.weight\", \"layer4.1.bn3.bias\", \"layer4.1.bn3.running_mean\", \"layer4.1.bn3.running_var\", \"layer4.2.conv1.weight\", \"layer4.2.bn1.weight\", \"layer4.2.bn1.bias\", \"layer4.2.bn1.running_mean\", \"layer4.2.bn1.running_var\", \"layer4.2.conv2.weight\", \"layer4.2.bn2.weight\", \"layer4.2.bn2.bias\", \"layer4.2.bn2.running_mean\", \"layer4.2.bn2.running_var\", \"layer4.2.conv3.weight\", \"layer4.2.bn3.weight\", \"layer4.2.bn3.bias\", \"layer4.2.bn3.running_mean\", \"layer4.2.bn3.running_var\", \"fc.fc1.weight\", \"fc.fc1.bias\", \"fc.fc2.weight\", \"fc.fc2.bias\". \n\tUnexpected key(s) in state_dict: \"0.0.weight\", \"0.1.weight\", \"0.1.bias\", \"0.1.running_mean\", \"0.1.running_var\", \"0.1.num_batches_tracked\", \"0.4.0.conv1.weight\", \"0.4.0.bn1.weight\", \"0.4.0.bn1.bias\", \"0.4.0.bn1.running_mean\", \"0.4.0.bn1.running_var\", \"0.4.0.bn1.num_batches_tracked\", \"0.4.0.conv2.weight\", \"0.4.0.bn2.weight\", \"0.4.0.bn2.bias\", \"0.4.0.bn2.running_mean\", \"0.4.0.bn2.running_var\", \"0.4.0.bn2.num_batches_tracked\", \"0.4.0.conv3.weight\", \"0.4.0.bn3.weight\", \"0.4.0.bn3.bias\", \"0.4.0.bn3.running_mean\", \"0.4.0.bn3.running_var\", \"0.4.0.bn3.num_batches_tracked\", \"0.4.0.downsample.0.weight\", \"0.4.0.downsample.1.weight\", \"0.4.0.downsample.1.bias\", \"0.4.0.downsample.1.running_mean\", \"0.4.0.downsample.1.running_var\", \"0.4.0.downsample.1.num_batches_tracked\", \"0.4.1.conv1.weight\", \"0.4.1.bn1.weight\", \"0.4.1.bn1.bias\", \"0.4.1.bn1.running_mean\", \"0.4.1.bn1.running_var\", \"0.4.1.bn1.num_batches_tracked\", \"0.4.1.conv2.weight\", \"0.4.1.bn2.weight\", \"0.4.1.bn2.bias\", \"0.4.1.bn2.running_mean\", \"0.4.1.bn2.running_var\", \"0.4.1.bn2.num_batches_tracked\", \"0.4.1.conv3.weight\", \"0.4.1.bn3.weight\", \"0.4.1.bn3.bias\", \"0.4.1.bn3.running_mean\", \"0.4.1.bn3.running_var\", \"0.4.1.bn3.num_batches_tracked\", \"0.4.2.conv1.weight\", \"0.4.2.bn1.weight\", \"0.4.2.bn1.bias\", \"0.4.2.bn1.running_mean\", \"0.4.2.bn1.running_var\", \"0.4.2.bn1.num_batches_tracked\", \"0.4.2.conv2.weight\", \"0.4.2.bn2.weight\", \"0.4.2.bn2.bias\", \"0.4.2.bn2.running_mean\", \"0.4.2.bn2.running_var\", \"0.4.2.bn2.num_batches_tracked\", \"0.4.2.conv3.weight\", \"0.4.2.bn3.weight\", \"0.4.2.bn3.bias\", \"0.4.2.bn3.running_mean\", \"0.4.2.bn3.running_var\", \"0.4.2.bn3.num_batches_tracked\", \"0.5.0.conv1.weight\", \"0.5.0.bn1.weight\", \"0.5.0.bn1.bias\", \"0.5.0.bn1.running_mean\", \"0.5.0.bn1.running_var\", \"0.5.0.bn1.num_batches_tracked\", \"0.5.0.conv2.weight\", \"0.5.0.bn2.weight\", \"0.5.0.bn2.bias\", \"0.5.0.bn2.running_mean\", \"0.5.0.bn2.running_var\", \"0.5.0.bn2.num_batches_tracked\", \"0.5.0.conv3.weight\", \"0.5.0.bn3.weight\", \"0.5.0.bn3.bias\", \"0.5.0.bn3.running_mean\", \"0.5.0.bn3.running_var\", \"0.5.0.bn3.num_batches_tracked\", \"0.5.0.downsample.0.weight\", \"0.5.0.downsample.1.weight\", \"0.5.0.downsample.1.bias\", \"0.5.0.downsample.1.running_mean\", \"0.5.0.downsample.1.running_var\", \"0.5.0.downsample.1.num_batches_tracked\", \"0.5.1.conv1.weight\", \"0.5.1.bn1.weight\", \"0.5.1.bn1.bias\", \"0.5.1.bn1.running_mean\", \"0.5.1.bn1.running_var\", \"0.5.1.bn1.num_batches_tracked\", \"0.5.1.conv2.weight\", \"0.5.1.bn2.weight\", \"0.5.1.bn2.bias\", \"0.5.1.bn2.running_mean\", \"0.5.1.bn2.running_var\", \"0.5.1.bn2.num_batches_tracked\", \"0.5.1.conv3.weight\", \"0.5.1.bn3.weight\", \"0.5.1.bn3.bias\", \"0.5.1.bn3.running_mean\", \"0.5.1.bn3.running_var\", \"0.5.1.bn3.num_batches_tracked\", \"0.5.2.conv1.weight\", \"0.5.2.bn1.weight\", \"0.5.2.bn1.bias\", \"0.5.2.bn1.running_mean\", \"0.5.2.bn1.running_var\", \"0.5.2.bn1.num_batches_tracked\", \"0.5.2.conv2.weight\", \"0.5.2.bn2.weight\", \"0.5.2.bn2.bias\", \"0.5.2.bn2.running_mean\", \"0.5.2.bn2.running_var\", \"0.5.2.bn2.num_batches_tracked\", \"0.5.2.conv3.weight\", \"0.5.2.bn3.weight\", \"0.5.2.bn3.bias\", \"0.5.2.bn3.running_mean\", \"0.5.2.bn3.running_var\", \"0.5.2.bn3.num_batches_tracked\", \"0.5.3.conv1.weight\", \"0.5.3.bn1.weight\", \"0.5.3.bn1.bias\", \"0.5.3.bn1.running_mean\", \"0.5.3.bn1.running_var\", \"0.5.3.bn1.num_batches_tracked\", \"0.5.3.conv2.weight\", \"0.5.3.bn2.weight\", \"0.5.3.bn2.bias\", \"0.5.3.bn2.running_mean\", \"0.5.3.bn2.running_var\", \"0.5.3.bn2.num_batches_tracked\", \"0.5.3.conv3.weight\", \"0.5.3.bn3.weight\", \"0.5.3.bn3.bias\", \"0.5.3.bn3.running_mean\", \"0.5.3.bn3.running_var\", \"0.5.3.bn3.num_batches_tracked\", \"0.6.0.conv1.weight\", \"0.6.0.bn1.weight\", \"0.6.0.bn1.bias\", \"0.6.0.bn1.running_mean\", \"0.6.0.bn1.running_var\", \"0.6.0.bn1.num_batches_tracked\", \"0.6.0.conv2.weight\", \"0.6.0.bn2.weight\", \"0.6.0.bn2.bias\", \"0.6.0.bn2.running_mean\", \"0.6.0.bn2.running_var\", \"0.6.0.bn2.num_batches_tracked\", \"0.6.0.conv3.weight\", \"0.6.0.bn3.weight\", \"0.6.0.bn3.bias\", \"0.6.0.bn3.running_mean\", \"0.6.0.bn3.running_var\", \"0.6.0.bn3.num_batches_tracked\", \"0.6.0.downsample.0.weight\", \"0.6.0.downsample.1.weight\", \"0.6.0.downsample.1.bias\", \"0.6.0.downsample.1.running_mean\", \"0.6.0.downsample.1.running_var\", \"0.6.0.downsample.1.num_batches_tracked\", \"0.6.1.conv1.weight\", \"0.6.1.bn1.weight\", \"0.6.1.bn1.bias\", \"0.6.1.bn1.running_mean\", \"0.6.1.bn1.running_var\", \"0.6.1.bn1.num_batches_tracked\", \"0.6.1.conv2.weight\", \"0.6.1.bn2.weight\", \"0.6.1.bn2.bias\", \"0.6.1.bn2.running_mean\", \"0.6.1.bn2.running_var\", \"0.6.1.bn2.num_batches_tracked\", \"0.6.1.conv3.weight\", \"0.6.1.bn3.weight\", \"0.6.1.bn3.bias\", \"0.6.1.bn3.running_mean\", \"0.6.1.bn3.running_var\", \"0.6.1.bn3.num_batches_tracked\", \"0.6.2.conv1.weight\", \"0.6.2.bn1.weight\", \"0.6.2.bn1.bias\", \"0.6.2.bn1.running_mean\", \"0.6.2.bn1.running_var\", \"0.6.2.bn1.num_batches_tracked\", \"0.6.2.conv2.weight\", \"0.6.2.bn2.weight\", \"0.6.2.bn2.bias\", \"0.6.2.bn2.running_mean\", \"0.6.2.bn2.running_var\", \"0.6.2.bn2.num_batches_tracked\", \"0.6.2.conv3.weight\", \"0.6.2.bn3.weight\", \"0.6.2.bn3.bias\", \"0.6.2.bn3.running_mean\", \"0.6.2.bn3.running_var\", \"0.6.2.bn3.num_batches_tracked\", \"0.6.3.conv1.weight\", \"0.6.3.bn1.weight\", \"0.6.3.bn1.bias\", \"0.6.3.bn1.running_mean\", \"0.6.3.bn1.running_var\", \"0.6.3.bn1.num_batches_tracked\", \"0.6.3.conv2.weight\", \"0.6.3.bn2.weight\", \"0.6.3.bn2.bias\", \"0.6.3.bn2.running_mean\", \"0.6.3.bn2.running_var\", \"0.6.3.bn2.num_batches_tracked\", \"0.6.3.conv3.weight\", \"0.6.3.bn3.weight\", \"0.6.3.bn3.bias\", \"0.6.3.bn3.running_mean\", \"0.6.3.bn3.running_var\", \"0.6.3.bn3.num_batches_tracked\", \"0.6.4.conv1.weight\", \"0.6.4.bn1.weight\", \"0.6.4.bn1.bias\", \"0.6.4.bn1.running_mean\", \"0.6.4.bn1.running_var\", \"0.6.4.bn1.num_batches_tracked\", \"0.6.4.conv2.weight\", \"0.6.4.bn2.weight\", \"0.6.4.bn2.bias\", \"0.6.4.bn2.running_mean\", \"0.6.4.bn2.running_var\", \"0.6.4.bn2.num_batches_tracked\", \"0.6.4.conv3.weight\", \"0.6.4.bn3.weight\", \"0.6.4.bn3.bias\", \"0.6.4.bn3.running_mean\", \"0.6.4.bn3.running_var\", \"0.6.4.bn3.num_batches_tracked\", \"0.6.5.conv1.weight\", \"0.6.5.bn1.weight\", \"0.6.5.bn1.bias\", \"0.6.5.bn1.running_mean\", \"0.6.5.bn1.running_var\", \"0.6.5.bn1.num_batches_tracked\", \"0.6.5.conv2.weight\", \"0.6.5.bn2.weight\", \"0.6.5.bn2.bias\", \"0.6.5.bn2.running_mean\", \"0.6.5.bn2.running_var\", \"0.6.5.bn2.num_batches_tracked\", \"0.6.5.conv3.weight\", \"0.6.5.bn3.weight\", \"0.6.5.bn3.bias\", \"0.6.5.bn3.running_mean\", \"0.6.5.bn3.running_var\", \"0.6.5.bn3.num_batches_tracked\", \"0.7.0.conv1.weight\", \"0.7.0.bn1.weight\", \"0.7.0.bn1.bias\", \"0.7.0.bn1.running_mean\", \"0.7.0.bn1.running_var\", \"0.7.0.bn1.num_batches_tracked\", \"0.7.0.conv2.weight\", \"0.7.0.bn2.weight\", \"0.7.0.bn2.bias\", \"0.7.0.bn2.running_mean\", \"0.7.0.bn2.running_var\", \"0.7.0.bn2.num_batches_tracked\", \"0.7.0.conv3.weight\", \"0.7.0.bn3.weight\", \"0.7.0.bn3.bias\", \"0.7.0.bn3.running_mean\", \"0.7.0.bn3.running_var\", \"0.7.0.bn3.num_batches_tracked\", \"0.7.0.downsample.0.weight\", \"0.7.0.downsample.1.weight\", \"0.7.0.downsample.1.bias\", \"0.7.0.downsample.1.running_mean\", \"0.7.0.downsample.1.running_var\", \"0.7.0.downsample.1.num_batches_tracked\", \"0.7.1.conv1.weight\", \"0.7.1.bn1.weight\", \"0.7.1.bn1.bias\", \"0.7.1.bn1.running_mean\", \"0.7.1.bn1.running_var\", \"0.7.1.bn1.num_batches_tracked\", \"0.7.1.conv2.weight\", \"0.7.1.bn2.weight\", \"0.7.1.bn2.bias\", \"0.7.1.bn2.running_mean\", \"0.7.1.bn2.running_var\", \"0.7.1.bn2.num_batches_tracked\", \"0.7.1.conv3.weight\", \"0.7.1.bn3.weight\", \"0.7.1.bn3.bias\", \"0.7.1.bn3.running_mean\", \"0.7.1.bn3.running_var\", \"0.7.1.bn3.num_batches_tracked\", \"0.7.2.conv1.weight\", \"0.7.2.bn1.weight\", \"0.7.2.bn1.bias\", \"0.7.2.bn1.running_mean\", \"0.7.2.bn1.running_var\", \"0.7.2.bn1.num_batches_tracked\", \"0.7.2.conv2.weight\", \"0.7.2.bn2.weight\", \"0.7.2.bn2.bias\", \"0.7.2.bn2.running_mean\", \"0.7.2.bn2.running_var\", \"0.7.2.bn2.num_batches_tracked\", \"0.7.2.conv3.weight\", \"0.7.2.bn3.weight\", \"0.7.2.bn3.bias\", \"0.7.2.bn3.running_mean\", \"0.7.2.bn3.running_var\", \"0.7.2.bn3.num_batches_tracked\", \"1.2.weight\", \"1.2.bias\", \"1.2.running_mean\", \"1.2.running_var\", \"1.2.num_batches_tracked\", \"1.4.weight\", \"1.6.weight\", \"1.6.bias\", \"1.6.running_mean\", \"1.6.running_var\", \"1.6.num_batches_tracked\", \"1.8.weight\". "
     ]
    }
   ],
   "source": [
    "# learn.save('dragonfly-netV2-1')\n",
    "learn.load('memeclassifier-1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sequential(\n",
       "  (0): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
       "    (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (2): ReLU(inplace=True)\n",
       "    (3): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
       "    (4): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "    )\n",
       "    (5): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "          (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (3): Bottleneck(\n",
       "        (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "    )\n",
       "    (6): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "          (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (3): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (4): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (5): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "    )\n",
       "    (7): Sequential(\n",
       "      (0): Bottleneck(\n",
       "        (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "        (downsample): Sequential(\n",
       "          (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
       "          (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (1): Bottleneck(\n",
       "        (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "      (2): Bottleneck(\n",
       "        (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "        (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (relu): ReLU(inplace=True)\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (1): Sequential(\n",
       "    (0): AdaptiveConcatPool2d(\n",
       "      (ap): AdaptiveAvgPool2d(output_size=1)\n",
       "      (mp): AdaptiveMaxPool2d(output_size=1)\n",
       "    )\n",
       "    (1): Flatten(full=False)\n",
       "    (2): BatchNorm1d(4096, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (3): Dropout(p=0.25, inplace=False)\n",
       "    (4): Linear(in_features=4096, out_features=512, bias=False)\n",
       "    (5): ReLU(inplace=True)\n",
       "    (6): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (7): Dropout(p=0.5, inplace=False)\n",
       "    (8): Linear(in_features=512, out_features=2, bias=False)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# labels_set = []\n",
    "# preds_set = []\n",
    "# for idx, data in enumerate(dataloaders['test']):\n",
    "#     # get the inputs\n",
    "#     model.eval()\n",
    "#     print(\"batch no. : \", idx)\n",
    "#     inputs, labels = data\n",
    "#     labels_set += labels.tolist()\n",
    "#     inputs, labels = inputs.cuda(), labels.cuda()\n",
    "# #     print(labels)\n",
    "#     outputs = model(inputs)\n",
    "#     _, preds = torch.max(outputs.data, 1)\n",
    "#     preds_set += preds.cpu().tolist()\n",
    "# #     print(labels_set, preds_set)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.metrics import accuracy_score\n",
    "# accuracy_score(y_pred=preds_set, y_true=labels_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import matplotlib as mpl\n",
    "font = {'weight' : 'bold',\n",
    "        'size'   : 3}\n",
    "\n",
    "mpl.rc('font', **font)\n",
    "mpl.rcParams['figure.dpi']= 50\n",
    "import seaborn as sns\n",
    "sns.set_context(\"notebook\", rc={\"lines.linewidth\": 1.0})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "batch no. :  0\n",
      "batch no. :  1\n",
      "batch no. :  2\n",
      "batch no. :  3\n",
      "batch no. :  4\n",
      "batch no. :  5\n",
      "batch no. :  6\n",
      "batch no. :  7\n",
      "batch no. :  8\n",
      "batch no. :  9\n",
      "batch no. :  10\n",
      "batch no. :  11\n",
      "batch no. :  12\n",
      "batch no. :  13\n",
      "batch no. :  14\n",
      "batch no. :  15\n",
      "batch no. :  16\n",
      "batch no. :  17\n",
      "batch no. :  18\n",
      "batch no. :  19\n",
      "batch no. :  20\n",
      "batch no. :  21\n",
      "batch no. :  22\n",
      "batch no. :  23\n",
      "batch no. :  24\n",
      "batch no. :  25\n",
      "batch no. :  26\n",
      "batch no. :  27\n",
      "batch no. :  28\n",
      "batch no. :  29\n",
      "batch no. :  30\n",
      "batch no. :  31\n",
      "batch no. :  32\n",
      "batch no. :  33\n",
      "batch no. :  34\n",
      "batch no. :  35\n",
      "batch no. :  36\n",
      "batch no. :  37\n",
      "batch no. :  38\n",
      "batch no. :  39\n",
      "batch no. :  40\n",
      "batch no. :  41\n",
      "batch no. :  42\n",
      "batch no. :  43\n",
      "batch no. :  44\n",
      "batch no. :  45\n",
      "batch no. :  46\n",
      "batch no. :  47\n",
      "batch no. :  48\n",
      "batch no. :  49\n",
      "batch no. :  50\n",
      "batch no. :  51\n",
      "batch no. :  52\n",
      "batch no. :  53\n",
      "batch no. :  54\n",
      "batch no. :  55\n",
      "batch no. :  56\n",
      "batch no. :  57\n",
      "batch no. :  58\n",
      "batch no. :  59\n",
      "batch no. :  60\n",
      "batch no. :  61\n",
      "batch no. :  62\n",
      "batch no. :  63\n",
      "batch no. :  64\n",
      "batch no. :  65\n",
      "batch no. :  66\n",
      "batch no. :  67\n",
      "batch no. :  68\n",
      "batch no. :  69\n",
      "batch no. :  70\n",
      "batch no. :  71\n",
      "batch no. :  72\n",
      "batch no. :  73\n",
      "batch no. :  74\n",
      "batch no. :  75\n",
      "batch no. :  76\n",
      "batch no. :  77\n",
      "batch no. :  78\n",
      "batch no. :  79\n",
      "batch no. :  80\n",
      "batch no. :  81\n",
      "batch no. :  82\n",
      "batch no. :  83\n",
      "batch no. :  84\n",
      "batch no. :  85\n",
      "batch no. :  86\n",
      "batch no. :  87\n",
      "batch no. :  88\n",
      "batch no. :  89\n",
      "batch no. :  90\n",
      "batch no. :  91\n",
      "batch no. :  92\n",
      "batch no. :  93\n",
      "batch no. :  94\n",
      "batch no. :  95\n",
      "batch no. :  96\n",
      "batch no. :  97\n",
      "batch no. :  98\n",
      "batch no. :  99\n",
      "batch no. :  100\n",
      "batch no. :  101\n",
      "batch no. :  102\n",
      "batch no. :  103\n",
      "batch no. :  104\n",
      "batch no. :  105\n",
      "batch no. :  106\n",
      "batch no. :  107\n",
      "batch no. :  108\n",
      "batch no. :  109\n",
      "batch no. :  110\n",
      "batch no. :  111\n",
      "batch no. :  112\n",
      "batch no. :  113\n",
      "batch no. :  114\n",
      "batch no. :  115\n",
      "batch no. :  116\n",
      "batch no. :  117\n",
      "batch no. :  118\n",
      "batch no. :  119\n",
      "batch no. :  120\n",
      "batch no. :  121\n",
      "batch no. :  122\n",
      "batch no. :  123\n",
      "batch no. :  124\n",
      "batch no. :  125\n",
      "batch no. :  126\n",
      "batch no. :  127\n",
      "batch no. :  128\n",
      "batch no. :  129\n",
      "batch no. :  130\n",
      "batch no. :  131\n",
      "batch no. :  132\n",
      "batch no. :  133\n",
      "batch no. :  134\n",
      "batch no. :  135\n",
      "batch no. :  136\n",
      "batch no. :  137\n",
      "batch no. :  138\n",
      "batch no. :  139\n",
      "batch no. :  140\n",
      "batch no. :  141\n",
      "batch no. :  142\n",
      "batch no. :  143\n",
      "batch no. :  144\n",
      "batch no. :  145\n",
      "batch no. :  146\n",
      "batch no. :  147\n",
      "batch no. :  148\n",
      "batch no. :  149\n",
      "batch no. :  150\n",
      "batch no. :  151\n",
      "batch no. :  152\n",
      "batch no. :  153\n",
      "batch no. :  154\n",
      "batch no. :  155\n",
      "batch no. :  156\n",
      "batch no. :  157\n",
      "batch no. :  158\n",
      "batch no. :  159\n",
      "batch no. :  160\n",
      "batch no. :  161\n",
      "batch no. :  162\n",
      "batch no. :  163\n",
      "batch no. :  164\n",
      "batch no. :  165\n",
      "batch no. :  166\n",
      "batch no. :  167\n",
      "batch no. :  168\n",
      "batch no. :  169\n",
      "batch no. :  170\n",
      "batch no. :  171\n",
      "batch no. :  172\n",
      "batch no. :  173\n",
      "batch no. :  174\n",
      "batch no. :  175\n",
      "batch no. :  176\n",
      "batch no. :  177\n",
      "batch no. :  178\n",
      "batch no. :  179\n",
      "batch no. :  180\n",
      "batch no. :  181\n",
      "batch no. :  182\n",
      "batch no. :  183\n",
      "batch no. :  184\n",
      "batch no. :  185\n",
      "batch no. :  186\n",
      "batch no. :  187\n",
      "batch no. :  188\n",
      "batch no. :  189\n",
      "batch no. :  190\n",
      "batch no. :  191\n",
      "batch no. :  192\n",
      "batch no. :  193\n",
      "batch no. :  194\n",
      "batch no. :  195\n",
      "batch no. :  196\n",
      "batch no. :  197\n",
      "batch no. :  198\n",
      "batch no. :  199\n",
      "batch no. :  200\n",
      "batch no. :  201\n",
      "batch no. :  202\n",
      "batch no. :  203\n",
      "batch no. :  204\n",
      "batch no. :  205\n",
      "batch no. :  206\n",
      "batch no. :  207\n",
      "batch no. :  208\n",
      "batch no. :  209\n",
      "batch no. :  210\n",
      "batch no. :  211\n",
      "batch no. :  212\n",
      "batch no. :  213\n",
      "batch no. :  214\n",
      "batch no. :  215\n",
      "batch no. :  216\n",
      "batch no. :  217\n",
      "batch no. :  218\n",
      "batch no. :  219\n",
      "batch no. :  220\n",
      "batch no. :  221\n",
      "batch no. :  222\n",
      "batch no. :  223\n",
      "batch no. :  224\n",
      "batch no. :  225\n",
      "batch no. :  226\n",
      "batch no. :  227\n",
      "batch no. :  228\n",
      "batch no. :  229\n",
      "batch no. :  230\n",
      "batch no. :  231\n",
      "batch no. :  232\n",
      "batch no. :  233\n",
      "batch no. :  234\n",
      "batch no. :  235\n",
      "batch no. :  236\n",
      "batch no. :  237\n",
      "batch no. :  238\n",
      "batch no. :  239\n",
      "batch no. :  240\n",
      "batch no. :  241\n",
      "batch no. :  242\n",
      "batch no. :  243\n",
      "batch no. :  244\n",
      "batch no. :  245\n",
      "batch no. :  246\n",
      "batch no. :  247\n",
      "batch no. :  248\n",
      "batch no. :  249\n",
      "batch no. :  250\n",
      "batch no. :  251\n",
      "batch no. :  252\n",
      "batch no. :  253\n",
      "batch no. :  254\n",
      "batch no. :  255\n",
      "batch no. :  256\n",
      "batch no. :  257\n",
      "batch no. :  258\n",
      "batch no. :  259\n",
      "batch no. :  260\n",
      "batch no. :  261\n",
      "batch no. :  262\n",
      "batch no. :  263\n",
      "batch no. :  264\n",
      "batch no. :  265\n",
      "batch no. :  266\n",
      "batch no. :  267\n",
      "batch no. :  268\n",
      "batch no. :  269\n",
      "batch no. :  270\n",
      "batch no. :  271\n",
      "batch no. :  272\n",
      "batch no. :  273\n",
      "batch no. :  274\n",
      "batch no. :  275\n",
      "batch no. :  276\n",
      "batch no. :  277\n",
      "batch no. :  278\n",
      "batch no. :  279\n",
      "batch no. :  280\n",
      "batch no. :  281\n",
      "batch no. :  282\n",
      "batch no. :  283\n",
      "batch no. :  284\n",
      "batch no. :  285\n",
      "batch no. :  286\n",
      "batch no. :  287\n",
      "batch no. :  288\n",
      "batch no. :  289\n",
      "batch no. :  290\n",
      "batch no. :  291\n",
      "batch no. :  292\n",
      "batch no. :  293\n",
      "batch no. :  294\n",
      "batch no. :  295\n",
      "batch no. :  296\n",
      "batch no. :  297\n",
      "batch no. :  298\n",
      "batch no. :  299\n",
      "batch no. :  300\n",
      "batch no. :  301\n",
      "batch no. :  302\n",
      "batch no. :  303\n",
      "batch no. :  304\n",
      "batch no. :  305\n",
      "batch no. :  306\n",
      "batch no. :  307\n",
      "batch no. :  308\n",
      "batch no. :  309\n",
      "batch no. :  310\n",
      "batch no. :  311\n",
      "batch no. :  312\n",
      "batch no. :  313\n",
      "batch no. :  314\n",
      "batch no. :  315\n",
      "batch no. :  316\n",
      "batch no. :  317\n",
      "batch no. :  318\n",
      "batch no. :  319\n",
      "batch no. :  320\n",
      "batch no. :  321\n",
      "batch no. :  322\n",
      "batch no. :  323\n",
      "batch no. :  324\n",
      "batch no. :  325\n",
      "batch no. :  326\n",
      "batch no. :  327\n",
      "batch no. :  328\n",
      "batch no. :  329\n",
      "batch no. :  330\n",
      "batch no. :  331\n",
      "batch no. :  332\n",
      "batch no. :  333\n",
      "batch no. :  334\n",
      "batch no. :  335\n",
      "batch no. :  336\n",
      "batch no. :  337\n",
      "batch no. :  338\n",
      "batch no. :  339\n",
      "batch no. :  340\n",
      "batch no. :  341\n",
      "batch no. :  342\n",
      "batch no. :  343\n",
      "batch no. :  344\n",
      "batch no. :  345\n",
      "batch no. :  346\n",
      "batch no. :  347\n",
      "batch no. :  348\n",
      "batch no. :  349\n",
      "batch no. :  350\n",
      "batch no. :  351\n",
      "batch no. :  352\n",
      "batch no. :  353\n",
      "batch no. :  354\n",
      "batch no. :  355\n",
      "batch no. :  356\n",
      "batch no. :  357\n",
      "batch no. :  358\n",
      "batch no. :  359\n",
      "batch no. :  360\n",
      "batch no. :  361\n",
      "batch no. :  362\n",
      "batch no. :  363\n",
      "batch no. :  364\n",
      "batch no. :  365\n",
      "batch no. :  366\n",
      "batch no. :  367\n",
      "batch no. :  368\n",
      "batch no. :  369\n",
      "batch no. :  370\n",
      "batch no. :  371\n",
      "batch no. :  372\n",
      "batch no. :  373\n",
      "batch no. :  374\n",
      "batch no. :  375\n",
      "batch no. :  376\n",
      "batch no. :  377\n",
      "batch no. :  378\n",
      "batch no. :  379\n",
      "batch no. :  380\n",
      "batch no. :  381\n",
      "batch no. :  382\n",
      "batch no. :  383\n",
      "batch no. :  384\n",
      "batch no. :  385\n",
      "batch no. :  386\n",
      "batch no. :  387\n",
      "batch no. :  388\n",
      "batch no. :  389\n",
      "batch no. :  390\n",
      "batch no. :  391\n",
      "batch no. :  392\n",
      "batch no. :  393\n",
      "batch no. :  394\n",
      "batch no. :  395\n",
      "batch no. :  396\n",
      "batch no. :  397\n",
      "batch no. :  398\n",
      "batch no. :  399\n",
      "batch no. :  400\n",
      "batch no. :  401\n",
      "batch no. :  402\n",
      "batch no. :  403\n",
      "batch no. :  404\n",
      "batch no. :  405\n",
      "batch no. :  406\n",
      "batch no. :  407\n",
      "batch no. :  408\n",
      "batch no. :  409\n",
      "batch no. :  410\n",
      "batch no. :  411\n",
      "batch no. :  412\n",
      "batch no. :  413\n",
      "batch no. :  414\n",
      "batch no. :  415\n",
      "batch no. :  416\n",
      "batch no. :  417\n",
      "batch no. :  418\n",
      "batch no. :  419\n",
      "batch no. :  420\n",
      "batch no. :  421\n",
      "batch no. :  422\n",
      "batch no. :  423\n",
      "batch no. :  424\n",
      "batch no. :  425\n",
      "batch no. :  426\n",
      "batch no. :  427\n",
      "batch no. :  428\n",
      "batch no. :  429\n",
      "batch no. :  430\n",
      "batch no. :  431\n",
      "batch no. :  432\n",
      "batch no. :  433\n",
      "batch no. :  434\n",
      "batch no. :  435\n",
      "batch no. :  436\n",
      "batch no. :  437\n",
      "batch no. :  438\n",
      "batch no. :  439\n",
      "batch no. :  440\n",
      "batch no. :  441\n",
      "batch no. :  442\n",
      "batch no. :  443\n",
      "batch no. :  444\n",
      "batch no. :  445\n",
      "batch no. :  446\n",
      "batch no. :  447\n",
      "batch no. :  448\n",
      "batch no. :  449\n",
      "batch no. :  450\n",
      "batch no. :  451\n",
      "batch no. :  452\n",
      "batch no. :  453\n",
      "batch no. :  454\n",
      "batch no. :  455\n",
      "batch no. :  456\n",
      "batch no. :  457\n",
      "batch no. :  458\n",
      "batch no. :  459\n",
      "batch no. :  460\n",
      "batch no. :  461\n",
      "batch no. :  462\n",
      "batch no. :  463\n",
      "batch no. :  464\n",
      "batch no. :  465\n",
      "batch no. :  466\n",
      "batch no. :  467\n",
      "batch no. :  468\n",
      "batch no. :  469\n",
      "batch no. :  470\n",
      "batch no. :  471\n",
      "batch no. :  472\n",
      "batch no. :  473\n",
      "batch no. :  474\n",
      "batch no. :  475\n",
      "batch no. :  476\n",
      "batch no. :  477\n",
      "batch no. :  478\n",
      "batch no. :  479\n",
      "batch no. :  480\n",
      "batch no. :  481\n",
      "batch no. :  482\n",
      "batch no. :  483\n",
      "batch no. :  484\n",
      "batch no. :  485\n",
      "batch no. :  486\n",
      "batch no. :  487\n",
      "batch no. :  488\n",
      "batch no. :  489\n",
      "batch no. :  490\n",
      "batch no. :  491\n",
      "batch no. :  492\n",
      "batch no. :  493\n",
      "batch no. :  494\n",
      "batch no. :  495\n",
      "batch no. :  496\n",
      "batch no. :  497\n",
      "batch no. :  498\n",
      "batch no. :  499\n",
      "batch no. :  500\n",
      "batch no. :  501\n",
      "batch no. :  502\n",
      "batch no. :  503\n",
      "batch no. :  504\n",
      "batch no. :  505\n",
      "batch no. :  506\n",
      "batch no. :  507\n",
      "batch no. :  508\n",
      "batch no. :  509\n",
      "batch no. :  510\n",
      "batch no. :  511\n",
      "batch no. :  512\n",
      "batch no. :  513\n",
      "batch no. :  514\n",
      "batch no. :  515\n",
      "batch no. :  516\n",
      "batch no. :  517\n",
      "batch no. :  518\n",
      "batch no. :  519\n",
      "batch no. :  520\n",
      "batch no. :  521\n",
      "batch no. :  522\n",
      "batch no. :  523\n",
      "batch no. :  524\n",
      "batch no. :  525\n",
      "batch no. :  526\n",
      "batch no. :  527\n",
      "batch no. :  528\n",
      "batch no. :  529\n",
      "batch no. :  530\n",
      "batch no. :  531\n",
      "batch no. :  532\n",
      "batch no. :  533\n",
      "batch no. :  534\n",
      "batch no. :  535\n",
      "batch no. :  536\n",
      "batch no. :  537\n",
      "batch no. :  538\n",
      "batch no. :  539\n",
      "batch no. :  540\n",
      "batch no. :  541\n",
      "batch no. :  542\n",
      "batch no. :  543\n",
      "batch no. :  544\n",
      "batch no. :  545\n",
      "batch no. :  546\n",
      "batch no. :  547\n",
      "batch no. :  548\n",
      "batch no. :  549\n",
      "batch no. :  550\n",
      "batch no. :  551\n",
      "batch no. :  552\n",
      "batch no. :  553\n",
      "batch no. :  554\n",
      "batch no. :  555\n",
      "batch no. :  556\n",
      "batch no. :  557\n",
      "batch no. :  558\n",
      "batch no. :  559\n",
      "batch no. :  560\n",
      "batch no. :  561\n",
      "batch no. :  562\n",
      "batch no. :  563\n",
      "batch no. :  564\n",
      "batch no. :  565\n",
      "batch no. :  566\n",
      "batch no. :  567\n",
      "batch no. :  568\n",
      "batch no. :  569\n",
      "batch no. :  570\n",
      "batch no. :  571\n",
      "batch no. :  572\n",
      "batch no. :  573\n",
      "batch no. :  574\n",
      "batch no. :  575\n",
      "batch no. :  576\n",
      "batch no. :  577\n",
      "batch no. :  578\n",
      "batch no. :  579\n",
      "batch no. :  580\n",
      "batch no. :  581\n",
      "batch no. :  582\n",
      "batch no. :  583\n",
      "batch no. :  584\n",
      "batch no. :  585\n",
      "batch no. :  586\n",
      "batch no. :  587\n",
      "batch no. :  588\n",
      "batch no. :  589\n",
      "batch no. :  590\n",
      "batch no. :  591\n",
      "batch no. :  592\n",
      "batch no. :  593\n",
      "batch no. :  594\n",
      "batch no. :  595\n",
      "batch no. :  596\n",
      "batch no. :  597\n",
      "batch no. :  598\n",
      "batch no. :  599\n",
      "batch no. :  600\n",
      "batch no. :  601\n",
      "batch no. :  602\n",
      "batch no. :  603\n",
      "batch no. :  604\n",
      "batch no. :  605\n",
      "batch no. :  606\n",
      "batch no. :  607\n",
      "batch no. :  608\n",
      "batch no. :  609\n",
      "batch no. :  610\n",
      "batch no. :  611\n",
      "batch no. :  612\n",
      "batch no. :  613\n",
      "batch no. :  614\n",
      "batch no. :  615\n",
      "batch no. :  616\n",
      "batch no. :  617\n",
      "batch no. :  618\n",
      "batch no. :  619\n",
      "batch no. :  620\n",
      "batch no. :  621\n",
      "batch no. :  622\n",
      "batch no. :  623\n",
      "batch no. :  624\n",
      "batch no. :  625\n",
      "batch no. :  626\n",
      "batch no. :  627\n",
      "batch no. :  628\n",
      "batch no. :  629\n",
      "batch no. :  630\n",
      "batch no. :  631\n",
      "batch no. :  632\n",
      "batch no. :  633\n",
      "batch no. :  634\n",
      "batch no. :  635\n",
      "batch no. :  636\n",
      "batch no. :  637\n",
      "batch no. :  638\n",
      "batch no. :  639\n",
      "batch no. :  640\n",
      "batch no. :  641\n",
      "batch no. :  642\n",
      "batch no. :  643\n",
      "batch no. :  644\n",
      "batch no. :  645\n",
      "batch no. :  646\n",
      "batch no. :  647\n",
      "batch no. :  648\n",
      "batch no. :  649\n",
      "batch no. :  650\n",
      "batch no. :  651\n",
      "batch no. :  652\n",
      "batch no. :  653\n",
      "batch no. :  654\n",
      "batch no. :  655\n",
      "batch no. :  656\n",
      "batch no. :  657\n",
      "batch no. :  658\n",
      "batch no. :  659\n",
      "batch no. :  660\n",
      "batch no. :  661\n",
      "batch no. :  662\n",
      "batch no. :  663\n",
      "batch no. :  664\n",
      "batch no. :  665\n",
      "batch no. :  666\n",
      "batch no. :  667\n",
      "batch no. :  668\n",
      "batch no. :  669\n",
      "batch no. :  670\n",
      "batch no. :  671\n",
      "batch no. :  672\n",
      "batch no. :  673\n",
      "batch no. :  674\n",
      "batch no. :  675\n",
      "batch no. :  676\n",
      "batch no. :  677\n",
      "batch no. :  678\n",
      "batch no. :  679\n",
      "batch no. :  680\n",
      "batch no. :  681\n",
      "batch no. :  682\n",
      "batch no. :  683\n",
      "batch no. :  684\n",
      "batch no. :  685\n",
      "batch no. :  686\n",
      "batch no. :  687\n",
      "batch no. :  688\n",
      "batch no. :  689\n",
      "batch no. :  690\n",
      "batch no. :  691\n",
      "batch no. :  692\n",
      "batch no. :  693\n",
      "batch no. :  694\n",
      "batch no. :  695\n",
      "batch no. :  696\n",
      "batch no. :  697\n",
      "batch no. :  698\n",
      "batch no. :  699\n",
      "batch no. :  700\n",
      "batch no. :  701\n",
      "batch no. :  702\n",
      "batch no. :  703\n",
      "batch no. :  704\n",
      "batch no. :  705\n",
      "batch no. :  706\n",
      "batch no. :  707\n",
      "batch no. :  708\n",
      "batch no. :  709\n",
      "batch no. :  710\n",
      "batch no. :  711\n",
      "batch no. :  712\n",
      "batch no. :  713\n",
      "batch no. :  714\n",
      "batch no. :  715\n",
      "batch no. :  716\n",
      "batch no. :  717\n",
      "batch no. :  718\n",
      "batch no. :  719\n",
      "batch no. :  720\n",
      "batch no. :  721\n",
      "batch no. :  722\n",
      "batch no. :  723\n",
      "batch no. :  724\n",
      "batch no. :  725\n",
      "batch no. :  726\n",
      "batch no. :  727\n",
      "batch no. :  728\n",
      "batch no. :  729\n",
      "batch no. :  730\n",
      "batch no. :  731\n",
      "batch no. :  732\n",
      "batch no. :  733\n",
      "batch no. :  734\n",
      "batch no. :  735\n",
      "batch no. :  736\n",
      "batch no. :  737\n",
      "batch no. :  738\n",
      "batch no. :  739\n",
      "batch no. :  740\n",
      "batch no. :  741\n",
      "batch no. :  742\n",
      "batch no. :  743\n",
      "batch no. :  744\n",
      "batch no. :  745\n",
      "batch no. :  746\n",
      "batch no. :  747\n",
      "batch no. :  748\n",
      "batch no. :  749\n",
      "batch no. :  750\n",
      "batch no. :  751\n",
      "batch no. :  752\n",
      "batch no. :  753\n",
      "batch no. :  754\n",
      "batch no. :  755\n",
      "batch no. :  756\n",
      "batch no. :  757\n",
      "batch no. :  758\n",
      "batch no. :  759\n",
      "batch no. :  760\n",
      "batch no. :  761\n",
      "batch no. :  762\n",
      "batch no. :  763\n",
      "batch no. :  764\n",
      "batch no. :  765\n",
      "batch no. :  766\n",
      "batch no. :  767\n",
      "batch no. :  768\n",
      "batch no. :  769\n",
      "batch no. :  770\n",
      "batch no. :  771\n",
      "batch no. :  772\n",
      "batch no. :  773\n",
      "batch no. :  774\n",
      "batch no. :  775\n",
      "batch no. :  776\n",
      "batch no. :  777\n",
      "batch no. :  778\n",
      "batch no. :  779\n",
      "batch no. :  780\n",
      "batch no. :  781\n",
      "batch no. :  782\n",
      "batch no. :  783\n",
      "batch no. :  784\n",
      "batch no. :  785\n",
      "batch no. :  786\n",
      "batch no. :  787\n",
      "batch no. :  788\n",
      "batch no. :  789\n",
      "batch no. :  790\n",
      "batch no. :  791\n",
      "batch no. :  792\n",
      "batch no. :  793\n",
      "batch no. :  794\n",
      "batch no. :  795\n",
      "batch no. :  796\n",
      "batch no. :  797\n",
      "batch no. :  798\n",
      "batch no. :  799\n",
      "batch no. :  800\n",
      "batch no. :  801\n",
      "batch no. :  802\n",
      "batch no. :  803\n",
      "batch no. :  804\n",
      "batch no. :  805\n",
      "batch no. :  806\n",
      "batch no. :  807\n",
      "batch no. :  808\n",
      "batch no. :  809\n",
      "batch no. :  810\n",
      "batch no. :  811\n",
      "batch no. :  812\n",
      "batch no. :  813\n",
      "batch no. :  814\n",
      "batch no. :  815\n",
      "batch no. :  816\n",
      "batch no. :  817\n",
      "batch no. :  818\n",
      "batch no. :  819\n",
      "batch no. :  820\n",
      "batch no. :  821\n",
      "batch no. :  822\n",
      "batch no. :  823\n",
      "batch no. :  824\n",
      "batch no. :  825\n",
      "batch no. :  826\n",
      "batch no. :  827\n",
      "batch no. :  828\n",
      "batch no. :  829\n",
      "batch no. :  830\n",
      "batch no. :  831\n",
      "batch no. :  832\n",
      "batch no. :  833\n",
      "batch no. :  834\n",
      "batch no. :  835\n",
      "batch no. :  836\n",
      "batch no. :  837\n",
      "batch no. :  838\n",
      "batch no. :  839\n",
      "batch no. :  840\n",
      "batch no. :  841\n",
      "batch no. :  842\n",
      "batch no. :  843\n",
      "batch no. :  844\n",
      "batch no. :  845\n",
      "batch no. :  846\n",
      "batch no. :  847\n",
      "batch no. :  848\n",
      "batch no. :  849\n",
      "batch no. :  850\n",
      "batch no. :  851\n",
      "batch no. :  852\n",
      "batch no. :  853\n",
      "batch no. :  854\n",
      "batch no. :  855\n",
      "batch no. :  856\n",
      "batch no. :  857\n",
      "batch no. :  858\n",
      "batch no. :  859\n",
      "batch no. :  860\n",
      "batch no. :  861\n",
      "batch no. :  862\n",
      "batch no. :  863\n",
      "batch no. :  864\n",
      "batch no. :  865\n",
      "batch no. :  866\n",
      "batch no. :  867\n",
      "batch no. :  868\n",
      "batch no. :  869\n",
      "batch no. :  870\n",
      "batch no. :  871\n",
      "batch no. :  872\n",
      "batch no. :  873\n",
      "batch no. :  874\n",
      "batch no. :  875\n",
      "batch no. :  876\n",
      "batch no. :  877\n",
      "batch no. :  878\n",
      "batch no. :  879\n",
      "batch no. :  880\n",
      "batch no. :  881\n",
      "batch no. :  882\n",
      "batch no. :  883\n",
      "batch no. :  884\n",
      "batch no. :  885\n",
      "batch no. :  886\n",
      "batch no. :  887\n",
      "batch no. :  888\n",
      "batch no. :  889\n",
      "batch no. :  890\n",
      "batch no. :  891\n",
      "batch no. :  892\n",
      "batch no. :  893\n",
      "batch no. :  894\n",
      "batch no. :  895\n",
      "batch no. :  896\n",
      "batch no. :  897\n",
      "batch no. :  898\n",
      "batch no. :  899\n",
      "batch no. :  900\n",
      "batch no. :  901\n",
      "batch no. :  902\n",
      "batch no. :  903\n",
      "batch no. :  904\n",
      "batch no. :  905\n",
      "batch no. :  906\n",
      "batch no. :  907\n",
      "batch no. :  908\n",
      "batch no. :  909\n",
      "batch no. :  910\n",
      "batch no. :  911\n",
      "batch no. :  912\n",
      "batch no. :  913\n",
      "batch no. :  914\n",
      "batch no. :  915\n",
      "batch no. :  916\n",
      "batch no. :  917\n",
      "batch no. :  918\n",
      "batch no. :  919\n",
      "batch no. :  920\n",
      "batch no. :  921\n",
      "batch no. :  922\n",
      "batch no. :  923\n",
      "batch no. :  924\n",
      "batch no. :  925\n",
      "batch no. :  926\n",
      "batch no. :  927\n",
      "batch no. :  928\n",
      "batch no. :  929\n",
      "batch no. :  930\n",
      "batch no. :  931\n",
      "batch no. :  932\n",
      "batch no. :  933\n",
      "batch no. :  934\n",
      "batch no. :  935\n",
      "batch no. :  936\n",
      "batch no. :  937\n",
      "batch no. :  938\n",
      "batch no. :  939\n",
      "batch no. :  940\n",
      "batch no. :  941\n",
      "batch no. :  942\n",
      "batch no. :  943\n",
      "batch no. :  944\n",
      "batch no. :  945\n",
      "batch no. :  946\n",
      "batch no. :  947\n",
      "batch no. :  948\n",
      "batch no. :  949\n",
      "batch no. :  950\n",
      "batch no. :  951\n",
      "batch no. :  952\n",
      "batch no. :  953\n",
      "batch no. :  954\n",
      "batch no. :  955\n",
      "batch no. :  956\n",
      "batch no. :  957\n",
      "batch no. :  958\n",
      "batch no. :  959\n",
      "batch no. :  960\n",
      "batch no. :  961\n",
      "batch no. :  962\n",
      "batch no. :  963\n",
      "batch no. :  964\n",
      "batch no. :  965\n",
      "batch no. :  966\n",
      "batch no. :  967\n",
      "batch no. :  968\n",
      "batch no. :  969\n",
      "batch no. :  970\n",
      "batch no. :  971\n",
      "batch no. :  972\n",
      "batch no. :  973\n",
      "batch no. :  974\n",
      "batch no. :  975\n",
      "batch no. :  976\n",
      "batch no. :  977\n",
      "batch no. :  978\n",
      "batch no. :  979\n",
      "batch no. :  980\n",
      "batch no. :  981\n",
      "batch no. :  982\n",
      "batch no. :  983\n",
      "batch no. :  984\n",
      "batch no. :  985\n",
      "batch no. :  986\n",
      "batch no. :  987\n",
      "batch no. :  988\n",
      "batch no. :  989\n",
      "batch no. :  990\n",
      "batch no. :  991\n",
      "batch no. :  992\n",
      "batch no. :  993\n",
      "batch no. :  994\n",
      "batch no. :  995\n",
      "batch no. :  996\n",
      "batch no. :  997\n",
      "batch no. :  998\n",
      "batch no. :  999\n",
      "batch no. :  1000\n",
      "batch no. :  1001\n",
      "batch no. :  1002\n",
      "batch no. :  1003\n",
      "batch no. :  1004\n",
      "batch no. :  1005\n",
      "batch no. :  1006\n",
      "batch no. :  1007\n",
      "batch no. :  1008\n",
      "batch no. :  1009\n",
      "batch no. :  1010\n",
      "batch no. :  1011\n",
      "batch no. :  1012\n",
      "batch no. :  1013\n",
      "batch no. :  1014\n",
      "batch no. :  1015\n",
      "batch no. :  1016\n",
      "batch no. :  1017\n",
      "batch no. :  1018\n",
      "batch no. :  1019\n",
      "batch no. :  1020\n",
      "batch no. :  1021\n",
      "batch no. :  1022\n",
      "batch no. :  1023\n",
      "batch no. :  1024\n",
      "batch no. :  1025\n",
      "batch no. :  1026\n",
      "batch no. :  1027\n",
      "batch no. :  1028\n",
      "batch no. :  1029\n",
      "batch no. :  1030\n",
      "batch no. :  1031\n",
      "batch no. :  1032\n",
      "batch no. :  1033\n",
      "batch no. :  1034\n",
      "batch no. :  1035\n",
      "batch no. :  1036\n",
      "batch no. :  1037\n",
      "batch no. :  1038\n",
      "batch no. :  1039\n",
      "batch no. :  1040\n",
      "batch no. :  1041\n",
      "batch no. :  1042\n",
      "batch no. :  1043\n",
      "batch no. :  1044\n",
      "batch no. :  1045\n",
      "batch no. :  1046\n",
      "batch no. :  1047\n",
      "batch no. :  1048\n",
      "batch no. :  1049\n",
      "batch no. :  1050\n",
      "batch no. :  1051\n",
      "batch no. :  1052\n",
      "batch no. :  1053\n",
      "batch no. :  1054\n",
      "batch no. :  1055\n",
      "batch no. :  1056\n",
      "batch no. :  1057\n",
      "batch no. :  1058\n",
      "batch no. :  1059\n",
      "batch no. :  1060\n",
      "batch no. :  1061\n",
      "batch no. :  1062\n",
      "batch no. :  1063\n",
      "batch no. :  1064\n",
      "batch no. :  1065\n",
      "batch no. :  1066\n",
      "batch no. :  1067\n",
      "batch no. :  1068\n",
      "batch no. :  1069\n",
      "batch no. :  1070\n",
      "batch no. :  1071\n",
      "batch no. :  1072\n",
      "batch no. :  1073\n",
      "batch no. :  1074\n",
      "batch no. :  1075\n",
      "batch no. :  1076\n",
      "batch no. :  1077\n",
      "batch no. :  1078\n",
      "batch no. :  1079\n",
      "batch no. :  1080\n",
      "batch no. :  1081\n",
      "batch no. :  1082\n",
      "batch no. :  1083\n",
      "batch no. :  1084\n",
      "batch no. :  1085\n",
      "batch no. :  1086\n",
      "batch no. :  1087\n",
      "batch no. :  1088\n",
      "batch no. :  1089\n",
      "batch no. :  1090\n",
      "batch no. :  1091\n",
      "batch no. :  1092\n",
      "batch no. :  1093\n",
      "batch no. :  1094\n",
      "batch no. :  1095\n",
      "batch no. :  1096\n",
      "batch no. :  1097\n",
      "batch no. :  1098\n",
      "batch no. :  1099\n",
      "batch no. :  1100\n",
      "batch no. :  1101\n",
      "batch no. :  1102\n",
      "batch no. :  1103\n",
      "batch no. :  1104\n",
      "batch no. :  1105\n",
      "batch no. :  1106\n",
      "batch no. :  1107\n",
      "batch no. :  1108\n",
      "batch no. :  1109\n",
      "batch no. :  1110\n",
      "batch no. :  1111\n",
      "batch no. :  1112\n",
      "batch no. :  1113\n",
      "batch no. :  1114\n",
      "batch no. :  1115\n",
      "batch no. :  1116\n",
      "batch no. :  1117\n",
      "batch no. :  1118\n",
      "batch no. :  1119\n",
      "batch no. :  1120\n",
      "batch no. :  1121\n",
      "batch no. :  1122\n",
      "batch no. :  1123\n",
      "batch no. :  1124\n",
      "batch no. :  1125\n",
      "batch no. :  1126\n",
      "batch no. :  1127\n",
      "batch no. :  1128\n",
      "batch no. :  1129\n",
      "batch no. :  1130\n",
      "batch no. :  1131\n",
      "batch no. :  1132\n",
      "batch no. :  1133\n",
      "batch no. :  1134\n",
      "batch no. :  1135\n",
      "batch no. :  1136\n",
      "batch no. :  1137\n",
      "batch no. :  1138\n",
      "batch no. :  1139\n",
      "batch no. :  1140\n",
      "batch no. :  1141\n",
      "batch no. :  1142\n",
      "batch no. :  1143\n",
      "batch no. :  1144\n",
      "batch no. :  1145\n",
      "batch no. :  1146\n",
      "batch no. :  1147\n",
      "batch no. :  1148\n",
      "batch no. :  1149\n",
      "batch no. :  1150\n",
      "batch no. :  1151\n",
      "batch no. :  1152\n",
      "batch no. :  1153\n",
      "batch no. :  1154\n",
      "batch no. :  1155\n",
      "batch no. :  1156\n",
      "batch no. :  1157\n",
      "batch no. :  1158\n",
      "batch no. :  1159\n",
      "batch no. :  1160\n",
      "batch no. :  1161\n",
      "batch no. :  1162\n",
      "batch no. :  1163\n",
      "batch no. :  1164\n",
      "batch no. :  1165\n",
      "batch no. :  1166\n",
      "batch no. :  1167\n",
      "batch no. :  1168\n",
      "batch no. :  1169\n",
      "batch no. :  1170\n",
      "batch no. :  1171\n",
      "batch no. :  1172\n",
      "batch no. :  1173\n",
      "batch no. :  1174\n",
      "batch no. :  1175\n",
      "batch no. :  1176\n",
      "batch no. :  1177\n",
      "batch no. :  1178\n",
      "batch no. :  1179\n",
      "batch no. :  1180\n",
      "batch no. :  1181\n",
      "batch no. :  1182\n",
      "batch no. :  1183\n",
      "batch no. :  1184\n",
      "batch no. :  1185\n",
      "batch no. :  1186\n",
      "batch no. :  1187\n",
      "batch no. :  1188\n",
      "batch no. :  1189\n",
      "batch no. :  1190\n",
      "batch no. :  1191\n",
      "batch no. :  1192\n",
      "batch no. :  1193\n",
      "batch no. :  1194\n",
      "batch no. :  1195\n",
      "batch no. :  1196\n",
      "batch no. :  1197\n",
      "batch no. :  1198\n",
      "batch no. :  1199\n",
      "batch no. :  1200\n",
      "batch no. :  1201\n",
      "batch no. :  1202\n",
      "batch no. :  1203\n",
      "batch no. :  1204\n",
      "batch no. :  1205\n",
      "batch no. :  1206\n",
      "batch no. :  1207\n",
      "batch no. :  1208\n",
      "batch no. :  1209\n",
      "batch no. :  1210\n",
      "batch no. :  1211\n",
      "batch no. :  1212\n",
      "batch no. :  1213\n",
      "batch no. :  1214\n",
      "batch no. :  1215\n",
      "batch no. :  1216\n",
      "batch no. :  1217\n",
      "batch no. :  1218\n",
      "batch no. :  1219\n",
      "batch no. :  1220\n",
      "batch no. :  1221\n",
      "batch no. :  1222\n",
      "batch no. :  1223\n",
      "batch no. :  1224\n",
      "batch no. :  1225\n",
      "batch no. :  1226\n",
      "batch no. :  1227\n",
      "batch no. :  1228\n",
      "batch no. :  1229\n",
      "batch no. :  1230\n",
      "batch no. :  1231\n",
      "batch no. :  1232\n",
      "batch no. :  1233\n",
      "batch no. :  1234\n",
      "batch no. :  1235\n",
      "batch no. :  1236\n",
      "batch no. :  1237\n",
      "batch no. :  1238\n",
      "batch no. :  1239\n",
      "batch no. :  1240\n",
      "batch no. :  1241\n",
      "batch no. :  1242\n",
      "batch no. :  1243\n",
      "batch no. :  1244\n",
      "batch no. :  1245\n",
      "batch no. :  1246\n",
      "batch no. :  1247\n",
      "batch no. :  1248\n",
      "batch no. :  1249\n",
      "batch no. :  1250\n",
      "batch no. :  1251\n",
      "batch no. :  1252\n",
      "batch no. :  1253\n",
      "batch no. :  1254\n",
      "batch no. :  1255\n",
      "batch no. :  1256\n",
      "batch no. :  1257\n",
      "batch no. :  1258\n",
      "batch no. :  1259\n",
      "batch no. :  1260\n",
      "batch no. :  1261\n",
      "batch no. :  1262\n",
      "batch no. :  1263\n",
      "batch no. :  1264\n",
      "batch no. :  1265\n",
      "batch no. :  1266\n",
      "batch no. :  1267\n",
      "batch no. :  1268\n",
      "batch no. :  1269\n",
      "batch no. :  1270\n",
      "batch no. :  1271\n",
      "batch no. :  1272\n",
      "batch no. :  1273\n",
      "batch no. :  1274\n",
      "batch no. :  1275\n",
      "batch no. :  1276\n",
      "batch no. :  1277\n",
      "batch no. :  1278\n",
      "batch no. :  1279\n",
      "batch no. :  1280\n",
      "batch no. :  1281\n",
      "batch no. :  1282\n",
      "batch no. :  1283\n",
      "batch no. :  1284\n",
      "batch no. :  1285\n",
      "batch no. :  1286\n",
      "batch no. :  1287\n",
      "batch no. :  1288\n",
      "batch no. :  1289\n",
      "batch no. :  1290\n",
      "batch no. :  1291\n",
      "batch no. :  1292\n",
      "batch no. :  1293\n",
      "batch no. :  1294\n",
      "batch no. :  1295\n",
      "batch no. :  1296\n",
      "batch no. :  1297\n",
      "batch no. :  1298\n",
      "batch no. :  1299\n",
      "batch no. :  1300\n",
      "batch no. :  1301\n",
      "batch no. :  1302\n",
      "batch no. :  1303\n",
      "batch no. :  1304\n",
      "batch no. :  1305\n",
      "batch no. :  1306\n",
      "batch no. :  1307\n",
      "batch no. :  1308\n",
      "batch no. :  1309\n",
      "batch no. :  1310\n",
      "batch no. :  1311\n",
      "batch no. :  1312\n",
      "batch no. :  1313\n",
      "batch no. :  1314\n",
      "batch no. :  1315\n",
      "batch no. :  1316\n",
      "batch no. :  1317\n",
      "batch no. :  1318\n",
      "batch no. :  1319\n",
      "batch no. :  1320\n",
      "batch no. :  1321\n",
      "batch no. :  1322\n",
      "batch no. :  1323\n",
      "batch no. :  1324\n",
      "batch no. :  1325\n",
      "batch no. :  1326\n",
      "batch no. :  1327\n",
      "batch no. :  1328\n",
      "batch no. :  1329\n",
      "batch no. :  1330\n",
      "batch no. :  1331\n",
      "batch no. :  1332\n",
      "batch no. :  1333\n",
      "batch no. :  1334\n",
      "batch no. :  1335\n",
      "batch no. :  1336\n",
      "batch no. :  1337\n",
      "batch no. :  1338\n",
      "batch no. :  1339\n",
      "batch no. :  1340\n",
      "batch no. :  1341\n",
      "batch no. :  1342\n",
      "batch no. :  1343\n",
      "batch no. :  1344\n",
      "batch no. :  1345\n",
      "batch no. :  1346\n",
      "batch no. :  1347\n",
      "batch no. :  1348\n",
      "batch no. :  1349\n",
      "batch no. :  1350\n",
      "batch no. :  1351\n",
      "batch no. :  1352\n",
      "batch no. :  1353\n",
      "batch no. :  1354\n",
      "batch no. :  1355\n",
      "batch no. :  1356\n",
      "batch no. :  1357\n",
      "batch no. :  1358\n",
      "batch no. :  1359\n",
      "batch no. :  1360\n",
      "batch no. :  1361\n",
      "batch no. :  1362\n",
      "batch no. :  1363\n",
      "batch no. :  1364\n",
      "batch no. :  1365\n",
      "batch no. :  1366\n",
      "batch no. :  1367\n",
      "batch no. :  1368\n",
      "batch no. :  1369\n",
      "batch no. :  1370\n",
      "batch no. :  1371\n",
      "batch no. :  1372\n",
      "batch no. :  1373\n",
      "batch no. :  1374\n",
      "batch no. :  1375\n",
      "batch no. :  1376\n",
      "batch no. :  1377\n",
      "batch no. :  1378\n",
      "batch no. :  1379\n",
      "batch no. :  1380\n",
      "batch no. :  1381\n",
      "batch no. :  1382\n",
      "batch no. :  1383\n",
      "batch no. :  1384\n",
      "batch no. :  1385\n",
      "batch no. :  1386\n",
      "batch no. :  1387\n",
      "batch no. :  1388\n",
      "batch no. :  1389\n",
      "batch no. :  1390\n",
      "batch no. :  1391\n",
      "batch no. :  1392\n",
      "batch no. :  1393\n",
      "batch no. :  1394\n",
      "batch no. :  1395\n",
      "batch no. :  1396\n",
      "batch no. :  1397\n",
      "batch no. :  1398\n",
      "batch no. :  1399\n",
      "batch no. :  1400\n",
      "batch no. :  1401\n",
      "batch no. :  1402\n",
      "batch no. :  1403\n",
      "batch no. :  1404\n",
      "batch no. :  1405\n",
      "batch no. :  1406\n",
      "batch no. :  1407\n",
      "batch no. :  1408\n",
      "batch no. :  1409\n",
      "batch no. :  1410\n",
      "batch no. :  1411\n",
      "batch no. :  1412\n",
      "batch no. :  1413\n",
      "batch no. :  1414\n",
      "batch no. :  1415\n",
      "batch no. :  1416\n",
      "batch no. :  1417\n",
      "batch no. :  1418\n",
      "batch no. :  1419\n",
      "batch no. :  1420\n",
      "batch no. :  1421\n",
      "batch no. :  1422\n",
      "batch no. :  1423\n",
      "batch no. :  1424\n",
      "batch no. :  1425\n",
      "batch no. :  1426\n",
      "batch no. :  1427\n",
      "batch no. :  1428\n",
      "batch no. :  1429\n",
      "batch no. :  1430\n",
      "batch no. :  1431\n",
      "batch no. :  1432\n",
      "batch no. :  1433\n",
      "batch no. :  1434\n",
      "batch no. :  1435\n",
      "batch no. :  1436\n",
      "batch no. :  1437\n",
      "batch no. :  1438\n",
      "batch no. :  1439\n",
      "batch no. :  1440\n",
      "batch no. :  1441\n",
      "batch no. :  1442\n",
      "batch no. :  1443\n",
      "batch no. :  1444\n",
      "batch no. :  1445\n",
      "batch no. :  1446\n",
      "batch no. :  1447\n",
      "batch no. :  1448\n",
      "batch no. :  1449\n",
      "batch no. :  1450\n",
      "batch no. :  1451\n",
      "batch no. :  1452\n",
      "batch no. :  1453\n",
      "batch no. :  1454\n",
      "batch no. :  1455\n",
      "batch no. :  1456\n",
      "batch no. :  1457\n",
      "batch no. :  1458\n",
      "batch no. :  1459\n",
      "batch no. :  1460\n",
      "batch no. :  1461\n",
      "batch no. :  1462\n",
      "batch no. :  1463\n",
      "batch no. :  1464\n",
      "batch no. :  1465\n",
      "batch no. :  1466\n",
      "batch no. :  1467\n",
      "batch no. :  1468\n",
      "batch no. :  1469\n",
      "batch no. :  1470\n",
      "batch no. :  1471\n",
      "batch no. :  1472\n",
      "batch no. :  1473\n",
      "batch no. :  1474\n",
      "batch no. :  1475\n",
      "batch no. :  1476\n",
      "batch no. :  1477\n",
      "batch no. :  1478\n",
      "batch no. :  1479\n",
      "batch no. :  1480\n",
      "batch no. :  1481\n",
      "batch no. :  1482\n",
      "batch no. :  1483\n",
      "batch no. :  1484\n",
      "batch no. :  1485\n",
      "batch no. :  1486\n",
      "batch no. :  1487\n",
      "batch no. :  1488\n",
      "batch no. :  1489\n",
      "batch no. :  1490\n",
      "batch no. :  1491\n",
      "batch no. :  1492\n",
      "batch no. :  1493\n",
      "batch no. :  1494\n",
      "batch no. :  1495\n",
      "batch no. :  1496\n",
      "batch no. :  1497\n",
      "batch no. :  1498\n",
      "batch no. :  1499\n",
      "batch no. :  1500\n",
      "batch no. :  1501\n",
      "batch no. :  1502\n",
      "batch no. :  1503\n",
      "batch no. :  1504\n",
      "batch no. :  1505\n",
      "batch no. :  1506\n",
      "batch no. :  1507\n"
     ]
    }
   ],
   "source": [
    "train_labels_set = []\n",
    "train_preds_set = []\n",
    "for idx, data in enumerate(dataloader.train):\n",
    "    # get the inputs\n",
    "    model.eval()\n",
    "    print(\"batch no. : \", idx)\n",
    "    inputs, labels = data\n",
    "    train_labels_set += labels.tolist()\n",
    "    inputs, labels = inputs.cuda(), labels.cuda()\n",
    "#     print(labels)\n",
    "    outputs = model(inputs)\n",
    "    _, preds = torch.max(outputs.data, 1)\n",
    "    train_preds_set += preds.cpu().tolist()\n",
    "#     print(labels_set, preds_set)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8728260869565218"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_pred=train_preds_set, y_true=train_labels_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.86      0.93      0.89      1031\n",
      "           1       0.90      0.80      0.85       809\n",
      "\n",
      "    accuracy                           0.87      1840\n",
      "   macro avg       0.88      0.87      0.87      1840\n",
      "weighted avg       0.87      0.87      0.87      1840\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_pred=train_preds_set, y_true=train_labels_set))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "batch no. :  0\n",
      "batch no. :  1\n",
      "batch no. :  2\n",
      "batch no. :  3\n",
      "batch no. :  4\n",
      "batch no. :  5\n",
      "batch no. :  6\n",
      "batch no. :  7\n",
      "batch no. :  8\n",
      "batch no. :  9\n",
      "batch no. :  10\n",
      "batch no. :  11\n",
      "batch no. :  12\n",
      "batch no. :  13\n",
      "batch no. :  14\n",
      "batch no. :  15\n",
      "batch no. :  16\n",
      "batch no. :  17\n",
      "batch no. :  18\n",
      "batch no. :  19\n",
      "batch no. :  20\n",
      "batch no. :  21\n",
      "batch no. :  22\n",
      "batch no. :  23\n",
      "batch no. :  24\n",
      "batch no. :  25\n",
      "batch no. :  26\n",
      "batch no. :  27\n",
      "batch no. :  28\n",
      "batch no. :  29\n",
      "batch no. :  30\n",
      "batch no. :  31\n",
      "batch no. :  32\n",
      "batch no. :  33\n",
      "batch no. :  34\n",
      "batch no. :  35\n",
      "batch no. :  36\n",
      "batch no. :  37\n",
      "batch no. :  38\n",
      "batch no. :  39\n",
      "batch no. :  40\n",
      "batch no. :  41\n",
      "batch no. :  42\n",
      "batch no. :  43\n",
      "batch no. :  44\n",
      "batch no. :  45\n",
      "batch no. :  46\n",
      "batch no. :  47\n",
      "batch no. :  48\n",
      "batch no. :  49\n",
      "batch no. :  50\n",
      "batch no. :  51\n",
      "batch no. :  52\n",
      "batch no. :  53\n",
      "batch no. :  54\n",
      "batch no. :  55\n",
      "batch no. :  56\n",
      "batch no. :  57\n",
      "batch no. :  58\n",
      "batch no. :  59\n",
      "batch no. :  60\n",
      "batch no. :  61\n",
      "batch no. :  62\n",
      "batch no. :  63\n",
      "batch no. :  64\n",
      "batch no. :  65\n",
      "batch no. :  66\n",
      "batch no. :  67\n",
      "batch no. :  68\n",
      "batch no. :  69\n",
      "batch no. :  70\n",
      "batch no. :  71\n",
      "batch no. :  72\n",
      "batch no. :  73\n",
      "batch no. :  74\n",
      "batch no. :  75\n",
      "batch no. :  76\n",
      "batch no. :  77\n",
      "batch no. :  78\n",
      "batch no. :  79\n",
      "batch no. :  80\n",
      "batch no. :  81\n",
      "batch no. :  82\n",
      "batch no. :  83\n",
      "batch no. :  84\n",
      "batch no. :  85\n",
      "batch no. :  86\n",
      "batch no. :  87\n",
      "batch no. :  88\n",
      "batch no. :  89\n",
      "batch no. :  90\n",
      "batch no. :  91\n",
      "batch no. :  92\n",
      "batch no. :  93\n",
      "batch no. :  94\n",
      "batch no. :  95\n",
      "batch no. :  96\n",
      "batch no. :  97\n",
      "batch no. :  98\n",
      "batch no. :  99\n",
      "batch no. :  100\n",
      "batch no. :  101\n",
      "batch no. :  102\n",
      "batch no. :  103\n",
      "batch no. :  104\n",
      "batch no. :  105\n",
      "batch no. :  106\n",
      "batch no. :  107\n",
      "batch no. :  108\n",
      "batch no. :  109\n",
      "batch no. :  110\n",
      "batch no. :  111\n",
      "batch no. :  112\n",
      "batch no. :  113\n",
      "batch no. :  114\n",
      "batch no. :  115\n",
      "batch no. :  116\n",
      "batch no. :  117\n",
      "batch no. :  118\n",
      "batch no. :  119\n",
      "batch no. :  120\n",
      "batch no. :  121\n",
      "batch no. :  122\n",
      "batch no. :  123\n",
      "batch no. :  124\n",
      "batch no. :  125\n",
      "batch no. :  126\n",
      "batch no. :  127\n",
      "batch no. :  128\n",
      "batch no. :  129\n",
      "batch no. :  130\n",
      "batch no. :  131\n",
      "batch no. :  132\n",
      "batch no. :  133\n",
      "batch no. :  134\n",
      "batch no. :  135\n",
      "batch no. :  136\n",
      "batch no. :  137\n",
      "batch no. :  138\n",
      "batch no. :  139\n",
      "batch no. :  140\n",
      "batch no. :  141\n",
      "batch no. :  142\n",
      "batch no. :  143\n",
      "batch no. :  144\n",
      "batch no. :  145\n",
      "batch no. :  146\n",
      "batch no. :  147\n",
      "batch no. :  148\n",
      "batch no. :  149\n",
      "batch no. :  150\n",
      "batch no. :  151\n",
      "batch no. :  152\n",
      "batch no. :  153\n",
      "batch no. :  154\n",
      "batch no. :  155\n",
      "batch no. :  156\n",
      "batch no. :  157\n",
      "batch no. :  158\n",
      "batch no. :  159\n",
      "batch no. :  160\n",
      "batch no. :  161\n",
      "batch no. :  162\n",
      "batch no. :  163\n",
      "batch no. :  164\n",
      "batch no. :  165\n",
      "batch no. :  166\n",
      "batch no. :  167\n",
      "batch no. :  168\n",
      "batch no. :  169\n",
      "batch no. :  170\n",
      "batch no. :  171\n",
      "batch no. :  172\n",
      "batch no. :  173\n",
      "batch no. :  174\n",
      "batch no. :  175\n",
      "batch no. :  176\n",
      "batch no. :  177\n",
      "batch no. :  178\n"
     ]
    }
   ],
   "source": [
    "valid_labels_set = []\n",
    "valid_preds_set = []\n",
    "for idx, data in enumerate(dataloader.valid):\n",
    "    # get the inputs\n",
    "    print(\"batch no. : \", idx)\n",
    "    inputs, labels = data\n",
    "    valid_labels_set += labels.tolist()\n",
    "    inputs, labels = inputs.cuda(), labels.cuda()\n",
    "#     print(labels)\n",
    "    outputs = model(inputs)\n",
    "    _, preds = torch.max(outputs.data, 1)\n",
    "    valid_preds_set += preds.cpu().tolist()\n",
    "#     print(labels_set, preds_set)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6731843575418994"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_pred=valid_preds_set, y_true=valid_labels_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.60      0.75         5\n",
      "           1       0.33      0.20      0.25         5\n",
      "           2       0.00      0.00      0.00         3\n",
      "           3       0.67      0.80      0.73        10\n",
      "           4       0.00      0.00      0.00         4\n",
      "           5       0.00      0.00      0.00         3\n",
      "           6       0.00      0.00      0.00         4\n",
      "           7       0.67      0.86      0.75         7\n",
      "           8       0.68      0.72      0.70        18\n",
      "           9       0.54      0.70      0.61        10\n",
      "          10       0.71      1.00      0.83         5\n",
      "          11       0.75      1.00      0.86         3\n",
      "          12       0.50      0.33      0.40         3\n",
      "          13       0.75      0.75      0.75         4\n",
      "          14       0.60      0.75      0.67         4\n",
      "          15       1.00      0.33      0.50         6\n",
      "          16       0.00      0.00      0.00         3\n",
      "          17       0.38      0.60      0.46         5\n",
      "          18       0.77      0.89      0.83        19\n",
      "          19       1.00      0.67      0.80         9\n",
      "          20       0.56      1.00      0.71         5\n",
      "          21       0.67      0.50      0.57         4\n",
      "          22       0.73      0.80      0.76        10\n",
      "          23       0.83      0.83      0.83         6\n",
      "          24       0.81      0.89      0.85        19\n",
      "          25       1.00      0.40      0.57         5\n",
      "          26       0.60      0.60      0.60         5\n",
      "          27       0.57      0.40      0.47        10\n",
      "          28       0.55      1.00      0.71        11\n",
      "          29       0.75      0.60      0.67         5\n",
      "          30       0.76      0.86      0.81        22\n",
      "          31       0.60      1.00      0.75         3\n",
      "          32       0.83      0.87      0.85        23\n",
      "          33       0.43      0.60      0.50        10\n",
      "          34       0.00      0.00      0.00         3\n",
      "          35       1.00      0.75      0.86         4\n",
      "          36       0.50      1.00      0.67         3\n",
      "          37       0.73      1.00      0.84         8\n",
      "          38       0.33      0.50      0.40         4\n",
      "          39       0.85      0.73      0.79        15\n",
      "          40       0.33      0.33      0.33         3\n",
      "          41       0.60      0.92      0.73        13\n",
      "          42       0.29      0.40      0.33         5\n",
      "          43       1.00      0.33      0.50         3\n",
      "          44       0.56      0.67      0.61        15\n",
      "          45       0.55      0.65      0.59        17\n",
      "          46       0.88      0.88      0.88         8\n",
      "          47       0.44      0.78      0.56         9\n",
      "          48       0.83      0.83      0.83         6\n",
      "          49       1.00      0.25      0.40         4\n",
      "          50       1.00      0.25      0.40         4\n",
      "          51       1.00      0.80      0.89         5\n",
      "          52       0.00      0.00      0.00         3\n",
      "          53       0.67      0.67      0.67         3\n",
      "          54       0.90      0.90      0.90        10\n",
      "          55       1.00      0.92      0.96        12\n",
      "          56       1.00      0.29      0.44         7\n",
      "          57       1.00      0.93      0.96        14\n",
      "          58       0.56      1.00      0.71         5\n",
      "          59       0.75      0.60      0.67         5\n",
      "          60       0.60      0.30      0.40        10\n",
      "          61       0.80      0.73      0.76        11\n",
      "          62       0.67      0.67      0.67        15\n",
      "          63       0.65      0.65      0.65        17\n",
      "          64       0.38      1.00      0.55         3\n",
      "          65       0.67      0.50      0.57         4\n",
      "          66       0.70      1.00      0.82         7\n",
      "          67       0.65      0.85      0.73        13\n",
      "          68       0.50      0.50      0.50         4\n",
      "          69       0.43      0.50      0.46         6\n",
      "          70       0.00      0.00      0.00         3\n",
      "          71       0.69      0.75      0.72        12\n",
      "          72       0.50      0.50      0.50         4\n",
      "          73       0.40      0.67      0.50         3\n",
      "          74       0.75      0.38      0.50         8\n",
      "          75       0.38      0.43      0.40         7\n",
      "          76       0.25      0.25      0.25         4\n",
      "          77       0.50      0.25      0.33        12\n",
      "          78       0.67      0.17      0.27        12\n",
      "          79       1.00      0.33      0.50         3\n",
      "          80       0.50      0.33      0.40         3\n",
      "          81       1.00      0.75      0.86         4\n",
      "          82       1.00      1.00      1.00         8\n",
      "          83       0.00      0.00      0.00         3\n",
      "          84       1.00      0.50      0.67         6\n",
      "          85       1.00      0.67      0.80         3\n",
      "          86       1.00      0.75      0.86         4\n",
      "          87       0.76      0.76      0.76        21\n",
      "          88       0.65      0.69      0.67        16\n",
      "          89       0.60      0.75      0.67         8\n",
      "          90       0.67      0.89      0.76         9\n",
      "          91       0.83      1.00      0.91         5\n",
      "          92       0.86      0.67      0.75         9\n",
      "          93       1.00      0.33      0.50         3\n",
      "\n",
      "    accuracy                           0.67       716\n",
      "   macro avg       0.64      0.60      0.59       716\n",
      "weighted avg       0.68      0.67      0.65       716\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_pred=valid_preds_set, y_true=valid_labels_set))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tensor([[-1.1673, -0.2164]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1175, -1.4215]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0916, -3.4097]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3203, -0.9870]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.9048, -1.5405]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0522, -2.8063]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1136, -2.3834]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7111, -3.9478]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.4123, -3.3514]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2776, -1.2252]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2520, -3.1235]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1294,  1.0293]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8395, -2.0392]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.6889, -3.1920]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7741, -0.1292]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1864, -3.5444]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5541, -1.2779]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.0713, 1.3056]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7842, -0.1954]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0858, -1.1561]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2493, -0.9807]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5672, -0.4622]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8731, 0.1573]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7977, -0.0190]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0106, 0.5682]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5795, -0.4562]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.2733, -5.2885]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1738, 0.4287]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.2398, -2.1526]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2043, -1.9697]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0256, -0.4382]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1617, -2.1974]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8668, -0.2576]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1572, 0.2950]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9823, -0.5004]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 6.2838e-05, -1.8517e+00]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0028, -1.9786]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0577, -2.6379]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.5963, 0.6486]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.8406, -3.2111]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1254, -0.0922]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.8227, -0.2319]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3301, -1.8567]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0117, -3.5311]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1803, -2.6689]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9588, -1.1507]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4955, -0.6411]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1880, -2.0008]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4970, -1.1149]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3872, -1.1890]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5505, -0.6584]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0718, -4.1589]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.5572, -0.2007]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0484, -0.7548]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0835, -2.5675]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4253, -2.9504]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9472, -0.4062]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.5228, -0.3454]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3905, -1.3289]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7872, -0.1771]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3593, -1.5574]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5007, -0.9292]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-2.0704, -2.3775]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1697, -3.6508]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6088, -1.4529]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3294, -0.2743]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0294, -2.9252]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.3963, 0.4332]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.6095, -4.4847]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6031, 0.8085]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.6312, -3.9942]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3070, -0.1071]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4553, -1.7609]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1045, -0.7977]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1791, 0.3531]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8385, 0.2887]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5758, -1.2775]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.5202, -6.2216]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7252, -0.5163]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7752, -0.4127]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2360, -0.8524]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6862, 0.9611]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0757,  0.2711]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9696, -0.1933]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1379, -0.7391]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.5132, 0.6981]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5150, 0.6093]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3434, -0.2869]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4603e-03, -2.1190e+00]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0844, -0.3300]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.1839, 0.0281]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6405, -2.3632]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.6082, -1.9826]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 8.1506e-01, -3.2523e-04]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9523, 0.1736]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3933, -2.1009]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6763, -0.9494]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5832, -1.3671]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0223, -0.6032]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9949, -2.5346]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6116, -1.2521]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4650, -1.5464]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.4831, 0.3547]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8521, 0.5180]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1992, 0.7503]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6142, -0.6342]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0644, -1.5430]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3526, -2.9997]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3643, -1.7772]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5609, -1.9488]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2732, -0.5377]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2860, -0.4733]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.9458, -1.7540]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5728, -0.3686]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5571, -2.2649]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3781, -0.4348]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4783, -2.0994]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0707, 1.0123]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1473, -1.5165]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3721, -3.0098]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3083, -0.8813]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1496, -0.6377]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2859, 0.6574]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2446, -2.2846]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3455, -0.6020]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1083, -2.0273]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5853, -1.1772]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4697, -5.3933]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9505, -0.9375]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0068, -0.9285]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5208, -0.1262]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3971, -0.1968]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5638, -2.4642]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4304, -0.1323]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3752, -2.7415]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.6398,  0.0583]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8843, -0.5594]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7022, -2.3892]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2965, -2.6241]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7575, -0.4081]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.2239, 0.4443]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1097, -2.4076]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1930, -0.5522]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.7396, -6.5396]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.8506, -3.2803]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4944, -2.2484]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4480, -2.7008]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7205, -1.4036]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0045, -0.5179]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0506, -0.4957]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4850, -1.7284]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0181, -2.9691]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.5552, 1.1028]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4703, -1.4501]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2159, -1.2924]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1336, -1.6634]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7165, -1.7608]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9598, 0.2126]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7846, -0.5448]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9444, 1.2854]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7631, -4.6301]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0956, -0.1041]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2350, -2.2595]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3373, -3.0154]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.9104, 0.9173]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2637, -0.3231]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0882, 0.2457]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0739, -3.1219]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3427, -1.9083]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1072, -1.9655]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5885, -0.6465]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7794, 0.1567]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.8627, -4.8995]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7574, -2.6044]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3254, -1.2128]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3043, -1.5315]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5858, -0.9811]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.3765, 1.0498]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7225, 1.1752]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6570, 1.7965]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0400, -0.3908]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.3378, 0.3434]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0904, -2.5823]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.2546, 0.2071]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.0058, 1.0910]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7888, 0.9465]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4558, -3.2402]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0118, -0.4489]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5555, -4.5323]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.2639, -4.7488]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9185, -1.8260]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2509, 0.4886]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.9205, -3.6179]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2579, -1.6891]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1406, -1.2175]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1314, 0.5433]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8620, 0.0233]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6841, -2.0965]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0352, -0.3605]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5525, -1.9340]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6084, -0.2990]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.6542, -2.6477]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6619, -0.6637]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2462, -0.1222]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4825, -0.2781]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.3278, 0.4645]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2411, -2.8536]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9453, -0.6593]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2337, 0.0777]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-2.4725, -5.0293]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.6614, -1.9119]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6682, -0.9926]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.8520, -2.8934]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0387, 0.5220]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0393, 0.5437]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6441, 1.2576]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.2174, 1.2973]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0514, -1.9004]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4104, -0.6419]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7015, -2.5724]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0182, -1.4353]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.6895, -3.0893]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.2995, -3.8825]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4827, -1.5381]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7317, -1.6312]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0562, -1.8516]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8000, -0.5525]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2290, 0.2286]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0204, -1.2423]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.4019, -3.6435]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5194, -1.9675]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[9.0636e-01, 4.4882e-05]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8025, -0.2325]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7378, -0.2105]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6556, 1.6677]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6411, 1.2308]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.7473, -1.9133]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2929, -0.2938]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8521, 1.8275]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6978, 1.3247]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8621, 0.5662]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5402, 0.7588]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.7169, -3.7939]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1543, -0.8531]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1115, -1.7547]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8370, -2.2908]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2057, -0.9380]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3789, -2.9074]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1655, -1.5906]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7310, -1.8390]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5545, -0.3962]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2473, -3.3986]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7416, -0.7887]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1961, 0.0017]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3845, -0.5072]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4074, -3.8871]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1156, -2.8772]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0054,  1.3011]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0205, -1.6140]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0139, -1.5141]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3582, -0.6461]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.8307, -0.1144]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1348, -4.5524]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1872, 0.2135]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1386, -0.2798]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2198, -1.4166]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7895, -1.6380]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1592, -3.6316]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7922, -0.3303]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3540, -2.3570]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7527, -1.3427]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1878, 0.5744]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3948, -1.6400]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.9485, -0.5157]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3174, -4.4381]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.6139, -1.3849]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6874, -0.6350]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7099, -4.2372]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1018, -1.9332]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.4147, -3.1168]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1742, -1.0902]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1726, -1.8467]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0546, -1.8240]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3779, -2.7615]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0501, -1.3183]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0595, -0.1505]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.5332, 0.2794]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3696, -0.4920]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9521, -0.9895]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3920, -0.7467]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7352, -1.7377]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.5914, -0.6191]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4333, -0.7817]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6785, 0.2195]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3796, -2.1166]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6259, -0.2460]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7263, -1.2481]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7041, -1.6683]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2236, 1.1780]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4513, -0.4899]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5070, -2.4922]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7016, -4.9132]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8004, -2.2229]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0459, -3.2011]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6612, -0.2608]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6414, 0.2059]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.8252, -1.2329]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3474, -1.0832]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6635, -1.0329]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8518, -0.2408]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3005, -1.6974]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0480, 0.1126]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.4777, 3.0459]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9109, 0.4535]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9986, -0.6319]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3128, -2.3280]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2064, -1.1941]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8847, -0.2355]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8921, 0.8651]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.5820, -2.1295]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1179, -3.8908]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2402, -0.0033]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1618, 0.9584]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4643, -0.4604]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0423, -2.3469]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0870, -1.4366]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5605, -0.3544]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5663, 0.3850]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4860, -0.7778]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6106, -1.6629]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2599, -0.1356]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4797, -0.0198]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2689, -3.2263]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3964, -0.1151]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1559, -1.1033]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7879, -1.2706]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3226, -1.8094]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4956, -1.6443]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7938, -0.9817]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1946, -2.1573]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1358, -4.3674]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6185, 0.0896]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5019, -2.3339]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5799, 0.2492]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7347, -0.8160]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4997, -3.5607]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6900, -0.9345]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.9665, -2.5185]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1151, -2.9529]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1087, 0.9007]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-3.2613, -5.0380]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1739, -1.8722]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.2891, 0.1808]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6949, -1.1488]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5888, -4.7416]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7892, -1.6882]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3994, -1.3697]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4412, -1.2463]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0655, -3.1536]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.6650, -0.2869]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5268, -0.8062]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0279, -1.2239]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.0553, 1.3936]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8710, -1.0665]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.6060, -3.8303]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0562, -2.6977]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.3683, 1.9927]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5529, -0.0624]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7874, -0.7867]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2537, -2.2790]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9207, -0.6283]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8710, 0.0589]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3328, -1.0560]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2837, -1.5463]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-2.1871, -3.7209]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.8523, -3.5819]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2985, -1.6785]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1989, -0.0232]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.9809, 0.3134]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0354, -0.7833]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7820, -2.2521]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8317, 1.7936]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3934, -0.4933]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2742, -1.4553]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.6596, -0.7426]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6276, 0.6574]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3801, -0.5003]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.5445, -1.2098]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2977, -0.1450]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4105, -1.7491]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6496, 1.8702]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.5137, 0.2222]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8240, 1.1197]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1208, -1.0618]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5309, -1.8064]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0484, -2.1431]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1631, -2.1731]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-2.3183, -3.6011]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9093, -0.8795]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4783, -0.0030]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5397, -2.0383]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0267, -1.1121]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4961,  0.2316]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1432, -0.2521]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2705, -0.1663]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0094, -2.6670]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9673, -0.7201]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3130, -4.3022]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4593, -0.7490]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8424, 0.7405]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7082, -3.8023]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8968, -0.2014]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9405, -1.3596]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-3.0999, -6.0852]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1015, -0.8606]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5319, -1.5757]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.6407,  0.1997]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4926, -2.7722]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1080, -0.9431]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2895, 0.9233]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-3.1169, -5.8688]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0476, 1.3413]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2827, -0.1314]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1546, -4.7721]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8926, -2.2830]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.9193, 0.5545]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0408, -2.0701]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4373, -0.0056]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1556, -0.6265]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7573, -3.7333]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.6291, -2.1687]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5032, 0.4625]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9755, -1.9556]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2037, -2.3303]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7232, 1.0361]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8622, 0.6700]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.1697, 0.6848]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3930, -0.4264]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8097, 0.5311]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9369, 2.7555]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.9514, -4.3412]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4103, -1.1492]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.1933, 1.5091]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7522, 2.6349]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2820, 0.4107]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9631, -2.3615]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1153, -4.4124]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5568, -0.6842]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1441, -0.7871]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8764, -2.2176]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3881, -1.7959]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2636, 1.0455]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9503, -0.5321]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2356, -1.2671]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5215, -0.3537]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7181, 1.4789]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2666, -1.7545]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6792, 0.3124]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2307, 0.6949]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2540, -0.4174]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2182,  1.0734]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5954, -0.9714]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2590, -1.7290]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7804, 0.5886]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-2.0282, -2.8919]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0306, -1.1622]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7906, 1.3787]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7643, -1.0876]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6445, -1.2697]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0673, -3.3423]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5481, -0.4700]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4084, -0.1736]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3538, -0.5792]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0511, -1.7692]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4807, -0.5720]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1808, -1.3580]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8064, 0.1597]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0997, -1.5919]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.8295, -2.7181]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.0006, 0.2164]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.3483, 1.2788]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1241, -1.8250]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8451, 0.3723]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4510, -0.7681]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8733, -0.8705]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6441, -0.0970]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3759, -0.3163]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2333, -0.9801]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0475, -4.2016]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5515, -2.2153]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6904, 0.7633]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0728, -1.8329]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9484, -1.0166]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0931, -1.9106]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5718, -3.6694]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1473, -3.9351]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8830, -0.5449]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3794, -4.0344]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2169, -1.6082]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9340, -1.2972]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.2468, -1.7264]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4504, -0.2705]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7985, -0.9238]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3219, -1.3874]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.5426, 0.0575]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0887, -0.7764]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1422, -1.9153]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6069, -0.3360]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0311, -0.8950]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0143, -1.5527]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7293, -3.1055]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1618, -1.4803]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.4866, 1.6226]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6115, 0.0135]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0198, -1.2339]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.0811, -0.6849]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0314, -0.5816]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2159, -3.8230]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9777, -0.5594]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7003, -1.1133]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5080, -4.5736]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5570, 0.5869]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4553, -3.4607]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8158, -3.4075]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2111, 0.0435]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1909, -4.1094]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0848, 1.5909]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9267, 0.4748]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3102, -0.9948]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0397, -3.5004]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[2.3955, 1.7508]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1940, -1.7580]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.3775, 1.0711]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1870, -0.3219]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.2433, 0.4605]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8649, 0.2720]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9370, -0.3423]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1483, -3.0533]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-2.5110, -4.2947]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8329, -2.3375]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3603, -2.8804]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.4482, 0.0478]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4646, -2.1172]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.0196, -1.9531]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9201, 0.3225]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3147, -0.9028]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7406, -2.1625]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3595, -1.7877]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3468, -0.0443]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.9495, -3.0798]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3688, -0.3074]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.5851, 0.1137]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.6026, -2.6798]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5645, -1.0849]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1661, -1.3827]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.2425, 0.9780]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1053, -1.1304]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.9258, -2.6760]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3023, -2.8460]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5617, -1.0146]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9499, -1.0198]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5897, -3.5274]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.5165, -1.8513]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3859, -1.3680]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7673, -1.9551]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1572, -0.8401]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.4732, 1.4332]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1801, -3.3591]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.1794, -1.3014]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1872, -0.8697]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4137, -1.9505]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.3550, 2.0835]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7925, 0.6600]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6918, -2.5138]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6214, -1.4909]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.4544, 0.1431]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1393,  0.5347]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0816, -2.4935]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.6187, -0.0335]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7458, -4.2572]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4651, -1.9313]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.1743, -2.7449]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8001, 0.3074]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3903, -0.2083]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5850, -1.6110]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6244, -1.7519]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.7842, 0.0053]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9588, -0.4149]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9546, -0.0851]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.4146, 0.4208]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.1573, 0.3606]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.3091, -0.9326]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.9880, 0.8345]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4779, -1.1054]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.4288, 0.8571]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4608, -0.6093]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3522, -2.8759]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.6977, 0.1185]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0467, -1.0167]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6972, -0.4724]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.4471, -5.5531]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2862, -1.5572]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.3268, -2.6613]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.2621, -0.0318]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6155, -1.7353]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6341, -1.6060]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3081, -1.2311]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.5675, -1.7683]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8115, 1.4494]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8144, -0.2226]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8952, -0.2370]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5782, 0.4145]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.5669, 0.3477]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.7589, 0.6936]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7697, -3.3416]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8734, 0.5210]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.3052, 0.2670]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.4238, -0.0483]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.3749, 0.9036]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.2863, -1.6988]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2457, -0.4887]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.2006, -0.6630]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.0543, -1.9770]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.6850, 0.3774]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.7516, -4.1780]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.3501, -0.5694]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4257, -0.1025]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6410, -0.6486]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.1495, -1.4545]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.4758, -2.6206]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.5274, -2.3287]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-1.3144, -5.1455]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.9926, 1.0033]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.1418, -1.3121]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.2774, 0.7618]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.0107, 0.4625]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.7698, -0.6860]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.6563, -2.1693]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 1.7915, -0.2018]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8789, -0.7258]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[0.8153, 0.1047]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.0317,  2.3950]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.4798, -0.8649]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8578, -0.1814]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.9189, -0.4678]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5805, -1.8266]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[-0.5877, -2.6481]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[1.8791, 0.4744]], device='cuda:0', grad_fn=<MmBackward>)\n",
      "tensor([[ 0.8929, -2.5863]], device='cuda:0', grad_fn=<MmBackward>)\n"
     ]
    }
   ],
   "source": [
    "test_labels_set = []\n",
    "test_preds_set = []\n",
    "for idx, data in enumerate(testdata.train):\n",
    "    # get the inputs\n",
    "    model.eval()\n",
    "    inputs, labels = data\n",
    "    test_labels_set += labels.tolist()\n",
    "    inputs, labels = inputs.cuda(), labels.cuda()\n",
    "#     print(labels)\n",
    "    outputs = model(inputs)\n",
    "    _, preds = torch.max(outputs.data, 1)\n",
    "    test_preds_set += preds.cpu().tolist()\n",
    "    print(outputs)\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6083333333333333"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_pred=test_preds_set, y_true=test_labels_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.00      0.00      0.00         7\n",
      "           1       0.38      0.43      0.40         7\n",
      "           2       0.00      0.00      0.00         4\n",
      "           3       0.71      0.91      0.80        11\n",
      "           4       0.83      0.83      0.83         6\n",
      "           5       1.00      0.50      0.67         4\n",
      "           6       0.00      0.00      0.00         6\n",
      "           7       0.44      0.57      0.50         7\n",
      "           8       0.56      0.53      0.54        19\n",
      "           9       0.57      0.73      0.64        11\n",
      "          10       0.43      0.50      0.46         6\n",
      "          11       1.00      0.40      0.57         5\n",
      "          12       1.00      0.25      0.40         4\n",
      "          13       0.75      0.60      0.67         5\n",
      "          14       1.00      0.50      0.67         6\n",
      "          15       1.00      0.57      0.73         7\n",
      "          16       0.00      0.00      0.00         5\n",
      "          17       0.62      0.83      0.71         6\n",
      "          18       0.67      0.84      0.74        19\n",
      "          19       0.50      0.90      0.64        10\n",
      "          20       0.50      0.33      0.40         6\n",
      "          21       0.80      0.80      0.80         5\n",
      "          22       0.58      1.00      0.73        11\n",
      "          23       0.33      0.25      0.29         8\n",
      "          24       0.43      0.76      0.55        21\n",
      "          25       0.67      0.80      0.73         5\n",
      "          26       0.75      0.50      0.60         6\n",
      "          27       0.20      0.17      0.18        12\n",
      "          28       0.24      0.42      0.30        12\n",
      "          29       0.67      0.86      0.75         7\n",
      "          30       0.76      0.86      0.81        22\n",
      "          31       0.22      0.50      0.31         4\n",
      "          32       0.67      0.80      0.73        25\n",
      "          33       0.53      0.83      0.65        12\n",
      "          34       0.50      0.40      0.44         5\n",
      "          35       0.80      0.67      0.73         6\n",
      "          36       0.33      0.50      0.40         4\n",
      "          37       0.89      0.89      0.89         9\n",
      "          38       0.14      0.20      0.17         5\n",
      "          39       0.73      0.94      0.82        17\n",
      "          40       0.67      0.50      0.57         4\n",
      "          41       0.65      0.73      0.69        15\n",
      "          42       1.00      0.33      0.50         6\n",
      "          43       1.00      0.50      0.67         4\n",
      "          44       0.67      0.50      0.57        16\n",
      "          45       0.50      0.50      0.50        18\n",
      "          46       1.00      0.67      0.80         9\n",
      "          47       0.29      0.40      0.33        10\n",
      "          48       0.62      0.62      0.62         8\n",
      "          49       0.50      0.33      0.40         6\n",
      "          50       0.50      0.17      0.25         6\n",
      "          51       0.40      0.33      0.36         6\n",
      "          52       0.00      0.00      0.00         4\n",
      "          53       0.00      0.00      0.00         4\n",
      "          54       0.83      0.83      0.83        12\n",
      "          55       0.81      1.00      0.90        13\n",
      "          56       0.80      0.50      0.62         8\n",
      "          57       1.00      0.81      0.90        16\n",
      "          58       0.50      1.00      0.67         6\n",
      "          59       0.75      0.50      0.60         6\n",
      "          60       0.60      0.25      0.35        12\n",
      "          61       0.29      0.18      0.22        11\n",
      "          62       1.00      0.88      0.94        17\n",
      "          63       0.72      0.72      0.72        18\n",
      "          64       1.00      1.00      1.00         5\n",
      "          65       0.50      0.40      0.44         5\n",
      "          66       0.73      0.89      0.80         9\n",
      "          67       0.55      0.86      0.67        14\n",
      "          68       0.50      0.40      0.44         5\n",
      "          69       0.38      0.71      0.50         7\n",
      "          70       0.00      0.00      0.00         4\n",
      "          71       0.79      0.85      0.81        13\n",
      "          72       0.00      0.00      0.00         6\n",
      "          73       0.00      0.00      0.00         3\n",
      "          74       0.60      0.33      0.43         9\n",
      "          75       0.00      0.00      0.00         9\n",
      "          76       1.00      0.40      0.57         5\n",
      "          77       0.33      0.31      0.32        13\n",
      "          78       0.44      0.31      0.36        13\n",
      "          79       0.00      0.00      0.00         3\n",
      "          80       0.60      0.60      0.60         5\n",
      "          81       0.67      0.67      0.67         6\n",
      "          82       0.91      1.00      0.95        10\n",
      "          83       1.00      0.40      0.57         5\n",
      "          84       0.86      0.75      0.80         8\n",
      "          85       0.67      0.40      0.50         5\n",
      "          86       1.00      0.40      0.57         5\n",
      "          87       0.86      0.83      0.84        23\n",
      "          88       0.64      0.53      0.58        17\n",
      "          89       0.67      0.80      0.73        10\n",
      "          90       0.82      0.90      0.86        10\n",
      "          91       0.80      0.57      0.67         7\n",
      "          92       0.67      1.00      0.80        10\n",
      "          93       0.67      0.50      0.57         4\n",
      "\n",
      "    accuracy                           0.61       840\n",
      "   macro avg       0.58      0.54      0.54       840\n",
      "weighted avg       0.61      0.61      0.59       840\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/opt/conda/lib/python3.6/site-packages/sklearn/metrics/classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_pred=test_preds_set, y_true=test_labels_set))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>imagename</th>\n",
       "      <th>captions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test_img_0.jpg</td>\n",
       "      <td>sugarkaga nadandhava vida figuregaga nadandhavan dhan adhigam</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test_img_1.jpg</td>\n",
       "      <td>i have come for my stones      stones thaane.. nalla cheap aana price la namma kalyan jewellers la kedaikkum</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>test_img_2.jpg</td>\n",
       "      <td>\"special porotta\" nu pottuierukke spacial kum sadhavukkum yenna vidhiyasam..?     rendukkum oru naal dhan vidhiyasam..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>test_img_3.jpg</td>\n",
       "      <td>*we : amma .. cooker 3 whistle vanthuchu off panniten...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>test_img_4.jpg</td>\n",
       "      <td>creating whatsapp group - 1st day vaanga ji.. vaanga ji.. ellarayum add pannunga ji TEAM WORK PANROM...ENJOY PANROMM..   AFTER ONE WEEK ... ENNAYA..? IVLO AMAITHIYA IRUKEENGA? ETHACHU PESUNGAYA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>test_img_5.jpg</td>\n",
       "      <td>*FOODIE : YAARUM PAARKKAMAL ENNAI PAARKIREN ENNAI ARIYAMAL UNNAI PARKIREN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>test_img_6.jpg</td>\n",
       "      <td>MARGALI MAADHA EFFECT YENNA KULLURU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>test_img_7.jpg</td>\n",
       "      <td>YENNAMMA FEEL PANNI YELUTHI ERUKRANDA AVAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>test_img_8.jpg</td>\n",
       "      <td>TAMIL SERIALS BE LIKE... INI AVARUKKU BADHIL..... IVAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>test_img_9.jpg</td>\n",
       "      <td>Mama ponnu Expectation reality</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>test_img_10.jpg</td>\n",
       "      <td>pondatti yennai thittum bodhu... eppavum unga amma pechai dhan kepingala.,,? en maganai thittum bodhu... amma pechai ketkave mattiiya...?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>test_img_11.jpg</td>\n",
       "      <td>mattai adakuvatharku en mutha ponna kattikoduppen... yoo maams kannukuttiya adakitten ilaya ponnu ennakkudhan...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>test_img_12.jpg</td>\n",
       "      <td>ada paavi payale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>test_img_13.jpg</td>\n",
       "      <td>*(me)ENGINEER : ennoda maasa  sambalam 30aayiram... unga ponna nalla pathukka idhu podhadha?   *mamanar : en ponnukku naan kodukkura pocket maneyye 25aayiram oru masathukku       sorry mama...adhaiyum sethudha 30aayiram endru kurinen...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>test_img_14.jpg</td>\n",
       "      <td>#10years challenge..... *2008 : NET RECHARGE PANNA 30RS KUDU AATHA... *2018 :  NET RECHARGE PANNA 300RS KUDU AATHA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>test_img_15.jpg</td>\n",
       "      <td>IDHUYENA PRAMADHAM ELECTION VARTTUM IDHAVIDA SPECIAL AACHIYANGALLAN KATHUERUKKU..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>test_img_16.jpg</td>\n",
       "      <td>*PEOPLE: MATRAM MUNNETRAM ENRIRE ADHARKU YENNA ARTHAM  *JINGLE BELL : PETTIGALAI KAI MATRI NAAN MUNNERUVEN  YENDRU ARTHAM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>test_img_17.jpg</td>\n",
       "      <td>*TN PEOPLE : ADA ARAIBODHA NAAYE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>test_img_18.jpg</td>\n",
       "      <td>*SIMBU : NAMMA MASSU ENNA NU KATANUM! ENAKU ANDAL PAAL UTHUNGA!     YENGA VEETULA ANDA ILLA!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>test_img_19.jpg</td>\n",
       "      <td>*MOM : VEETULA IRUNDHA ANDAA VAYUM PAAL PACKET UM KANOM      *BANNER JEE FANS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>test_img_20.jpg</td>\n",
       "      <td>ONLY FOR 90S KIDS SCHOOL DAYS   *PANAKARA VEETU PULLAIYA ERUPPARO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>test_img_21.jpg</td>\n",
       "      <td>BEAUTY CARE  #OTHER GIRLS    FOFR MY BESTIE     #ME : NEE DHINAM KULICHAA POTHUMAE, VERA YATHUVUM VENDAAMAE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>test_img_22.jpg</td>\n",
       "      <td>VATHIYAR : CURRENT POGUM BODHU MUTHALA YEDHAI YEDUPINGA? TORCH LIGHT, EMERGENCY LIGHT?  RENDUMILLAI SIR, APPA PACKETLA ERUKKIRA PANATHAI YEDUPPEN SIR...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>test_img_23.jpg</td>\n",
       "      <td>DEI NAMMA ANDREA DA..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>test_img_24.jpg</td>\n",
       "      <td>#2K KIDS : INNAKKI VENNUMNA NEENGA ORE VEDIYILA 7 SHOT VEDIKKALAM......  #90 KIDS : AANA NANGA ANNAIKE 7 BIJILI VEDIYA ONNA THIRICHU ORE VEDIYILA 7SHOT VEDICHAVUKAPU!!!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>test_img_25.jpg</td>\n",
       "      <td>100CR... 125CR... NEENGA EDHACHU FEEL PANRINGALA??  *SIMBU : CHA CHA NAAN RAJA AH VANTHA ASSAULT AH 150 CR EDUPANGA..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>test_img_26.jpg</td>\n",
       "      <td>DECENTA KENJUFYING MY SISTER .. PROGRESS REPORT LA APPA KITTA SIGN VAANGI KUDUKARIYAA PLEASE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>test_img_27.jpg</td>\n",
       "      <td>THONDARGAL    *CAR OOTUNAR   *VEETUKARAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>test_img_28.jpg</td>\n",
       "      <td>WHEN 90S KIDS WATCHING THIS SERIAL...  ~90'S KIDS  : NAMAKU VEELAIKKAARI AMAIYULA... VEELAIKKARI VAIKURA ALAVUKKU PANACASITHYUM AMAIYULS...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>test_img_29.jpg</td>\n",
       "      <td>COLLEGE AMAIVATHU ELLAM +2 KODUTHA VARAM... OFFICE AMAIVATHU ELLAM AVAN AVAN SEITHA VINAI... NAMAKU COLLEGE'UM OZHUNGA AMAIYALA...OFFICE'UM OZHUNGA AMAIYALA INTHA PAGE'AAVATHU MAINTAIN PANNUVOM..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>test_img_637.jpg</td>\n",
       "      <td>IF YOU ARE BAD  I'M YOUR DAD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>638</th>\n",
       "      <td>test_img_638.jpg</td>\n",
       "      <td>MEERA: NAMMA HUMAN BEING NAMALUKU IRUKKURA MAARI POWER ETHUMAE KEDAIYAATHU ATHA SIXTH SENSE.. ORUTHAR NAMBA PESUMBOTHAE ATHA ENNA 'NU FEEL PANNA MUDIUM.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>639</th>\n",
       "      <td>test_img_639.jpg</td>\n",
       "      <td>GYM TRAINER TO NEW YEAR MOTIVATIONAL GUYS  UNNA KADAISIYA PONA JANUARY 2 DHETHY PATHATHU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>640</th>\n",
       "      <td>test_img_640.jpg</td>\n",
       "      <td>NAA POREN JAANNU   DEII RAM.. VENAM DA SONNA KELU DA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>641</th>\n",
       "      <td>test_img_641.jpg</td>\n",
       "      <td>FOUR TYPES OF BIRTHDAY TREATS  ON EVERY GANG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>642</th>\n",
       "      <td>test_img_642.jpg</td>\n",
       "      <td>90S KIDS DURING INTERVALS  *ME: HEIGHT SOLLADHA .. AVAN KITTA UNDERTAKER IRUKKU!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>643</th>\n",
       "      <td>test_img_643.jpg</td>\n",
       "      <td>THANI ORUVAN UNGA EDAM UNGA AALUNGA.. ENAKU BAYAM ILLA, ATHUKU ITHAAN SAMPLE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>644</th>\n",
       "      <td>test_img_644.jpg</td>\n",
       "      <td>LETS TAKE A SELFIE PULA   *SELFIE KUMAR: SHROOOVVV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>645</th>\n",
       "      <td>test_img_645.jpg</td>\n",
       "      <td>~IN KAATRU VELIYIDAI MOVIE   IDE MAARI COLOUR FULL ANA SONG AH ENGAYO NAMMA PATHURUKAMEY....!!?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>646</th>\n",
       "      <td>test_img_646.jpg</td>\n",
       "      <td>5) TAMIL MATHANGAL KURIPU ELUDHUGA:  1) CHITTRAI  2}VAIKASI 3) AANI 4) AAVADI 5) PURATTASI</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>647</th>\n",
       "      <td>test_img_647.jpg</td>\n",
       "      <td>NO COMMENTS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>648</th>\n",
       "      <td>test_img_648.jpg</td>\n",
       "      <td>KAILA VERA PATHUPAISA ILLA IVAN VERA NAMBALA LOVE PANDRENNU SOLDRAN IVAN KUDA SENDHU APPADIYE VANDIYA OOTRIVENDI DHAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>649</th>\n",
       "      <td>test_img_649.jpg</td>\n",
       "      <td>ENERGY LOW AAGUM VARAI VELAIYADIYADHU ANDHA KALAM  BATTERY LOW AAGUM VARAI VELIYADHUVADHU INDHA KALAM.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>650</th>\n",
       "      <td>test_img_650.jpg</td>\n",
       "      <td>ayyey vaaya paaru</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>651</th>\n",
       "      <td>test_img_651.jpg</td>\n",
       "      <td>no comments</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>652</th>\n",
       "      <td>test_img_652.jpg</td>\n",
       "      <td>MEETS YOU BEFORE EXAM ASKS YOU QUESTIONS YOU'VE NEVER HEARD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>653</th>\n",
       "      <td>test_img_653.jpg</td>\n",
       "      <td>CURRENT THOTTAA SAADHARANA MANUSHANUKU THAAN SHOCK ADIKUM NAAN NARASIMMA.. ENNA THOTTAA ANDHA CURRENTUKE SHOCK ADIKKUM    CAPTAIN Y U NO PRODUCE CURRENT FOR TAMILNADU POWERCUT?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>654</th>\n",
       "      <td>test_img_654.jpg</td>\n",
       "      <td>DAI NANACHUM PADAM MATTUM THA RELEASE PANREN... YENNA NAMBI NEENGA PADAM PAKALAMDA...  AANA IVANAILAM ORU AALA MATHICHU IVANUGALA NAMBI PADATHUKU POVATHINGADA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>655</th>\n",
       "      <td>test_img_655.jpg</td>\n",
       "      <td>ONGI ADICHA ONDRA TON WEIGHT AH??  AIYO, ATHU SATHIYAMA NAAN ILLAINGA NA.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>656</th>\n",
       "      <td>test_img_656.jpg</td>\n",
       "      <td>YENDA UNGA APPA PEYARAI FRIDGEKULLA YELUDHI VACHURUKA.? AVARDHANDA SONNARU EN PEYAR KETTUPOGAMMA PAARTHUKONNU....!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>657</th>\n",
       "      <td>test_img_657.jpg</td>\n",
       "      <td>WATCHING OTHER LANGUAGE MOVIE IN THEATRE SOOOPER APPU!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>658</th>\n",
       "      <td>test_img_658.jpg</td>\n",
       "      <td>ENAKU PIDITHA YENNALME THOLAI THURATHULA ANDRU NILLA.. INDRU NEE..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>659</th>\n",
       "      <td>test_img_659.jpg</td>\n",
       "      <td>EVERUONE IS A GANGSTER  UNTIL THE REAL GANSTER ARRIVES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>660</th>\n",
       "      <td>test_img_660.jpg</td>\n",
       "      <td>*TN GIRLS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>661</th>\n",
       "      <td>test_img_661.jpg</td>\n",
       "      <td>CLIMAX SCENE LA MAGALA KOOTITU VANDHU ALUDHA ELLARUM SUNA PANA AAGIDA MUDIYUMA? DA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>662</th>\n",
       "      <td>test_img_662.jpg</td>\n",
       "      <td>DURING SCHOOL DAYS  ME &amp; FRNDS: ANNA MAIL CHECK PANNANUM LAST SYSTEM KODUNGA  BROWSING CENTER ANNA:</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>663</th>\n",
       "      <td>test_img_663.jpg</td>\n",
       "      <td>REMEMBER THIS KID    THIS KID IS RIGHT NOW</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>664</th>\n",
       "      <td>test_img_664.jpg</td>\n",
       "      <td>*SINGLES: IVAN VERA ENGALA ROMBA TORCHER PANDRANDA...!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665</th>\n",
       "      <td>test_img_665.jpg</td>\n",
       "      <td>ENAIKI PUBG LA CHICKEN DINNER VANGURA PAYALUGALAM YARUUU  ORU KALATHULA DUCK KA SHOOT PANNA MUDIYAMA ENDHA NAI KITAA HE HE HE VANGUNA PAYALUGA THANN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>666</th>\n",
       "      <td>test_img_666.jpg</td>\n",
       "      <td>LATER THAT DAY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>667 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            imagename  \\\n",
       "0      test_img_0.jpg   \n",
       "1      test_img_1.jpg   \n",
       "2      test_img_2.jpg   \n",
       "3      test_img_3.jpg   \n",
       "4      test_img_4.jpg   \n",
       "5      test_img_5.jpg   \n",
       "6      test_img_6.jpg   \n",
       "7      test_img_7.jpg   \n",
       "8      test_img_8.jpg   \n",
       "9      test_img_9.jpg   \n",
       "10    test_img_10.jpg   \n",
       "11    test_img_11.jpg   \n",
       "12    test_img_12.jpg   \n",
       "13    test_img_13.jpg   \n",
       "14    test_img_14.jpg   \n",
       "15    test_img_15.jpg   \n",
       "16    test_img_16.jpg   \n",
       "17    test_img_17.jpg   \n",
       "18    test_img_18.jpg   \n",
       "19    test_img_19.jpg   \n",
       "20    test_img_20.jpg   \n",
       "21    test_img_21.jpg   \n",
       "22    test_img_22.jpg   \n",
       "23    test_img_23.jpg   \n",
       "24    test_img_24.jpg   \n",
       "25    test_img_25.jpg   \n",
       "26    test_img_26.jpg   \n",
       "27    test_img_27.jpg   \n",
       "28    test_img_28.jpg   \n",
       "29    test_img_29.jpg   \n",
       "..                ...   \n",
       "637  test_img_637.jpg   \n",
       "638  test_img_638.jpg   \n",
       "639  test_img_639.jpg   \n",
       "640  test_img_640.jpg   \n",
       "641  test_img_641.jpg   \n",
       "642  test_img_642.jpg   \n",
       "643  test_img_643.jpg   \n",
       "644  test_img_644.jpg   \n",
       "645  test_img_645.jpg   \n",
       "646  test_img_646.jpg   \n",
       "647  test_img_647.jpg   \n",
       "648  test_img_648.jpg   \n",
       "649  test_img_649.jpg   \n",
       "650  test_img_650.jpg   \n",
       "651  test_img_651.jpg   \n",
       "652  test_img_652.jpg   \n",
       "653  test_img_653.jpg   \n",
       "654  test_img_654.jpg   \n",
       "655  test_img_655.jpg   \n",
       "656  test_img_656.jpg   \n",
       "657  test_img_657.jpg   \n",
       "658  test_img_658.jpg   \n",
       "659  test_img_659.jpg   \n",
       "660  test_img_660.jpg   \n",
       "661  test_img_661.jpg   \n",
       "662  test_img_662.jpg   \n",
       "663  test_img_663.jpg   \n",
       "664  test_img_664.jpg   \n",
       "665  test_img_665.jpg   \n",
       "666  test_img_666.jpg   \n",
       "\n",
       "                                                                                                                                                                                                                                         captions  \n",
       "0                                                                                                                                                                                   sugarkaga nadandhava vida figuregaga nadandhavan dhan adhigam  \n",
       "1                                                                                                                                    i have come for my stones      stones thaane.. nalla cheap aana price la namma kalyan jewellers la kedaikkum  \n",
       "2                                                                                                                          \"special porotta\" nu pottuierukke spacial kum sadhavukkum yenna vidhiyasam..?     rendukkum oru naal dhan vidhiyasam..  \n",
       "3                                                                                                                                                                                        *we : amma .. cooker 3 whistle vanthuchu off panniten...  \n",
       "4                                               creating whatsapp group - 1st day vaanga ji.. vaanga ji.. ellarayum add pannunga ji TEAM WORK PANROM...ENJOY PANROMM..   AFTER ONE WEEK ... ENNAYA..? IVLO AMAITHIYA IRUKEENGA? ETHACHU PESUNGAYA  \n",
       "5                                                                                                                                                                       *FOODIE : YAARUM PAARKKAMAL ENNAI PAARKIREN ENNAI ARIYAMAL UNNAI PARKIREN  \n",
       "6                                                                                                                                                                                                             MARGALI MAADHA EFFECT YENNA KULLURU  \n",
       "7                                                                                                                                                                                                      YENNAMMA FEEL PANNI YELUTHI ERUKRANDA AVAN  \n",
       "8                                                                                                                                                                                          TAMIL SERIALS BE LIKE... INI AVARUKKU BADHIL..... IVAR  \n",
       "9                                                                                                                                                                                                                  Mama ponnu Expectation reality  \n",
       "10                                                                                                      pondatti yennai thittum bodhu... eppavum unga amma pechai dhan kepingala.,,? en maganai thittum bodhu... amma pechai ketkave mattiiya...?  \n",
       "11                                                                                                                               mattai adakuvatharku en mutha ponna kattikoduppen... yoo maams kannukuttiya adakitten ilaya ponnu ennakkudhan...  \n",
       "12                                                                                                                                                                                                                               ada paavi payale  \n",
       "13   *(me)ENGINEER : ennoda maasa  sambalam 30aayiram... unga ponna nalla pathukka idhu podhadha?   *mamanar : en ponnukku naan kodukkura pocket maneyye 25aayiram oru masathukku       sorry mama...adhaiyum sethudha 30aayiram endru kurinen...  \n",
       "14                                                                                                                             #10years challenge..... *2008 : NET RECHARGE PANNA 30RS KUDU AATHA... *2018 :  NET RECHARGE PANNA 300RS KUDU AATHA  \n",
       "15                                                                                                                                                              IDHUYENA PRAMADHAM ELECTION VARTTUM IDHAVIDA SPECIAL AACHIYANGALLAN KATHUERUKKU..  \n",
       "16                                                                                                                      *PEOPLE: MATRAM MUNNETRAM ENRIRE ADHARKU YENNA ARTHAM  *JINGLE BELL : PETTIGALAI KAI MATRI NAAN MUNNERUVEN  YENDRU ARTHAM  \n",
       "17                                                                                                                                                                                                               *TN PEOPLE : ADA ARAIBODHA NAAYE  \n",
       "18                                                                                                                                                   *SIMBU : NAMMA MASSU ENNA NU KATANUM! ENAKU ANDAL PAAL UTHUNGA!     YENGA VEETULA ANDA ILLA!  \n",
       "19                                                                                                                                                                  *MOM : VEETULA IRUNDHA ANDAA VAYUM PAAL PACKET UM KANOM      *BANNER JEE FANS  \n",
       "20                                                                                                                                                                              ONLY FOR 90S KIDS SCHOOL DAYS   *PANAKARA VEETU PULLAIYA ERUPPARO  \n",
       "21                                                                                                                                    BEAUTY CARE  #OTHER GIRLS    FOFR MY BESTIE     #ME : NEE DHINAM KULICHAA POTHUMAE, VERA YATHUVUM VENDAAMAE  \n",
       "22                                                                                       VATHIYAR : CURRENT POGUM BODHU MUTHALA YEDHAI YEDUPINGA? TORCH LIGHT, EMERGENCY LIGHT?  RENDUMILLAI SIR, APPA PACKETLA ERUKKIRA PANATHAI YEDUPPEN SIR...  \n",
       "23                                                                                                                                                                                                                          DEI NAMMA ANDREA DA..  \n",
       "24                                                                       #2K KIDS : INNAKKI VENNUMNA NEENGA ORE VEDIYILA 7 SHOT VEDIKKALAM......  #90 KIDS : AANA NANGA ANNAIKE 7 BIJILI VEDIYA ONNA THIRICHU ORE VEDIYILA 7SHOT VEDICHAVUKAPU!!!  \n",
       "25                                                                                                                          100CR... 125CR... NEENGA EDHACHU FEEL PANRINGALA??  *SIMBU : CHA CHA NAAN RAJA AH VANTHA ASSAULT AH 150 CR EDUPANGA..  \n",
       "26                                                                                                                                                   DECENTA KENJUFYING MY SISTER .. PROGRESS REPORT LA APPA KITTA SIGN VAANGI KUDUKARIYAA PLEASE  \n",
       "27                                                                                                                                                                                                       THONDARGAL    *CAR OOTUNAR   *VEETUKARAR  \n",
       "28                                                                                                    WHEN 90S KIDS WATCHING THIS SERIAL...  ~90'S KIDS  : NAMAKU VEELAIKKAARI AMAIYULA... VEELAIKKARI VAIKURA ALAVUKKU PANACASITHYUM AMAIYULS...  \n",
       "29                                            COLLEGE AMAIVATHU ELLAM +2 KODUTHA VARAM... OFFICE AMAIVATHU ELLAM AVAN AVAN SEITHA VINAI... NAMAKU COLLEGE'UM OZHUNGA AMAIYALA...OFFICE'UM OZHUNGA AMAIYALA INTHA PAGE'AAVATHU MAINTAIN PANNUVOM..  \n",
       "..                                                                                                                                                                                                                                            ...  \n",
       "637                                                                                                                                                                                                                  IF YOU ARE BAD  I'M YOUR DAD  \n",
       "638                                                                                      MEERA: NAMMA HUMAN BEING NAMALUKU IRUKKURA MAARI POWER ETHUMAE KEDAIYAATHU ATHA SIXTH SENSE.. ORUTHAR NAMBA PESUMBOTHAE ATHA ENNA 'NU FEEL PANNA MUDIUM.  \n",
       "639                                                                                                                                                      GYM TRAINER TO NEW YEAR MOTIVATIONAL GUYS  UNNA KADAISIYA PONA JANUARY 2 DHETHY PATHATHU  \n",
       "640                                                                                                                                                                                          NAA POREN JAANNU   DEII RAM.. VENAM DA SONNA KELU DA  \n",
       "641                                                                                                                                                                                                  FOUR TYPES OF BIRTHDAY TREATS  ON EVERY GANG  \n",
       "642                                                                                                                                                              90S KIDS DURING INTERVALS  *ME: HEIGHT SOLLADHA .. AVAN KITTA UNDERTAKER IRUKKU!  \n",
       "643                                                                                                                                                                  THANI ORUVAN UNGA EDAM UNGA AALUNGA.. ENAKU BAYAM ILLA, ATHUKU ITHAAN SAMPLE  \n",
       "644                                                                                                                                                                                            LETS TAKE A SELFIE PULA   *SELFIE KUMAR: SHROOOVVV  \n",
       "645                                                                                                                                               ~IN KAATRU VELIYIDAI MOVIE   IDE MAARI COLOUR FULL ANA SONG AH ENGAYO NAMMA PATHURUKAMEY....!!?  \n",
       "646                                                                                                                                                    5) TAMIL MATHANGAL KURIPU ELUDHUGA:  1) CHITTRAI  2}VAIKASI 3) AANI 4) AAVADI 5) PURATTASI  \n",
       "647                                                                                                                                                                                                                                   NO COMMENTS  \n",
       "648                                                                                                                         KAILA VERA PATHUPAISA ILLA IVAN VERA NAMBALA LOVE PANDRENNU SOLDRAN IVAN KUDA SENDHU APPADIYE VANDIYA OOTRIVENDI DHAN  \n",
       "649                                                                                                                                        ENERGY LOW AAGUM VARAI VELAIYADIYADHU ANDHA KALAM  BATTERY LOW AAGUM VARAI VELIYADHUVADHU INDHA KALAM.  \n",
       "650                                                                                                                                                                                                                             ayyey vaaya paaru  \n",
       "651                                                                                                                                                                                                                                   no comments  \n",
       "652                                                                                                                                                                                   MEETS YOU BEFORE EXAM ASKS YOU QUESTIONS YOU'VE NEVER HEARD  \n",
       "653                                                              CURRENT THOTTAA SAADHARANA MANUSHANUKU THAAN SHOCK ADIKUM NAAN NARASIMMA.. ENNA THOTTAA ANDHA CURRENTUKE SHOCK ADIKKUM    CAPTAIN Y U NO PRODUCE CURRENT FOR TAMILNADU POWERCUT?  \n",
       "654                                                                             DAI NANACHUM PADAM MATTUM THA RELEASE PANREN... YENNA NAMBI NEENGA PADAM PAKALAMDA...  AANA IVANAILAM ORU AALA MATHICHU IVANUGALA NAMBI PADATHUKU POVATHINGADA...  \n",
       "655                                                                                                                                                                     ONGI ADICHA ONDRA TON WEIGHT AH??  AIYO, ATHU SATHIYAMA NAAN ILLAINGA NA.  \n",
       "656                                                                                                                            YENDA UNGA APPA PEYARAI FRIDGEKULLA YELUDHI VACHURUKA.? AVARDHANDA SONNARU EN PEYAR KETTUPOGAMMA PAARTHUKONNU....!  \n",
       "657                                                                                                                                                                                        WATCHING OTHER LANGUAGE MOVIE IN THEATRE SOOOPER APPU!  \n",
       "658                                                                                                                                                                            ENAKU PIDITHA YENNALME THOLAI THURATHULA ANDRU NILLA.. INDRU NEE..  \n",
       "659                                                                                                                                                                                        EVERUONE IS A GANGSTER  UNTIL THE REAL GANSTER ARRIVES  \n",
       "660                                                                                                                                                                                                                                     *TN GIRLS  \n",
       "661                                                                                                                                                            CLIMAX SCENE LA MAGALA KOOTITU VANDHU ALUDHA ELLARUM SUNA PANA AAGIDA MUDIYUMA? DA  \n",
       "662                                                                                                                                           DURING SCHOOL DAYS  ME & FRNDS: ANNA MAIL CHECK PANNANUM LAST SYSTEM KODUNGA  BROWSING CENTER ANNA:  \n",
       "663                                                                                                                                                                                                    REMEMBER THIS KID    THIS KID IS RIGHT NOW  \n",
       "664                                                                                                                                                                                        *SINGLES: IVAN VERA ENGALA ROMBA TORCHER PANDRANDA...!  \n",
       "665                                                                                          ENAIKI PUBG LA CHICKEN DINNER VANGURA PAYALUGALAM YARUUU  ORU KALATHULA DUCK KA SHOOT PANNA MUDIYAMA ENDHA NAI KITAA HE HE HE VANGUNA PAYALUGA THANN  \n",
       "666                                                                                                                                                                                                                                LATER THAT DAY  \n",
       "\n",
       "[667 rows x 2 columns]"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = text_classifier_learner(text_dls,AWD_LSTM, drop_mult=0.5, metrics=accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.680365</td>\n",
       "      <td>0.290143</td>\n",
       "      <td>0.906522</td>\n",
       "      <td>00:13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.601811</td>\n",
       "      <td>0.403447</td>\n",
       "      <td>0.850000</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.609261</td>\n",
       "      <td>0.388994</td>\n",
       "      <td>0.906522</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.593565</td>\n",
       "      <td>0.386354</td>\n",
       "      <td>0.893478</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.571990</td>\n",
       "      <td>0.292213</td>\n",
       "      <td>0.928261</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fine_tune(4, 1e-2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.584801</td>\n",
       "      <td>0.230426</td>\n",
       "      <td>0.923913</td>\n",
       "      <td>00:13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.585832</td>\n",
       "      <td>0.663096</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.613570</td>\n",
       "      <td>0.307460</td>\n",
       "      <td>0.915217</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.607570</td>\n",
       "      <td>0.334783</td>\n",
       "      <td>0.902174</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.506133</td>\n",
       "      <td>0.289527</td>\n",
       "      <td>0.923913</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fine_tune(4, 1e-2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "      <th>category</th>\n",
       "      <th>category_</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>xxbos xxup adra xxup adra … xxup xxunk : xxup xxunk xxup xxunk xxup xxunk xxup xxunk xxup alagana xxup xxunk xxup xxunk xxup xxunk . ? xxup namakku xxup yen xxup appadi xxup xxunk xxup xxunk . xxup xxunk \" xxunk \" xxup xxunk xxup xxunk . xxup xxunk : xxup adhuva xxup chellam … xxup naam xxup xxunk xxup piragu xxup xxunk xxup xxunk .. xxup xxunk , xxup chicken xxup xxunk , xxup chicken xxup xxunk , xxup chicken xxup kuruma , xxup chicken xxup kari , xxup xxunk xxup chicken , xxup pepper xxup chicken , xxup xxunk xxup chicken , xxup xxunk xxup chicken , xxup xxunk . xxup ippadi xxup pala xxu...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>xxbos xxup after xxup xxunk xxup with xxup girlfriend … * boy : xxup xxunk ? * girl : xxup hum xxup xxunk xxup xxunk * boy : xxup enna xxup pandra ? * girl : xxup hum xxup xxunk xxup irukan * boy : xxup how xxup was xxup the xxup day ? * girl : xxup hum xxup hum xxup nice xxup nice * boy : xxup ipo xxup olunga xxup xxunk xxup xxunk ? * girls : xxup xxunk xxup sanda xxup xxunk xxup la xxup ipo xxup mattum xxup naan xxup nalla xxup xxunk xxup ah ?</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>xxbos xxunk 's xxup la xxup oru xxup xxunk xxup xxunk xxunk xxup xxunk xxup dhan xxup irukum … xxunk / xxunk … xxup apo xxup oru xxup xxunk xxup ku xxup xxunk xxup pota xxup unaku xxunk xxup xxunk xxup dhana ? ? .. xxup apdi xxup oru xxup coin xxup hey xxup illaye .. xxup enda xxup xxunk xxup xxunk xxup xxunk ? xxup dai xxup xxunk xxrep 3 !</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>xxbos xxup physics xxup oru xxup tution … xxunk xxup ku xxup oru xxup tution … … xxup xxunk 's xxup ku xxup oru xxup tution … .. ~ xxup engineering xxup xxunk ) : xxup xxunk xxup tution xxup xxunk xxup padicha xxup xxunk xxup vela xxup xxunk xxup xxunk xxup xxunk xxup xxunk xxup college xxup pona xxup yennakku xxup xxunk xxup vela xxup xxunk xxup xxunk xxup poda</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.show_results()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SequentialRNN(\n",
       "  (0): SentenceEncoder(\n",
       "    (module): AWD_LSTM(\n",
       "      (encoder): Embedding(1432, 400, padding_idx=1)\n",
       "      (encoder_dp): EmbeddingDropout(\n",
       "        (emb): Embedding(1432, 400, padding_idx=1)\n",
       "      )\n",
       "      (rnns): ModuleList(\n",
       "        (0): WeightDropout(\n",
       "          (module): LSTM(400, 1152, batch_first=True)\n",
       "        )\n",
       "        (1): WeightDropout(\n",
       "          (module): LSTM(1152, 1152, batch_first=True)\n",
       "        )\n",
       "        (2): WeightDropout(\n",
       "          (module): LSTM(1152, 400, batch_first=True)\n",
       "        )\n",
       "      )\n",
       "      (input_dp): RNNDropout()\n",
       "      (hidden_dps): ModuleList(\n",
       "        (0): RNNDropout()\n",
       "        (1): RNNDropout()\n",
       "        (2): RNNDropout()\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (1): PoolingLinearClassifier(\n",
       "    (layers): Sequential(\n",
       "      (0): LinBnDrop(\n",
       "        (0): BatchNorm1d(1200, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (1): Dropout(p=0.2, inplace=False)\n",
       "        (2): Linear(in_features=1200, out_features=50, bias=False)\n",
       "        (3): ReLU(inplace=True)\n",
       "      )\n",
       "      (1): LinBnDrop(\n",
       "        (0): BatchNorm1d(50, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (1): Dropout(p=0.1, inplace=False)\n",
       "        (2): Linear(in_features=50, out_features=2, bias=False)\n",
       "      )\n",
       "    )\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text_model = learn.model\n",
    "text_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "batch no. :  0\n",
      "batch no. :  1\n",
      "batch no. :  2\n",
      "batch no. :  3\n",
      "batch no. :  4\n",
      "batch no. :  5\n",
      "batch no. :  6\n",
      "batch no. :  7\n",
      "batch no. :  8\n",
      "batch no. :  9\n",
      "batch no. :  10\n",
      "batch no. :  11\n",
      "batch no. :  12\n",
      "batch no. :  13\n",
      "batch no. :  14\n",
      "batch no. :  15\n",
      "batch no. :  16\n",
      "batch no. :  17\n",
      "batch no. :  18\n",
      "batch no. :  19\n",
      "batch no. :  20\n",
      "batch no. :  21\n",
      "batch no. :  22\n",
      "batch no. :  23\n",
      "batch no. :  24\n",
      "batch no. :  25\n",
      "batch no. :  26\n",
      "batch no. :  27\n",
      "batch no. :  28\n",
      "batch no. :  29\n",
      "batch no. :  30\n",
      "batch no. :  31\n",
      "batch no. :  32\n",
      "batch no. :  33\n",
      "batch no. :  34\n",
      "batch no. :  35\n",
      "batch no. :  36\n",
      "batch no. :  37\n",
      "batch no. :  38\n",
      "batch no. :  39\n",
      "batch no. :  40\n",
      "batch no. :  41\n",
      "batch no. :  42\n",
      "batch no. :  43\n",
      "batch no. :  44\n",
      "batch no. :  45\n",
      "batch no. :  46\n",
      "batch no. :  47\n",
      "batch no. :  48\n",
      "batch no. :  49\n",
      "batch no. :  50\n",
      "batch no. :  51\n",
      "batch no. :  52\n",
      "batch no. :  53\n",
      "batch no. :  54\n",
      "batch no. :  55\n",
      "batch no. :  56\n",
      "batch no. :  57\n",
      "batch no. :  58\n",
      "batch no. :  59\n",
      "batch no. :  60\n",
      "batch no. :  61\n",
      "batch no. :  62\n",
      "batch no. :  63\n",
      "batch no. :  64\n",
      "batch no. :  65\n",
      "batch no. :  66\n",
      "batch no. :  67\n",
      "batch no. :  68\n",
      "batch no. :  69\n",
      "batch no. :  70\n",
      "batch no. :  71\n",
      "batch no. :  72\n",
      "batch no. :  73\n",
      "batch no. :  74\n",
      "batch no. :  75\n",
      "batch no. :  76\n",
      "batch no. :  77\n",
      "batch no. :  78\n",
      "batch no. :  79\n",
      "batch no. :  80\n",
      "batch no. :  81\n",
      "batch no. :  82\n",
      "batch no. :  83\n",
      "batch no. :  84\n",
      "batch no. :  85\n",
      "batch no. :  86\n",
      "batch no. :  87\n",
      "batch no. :  88\n",
      "batch no. :  89\n",
      "batch no. :  90\n",
      "batch no. :  91\n",
      "batch no. :  92\n",
      "batch no. :  93\n",
      "batch no. :  94\n",
      "batch no. :  95\n",
      "batch no. :  96\n",
      "batch no. :  97\n",
      "batch no. :  98\n",
      "batch no. :  99\n",
      "batch no. :  100\n",
      "batch no. :  101\n",
      "batch no. :  102\n",
      "batch no. :  103\n",
      "batch no. :  104\n",
      "batch no. :  105\n",
      "batch no. :  106\n",
      "batch no. :  107\n",
      "batch no. :  108\n",
      "batch no. :  109\n",
      "batch no. :  110\n",
      "batch no. :  111\n",
      "batch no. :  112\n",
      "batch no. :  113\n",
      "batch no. :  114\n",
      "batch no. :  115\n",
      "batch no. :  116\n",
      "batch no. :  117\n",
      "batch no. :  118\n",
      "batch no. :  119\n",
      "batch no. :  120\n",
      "batch no. :  121\n",
      "batch no. :  122\n",
      "batch no. :  123\n",
      "batch no. :  124\n",
      "batch no. :  125\n",
      "batch no. :  126\n",
      "batch no. :  127\n",
      "batch no. :  128\n",
      "batch no. :  129\n",
      "batch no. :  130\n",
      "batch no. :  131\n",
      "batch no. :  132\n",
      "batch no. :  133\n",
      "batch no. :  134\n",
      "batch no. :  135\n",
      "batch no. :  136\n",
      "batch no. :  137\n",
      "batch no. :  138\n",
      "batch no. :  139\n",
      "batch no. :  140\n",
      "batch no. :  141\n",
      "batch no. :  142\n",
      "batch no. :  143\n",
      "batch no. :  144\n",
      "batch no. :  145\n",
      "batch no. :  146\n",
      "batch no. :  147\n",
      "batch no. :  148\n",
      "batch no. :  149\n",
      "batch no. :  150\n",
      "batch no. :  151\n",
      "batch no. :  152\n",
      "batch no. :  153\n",
      "batch no. :  154\n",
      "batch no. :  155\n",
      "batch no. :  156\n",
      "batch no. :  157\n",
      "batch no. :  158\n",
      "batch no. :  159\n",
      "batch no. :  160\n",
      "batch no. :  161\n",
      "batch no. :  162\n",
      "batch no. :  163\n",
      "batch no. :  164\n",
      "batch no. :  165\n",
      "batch no. :  166\n",
      "batch no. :  167\n",
      "batch no. :  168\n",
      "batch no. :  169\n",
      "batch no. :  170\n",
      "batch no. :  171\n",
      "batch no. :  172\n",
      "batch no. :  173\n",
      "batch no. :  174\n",
      "batch no. :  175\n",
      "batch no. :  176\n",
      "batch no. :  177\n",
      "batch no. :  178\n",
      "batch no. :  179\n",
      "batch no. :  180\n",
      "batch no. :  181\n",
      "batch no. :  182\n",
      "batch no. :  183\n",
      "batch no. :  184\n",
      "batch no. :  185\n",
      "batch no. :  186\n",
      "batch no. :  187\n",
      "batch no. :  188\n",
      "batch no. :  189\n",
      "batch no. :  190\n",
      "batch no. :  191\n",
      "batch no. :  192\n",
      "batch no. :  193\n",
      "batch no. :  194\n",
      "batch no. :  195\n",
      "batch no. :  196\n",
      "batch no. :  197\n",
      "batch no. :  198\n",
      "batch no. :  199\n",
      "batch no. :  200\n",
      "batch no. :  201\n",
      "batch no. :  202\n",
      "batch no. :  203\n",
      "batch no. :  204\n",
      "batch no. :  205\n",
      "batch no. :  206\n",
      "batch no. :  207\n",
      "batch no. :  208\n",
      "batch no. :  209\n",
      "batch no. :  210\n",
      "batch no. :  211\n",
      "batch no. :  212\n",
      "batch no. :  213\n",
      "batch no. :  214\n",
      "batch no. :  215\n",
      "batch no. :  216\n",
      "batch no. :  217\n",
      "batch no. :  218\n",
      "batch no. :  219\n",
      "batch no. :  220\n",
      "batch no. :  221\n",
      "batch no. :  222\n",
      "batch no. :  223\n",
      "batch no. :  224\n",
      "batch no. :  225\n",
      "batch no. :  226\n",
      "batch no. :  227\n",
      "batch no. :  228\n",
      "batch no. :  229\n",
      "batch no. :  230\n",
      "batch no. :  231\n",
      "batch no. :  232\n",
      "batch no. :  233\n",
      "batch no. :  234\n",
      "batch no. :  235\n",
      "batch no. :  236\n",
      "batch no. :  237\n",
      "batch no. :  238\n",
      "batch no. :  239\n",
      "batch no. :  240\n",
      "batch no. :  241\n",
      "batch no. :  242\n",
      "batch no. :  243\n",
      "batch no. :  244\n",
      "batch no. :  245\n",
      "batch no. :  246\n",
      "batch no. :  247\n",
      "batch no. :  248\n",
      "batch no. :  249\n",
      "batch no. :  250\n",
      "batch no. :  251\n",
      "batch no. :  252\n",
      "batch no. :  253\n",
      "batch no. :  254\n",
      "batch no. :  255\n",
      "batch no. :  256\n",
      "batch no. :  257\n",
      "batch no. :  258\n",
      "batch no. :  259\n",
      "batch no. :  260\n",
      "batch no. :  261\n",
      "batch no. :  262\n",
      "batch no. :  263\n",
      "batch no. :  264\n",
      "batch no. :  265\n",
      "batch no. :  266\n",
      "batch no. :  267\n",
      "batch no. :  268\n",
      "batch no. :  269\n",
      "batch no. :  270\n",
      "batch no. :  271\n",
      "batch no. :  272\n",
      "batch no. :  273\n",
      "batch no. :  274\n",
      "batch no. :  275\n",
      "batch no. :  276\n",
      "batch no. :  277\n",
      "batch no. :  278\n",
      "batch no. :  279\n",
      "batch no. :  280\n",
      "batch no. :  281\n",
      "batch no. :  282\n",
      "batch no. :  283\n",
      "batch no. :  284\n",
      "batch no. :  285\n",
      "batch no. :  286\n",
      "batch no. :  287\n",
      "batch no. :  288\n",
      "batch no. :  289\n",
      "batch no. :  290\n",
      "batch no. :  291\n",
      "batch no. :  292\n",
      "batch no. :  293\n",
      "batch no. :  294\n",
      "batch no. :  295\n",
      "batch no. :  296\n",
      "batch no. :  297\n",
      "batch no. :  298\n",
      "batch no. :  299\n",
      "batch no. :  300\n",
      "batch no. :  301\n",
      "batch no. :  302\n",
      "batch no. :  303\n",
      "batch no. :  304\n",
      "batch no. :  305\n",
      "batch no. :  306\n",
      "batch no. :  307\n",
      "batch no. :  308\n",
      "batch no. :  309\n",
      "batch no. :  310\n",
      "batch no. :  311\n",
      "batch no. :  312\n",
      "batch no. :  313\n",
      "batch no. :  314\n",
      "batch no. :  315\n",
      "batch no. :  316\n",
      "batch no. :  317\n",
      "batch no. :  318\n",
      "batch no. :  319\n",
      "batch no. :  320\n",
      "batch no. :  321\n",
      "batch no. :  322\n",
      "batch no. :  323\n",
      "batch no. :  324\n",
      "batch no. :  325\n",
      "batch no. :  326\n",
      "batch no. :  327\n",
      "batch no. :  328\n",
      "batch no. :  329\n",
      "batch no. :  330\n",
      "batch no. :  331\n",
      "batch no. :  332\n",
      "batch no. :  333\n",
      "batch no. :  334\n",
      "batch no. :  335\n",
      "batch no. :  336\n",
      "batch no. :  337\n",
      "batch no. :  338\n",
      "batch no. :  339\n",
      "batch no. :  340\n",
      "batch no. :  341\n",
      "batch no. :  342\n",
      "batch no. :  343\n",
      "batch no. :  344\n",
      "batch no. :  345\n",
      "batch no. :  346\n",
      "batch no. :  347\n",
      "batch no. :  348\n",
      "batch no. :  349\n",
      "batch no. :  350\n",
      "batch no. :  351\n",
      "batch no. :  352\n",
      "batch no. :  353\n",
      "batch no. :  354\n",
      "batch no. :  355\n",
      "batch no. :  356\n",
      "batch no. :  357\n",
      "batch no. :  358\n",
      "batch no. :  359\n",
      "batch no. :  360\n",
      "batch no. :  361\n",
      "batch no. :  362\n",
      "batch no. :  363\n",
      "batch no. :  364\n",
      "batch no. :  365\n",
      "batch no. :  366\n",
      "batch no. :  367\n",
      "batch no. :  368\n",
      "batch no. :  369\n",
      "batch no. :  370\n",
      "batch no. :  371\n",
      "batch no. :  372\n",
      "batch no. :  373\n",
      "batch no. :  374\n",
      "batch no. :  375\n",
      "batch no. :  376\n",
      "batch no. :  377\n",
      "batch no. :  378\n",
      "batch no. :  379\n",
      "batch no. :  380\n",
      "batch no. :  381\n",
      "batch no. :  382\n",
      "batch no. :  383\n",
      "batch no. :  384\n",
      "batch no. :  385\n",
      "batch no. :  386\n",
      "batch no. :  387\n",
      "batch no. :  388\n",
      "batch no. :  389\n",
      "batch no. :  390\n",
      "batch no. :  391\n",
      "batch no. :  392\n",
      "batch no. :  393\n",
      "batch no. :  394\n",
      "batch no. :  395\n",
      "batch no. :  396\n",
      "batch no. :  397\n",
      "batch no. :  398\n",
      "batch no. :  399\n",
      "batch no. :  400\n",
      "batch no. :  401\n",
      "batch no. :  402\n",
      "batch no. :  403\n",
      "batch no. :  404\n",
      "batch no. :  405\n",
      "batch no. :  406\n",
      "batch no. :  407\n",
      "batch no. :  408\n",
      "batch no. :  409\n",
      "batch no. :  410\n",
      "batch no. :  411\n",
      "batch no. :  412\n",
      "batch no. :  413\n",
      "batch no. :  414\n",
      "batch no. :  415\n",
      "batch no. :  416\n",
      "batch no. :  417\n",
      "batch no. :  418\n",
      "batch no. :  419\n",
      "batch no. :  420\n",
      "batch no. :  421\n",
      "batch no. :  422\n",
      "batch no. :  423\n",
      "batch no. :  424\n",
      "batch no. :  425\n",
      "batch no. :  426\n",
      "batch no. :  427\n",
      "batch no. :  428\n",
      "batch no. :  429\n",
      "batch no. :  430\n",
      "batch no. :  431\n",
      "batch no. :  432\n",
      "batch no. :  433\n",
      "batch no. :  434\n",
      "batch no. :  435\n",
      "batch no. :  436\n",
      "batch no. :  437\n",
      "batch no. :  438\n",
      "batch no. :  439\n",
      "batch no. :  440\n",
      "batch no. :  441\n",
      "batch no. :  442\n",
      "batch no. :  443\n",
      "batch no. :  444\n",
      "batch no. :  445\n",
      "batch no. :  446\n",
      "batch no. :  447\n",
      "batch no. :  448\n",
      "batch no. :  449\n",
      "batch no. :  450\n",
      "batch no. :  451\n",
      "batch no. :  452\n",
      "batch no. :  453\n",
      "batch no. :  454\n",
      "batch no. :  455\n",
      "batch no. :  456\n",
      "batch no. :  457\n",
      "batch no. :  458\n",
      "batch no. :  459\n"
     ]
    }
   ],
   "source": [
    "train_labels_set_text = []\n",
    "train_preds_set_text = []\n",
    "for idx, data in enumerate(text_dls.train):\n",
    "    # get the inputs\n",
    "    text_model.eval()\n",
    "    print(\"batch no. : \", idx)\n",
    "    inputs, labels = data\n",
    "    train_labels_set_text += labels.tolist()\n",
    "    inputs, labels = inputs.cuda(), labels.cuda()\n",
    "#     print(labels)\n",
    "    outputs = text_model(inputs)[0]\n",
    "    _, preds = torch.max(outputs, 1)\n",
    "    train_preds_set_text += preds.cpu().tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9211956521739131"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_pred=train_preds_set_text, y_true=train_labels_set_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "class MyEnsemble(nn.Module):\n",
    "    def __init__(self, modelA, modelB):\n",
    "        super(MyEnsemble, self).__init__()\n",
    "        self.modelA = modelA\n",
    "        self.modelB = modelB\n",
    "        self.classifier = nn.Linear(4, 2)\n",
    "        \n",
    "    def forward(self, x1, x2):\n",
    "        x1 = self.modelA(x1)\n",
    "        x2 = self.modelB(x2)[0]\n",
    "        x = torch.cat((x1, x2), dim=1)\n",
    "        x = self.classifier(F.relu(x))\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 0,\n",
       " 0,\n",
       " 0,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " 1,\n",
       " ...]"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_preds_set_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/numpy/core/_asarray.py:83: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order)\n"
     ]
    }
   ],
   "source": [
    "data_text = TextDataLoaders.from_df(df, valid_pct=0, text_col='captions', label_col='label', shuffle_train=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "batch no. :  0\n",
      "TensorText([[   2,  369,  232,   12,   34,    0,  817,   10,  915,   57,  567,    0,\n",
      "          915,   57,  168,   13,  567,    0,   16,   43,    0,   14,  369,   90,\n",
      "           12,   43,  258,  436,  466,   67,    0,   15,    7,    0,    7,  915,\n",
      "            7,   57,    7,  122,    7, 1222,  244,    7,   63,    7,  581,    7,\n",
      "           31,    7,  548,   15,    7,   63,    7,  258,  363,    7,  466,    7,\n",
      "          567,    7,    0,   13,    7,    0,   16,    7,    0,    7,  581,    7,\n",
      "           63,    7,  548,    7,  137,    7,  102,    7,   71,    7,  537,    7,\n",
      "            0,    7,    0,    7,  567,    7,    0,    7,    0,    7,   63,    7,\n",
      "            0,   11,   11,    7,  222,   47,    7,    0,   10,   18,    0,    7,\n",
      "            0,   11],\n",
      "        [   2,    7,    0,    7,   38,    7,    0,    7,    0,    7,    0,   12,\n",
      "            7,  118,   90,    7,  486,    7,   82,    7, 1004,    9,    7,  229,\n",
      "            7,  996,    7,   97,    7, 1397,    9,  454,    7,   82,    7, 1004,\n",
      "            9,  454,    7,  996,    7,   97,    7, 1397,    9,  454,    7,   82,\n",
      "            7, 1004,    9,  454,    7,  996,    7,   97,    7, 1397,    9,  454,\n",
      "            7,   82,    7, 1004,    9,  266,  368,    5,   17,   29,    7,    0,\n",
      "            7,  174,    7,    0,    7,  646,    7,   53,    7,   64,    7,    0,\n",
      "            7,  999,    7,    0,    7,  281,    9,    1,    1,    1,    1,    1,\n",
      "            1,    1,    1,    1,    1,    1,    1,    1,    1,    1,    1,    1,\n",
      "            1,    1],\n",
      "        [   2,    7,    0,    7,    0,    7,    0,    7,    0,    7,   81,   29,\n",
      "            7, 1322,    7,    0,  363,    7,    0,    7,    0,    7,  994,   29,\n",
      "            7,    0,  244,   26,    7,  242,    7,  889,  363,    7,    0,    7,\n",
      "          400,    7,    0,    7,   46,    7,  370,    7,    0,    7,    0,   15,\n",
      "            7,   71,    7,    0,    7,  240,    7,  294,    7,    0,   29,   26,\n",
      "            7,    0,    7,  146,    7,  242,    7,   98,    7,   46,    7, 1336,\n",
      "           15,   26,    7,    0,   11,    7,  546,    7,   98,    7,   46,    7,\n",
      "            0,    7, 1180,    5,  197,  163,    1,    1,    1,    1,    1,    1,\n",
      "            1,    1,    1,    1,    1,    1,    1,    1,    1,    1,    1,    1,\n",
      "            1,    1],\n",
      "        [   2,    7,    0,    7, 1231,    7,  326,    7, 1232,    7,    0,    7,\n",
      "          127,    7,    0,    7,   32,    7,    0,   13,    7,  178,    7, 1233,\n",
      "            7, 1231,    7,   58,    7,  326,    7, 1232,    7, 1234,    7,    0,\n",
      "            7,   32,    7,    0,   13,    7,   27,    7,   32,    7,  746,    7,\n",
      "          643,   13,    7,    0,    9,    7,  666,    7,    0,    7,  326,    7,\n",
      "            0,    7,    0,    7,    0,   13,   13,    7,  838,    7,  892,    7,\n",
      "           32,    7,  194,   10,    7,  892,    7,    0,   10,    7,    0,    7,\n",
      "          493,    7,  504,    7,   68,    1,    1,    1,    1,    1,    1,    1,\n",
      "            1,    1,    1,    1,    1,    1,    1,    1,    1,    1,    1,    1,\n",
      "            1,    1]], device='cuda:0')\n",
      "TensorCategory([1, 1, 1, 1], device='cuda:0')\n"
     ]
    }
   ],
   "source": [
    "for idx, data in enumerate(text_dls.train.new(shuffle=False)):\n",
    "    # get the inputs\n",
    "    text_model.eval()\n",
    "    print(\"batch no. : \", idx)\n",
    "    inputs, labels = data\n",
    "    print(inputs)\n",
    "    print(labels)\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>imagename</th>\n",
       "      <th>captions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>test_img_0.jpg</td>\n",
       "      <td>sugarkaga nadandhava vida figuregaga nadandhavan dhan adhigam</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>test_img_1.jpg</td>\n",
       "      <td>i have come for my stones      stones thaane.. nalla cheap aana price la namma kalyan jewellers la kedaikkum</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>test_img_2.jpg</td>\n",
       "      <td>\"special porotta\" nu pottuierukke spacial kum sadhavukkum yenna vidhiyasam..?     rendukkum oru naal dhan vidhiyasam..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>test_img_3.jpg</td>\n",
       "      <td>*we : amma .. cooker 3 whistle vanthuchu off panniten...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>test_img_4.jpg</td>\n",
       "      <td>creating whatsapp group - 1st day vaanga ji.. vaanga ji.. ellarayum add pannunga ji TEAM WORK PANROM...ENJOY PANROMM..   AFTER ONE WEEK ... ENNAYA..? IVLO AMAITHIYA IRUKEENGA? ETHACHU PESUNGAYA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>test_img_5.jpg</td>\n",
       "      <td>*FOODIE : YAARUM PAARKKAMAL ENNAI PAARKIREN ENNAI ARIYAMAL UNNAI PARKIREN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>test_img_6.jpg</td>\n",
       "      <td>MARGALI MAADHA EFFECT YENNA KULLURU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>test_img_7.jpg</td>\n",
       "      <td>YENNAMMA FEEL PANNI YELUTHI ERUKRANDA AVAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>test_img_8.jpg</td>\n",
       "      <td>TAMIL SERIALS BE LIKE... INI AVARUKKU BADHIL..... IVAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>test_img_9.jpg</td>\n",
       "      <td>Mama ponnu Expectation reality</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>test_img_10.jpg</td>\n",
       "      <td>pondatti yennai thittum bodhu... eppavum unga amma pechai dhan kepingala.,,? en maganai thittum bodhu... amma pechai ketkave mattiiya...?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>test_img_11.jpg</td>\n",
       "      <td>mattai adakuvatharku en mutha ponna kattikoduppen... yoo maams kannukuttiya adakitten ilaya ponnu ennakkudhan...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>test_img_12.jpg</td>\n",
       "      <td>ada paavi payale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>test_img_13.jpg</td>\n",
       "      <td>*(me)ENGINEER : ennoda maasa  sambalam 30aayiram... unga ponna nalla pathukka idhu podhadha?   *mamanar : en ponnukku naan kodukkura pocket maneyye 25aayiram oru masathukku       sorry mama...adhaiyum sethudha 30aayiram endru kurinen...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>test_img_14.jpg</td>\n",
       "      <td>#10years challenge..... *2008 : NET RECHARGE PANNA 30RS KUDU AATHA... *2018 :  NET RECHARGE PANNA 300RS KUDU AATHA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>test_img_15.jpg</td>\n",
       "      <td>IDHUYENA PRAMADHAM ELECTION VARTTUM IDHAVIDA SPECIAL AACHIYANGALLAN KATHUERUKKU..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>test_img_16.jpg</td>\n",
       "      <td>*PEOPLE: MATRAM MUNNETRAM ENRIRE ADHARKU YENNA ARTHAM  *JINGLE BELL : PETTIGALAI KAI MATRI NAAN MUNNERUVEN  YENDRU ARTHAM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>test_img_17.jpg</td>\n",
       "      <td>*TN PEOPLE : ADA ARAIBODHA NAAYE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>test_img_18.jpg</td>\n",
       "      <td>*SIMBU : NAMMA MASSU ENNA NU KATANUM! ENAKU ANDAL PAAL UTHUNGA!     YENGA VEETULA ANDA ILLA!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>test_img_19.jpg</td>\n",
       "      <td>*MOM : VEETULA IRUNDHA ANDAA VAYUM PAAL PACKET UM KANOM      *BANNER JEE FANS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>test_img_20.jpg</td>\n",
       "      <td>ONLY FOR 90S KIDS SCHOOL DAYS   *PANAKARA VEETU PULLAIYA ERUPPARO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>test_img_21.jpg</td>\n",
       "      <td>BEAUTY CARE  #OTHER GIRLS    FOFR MY BESTIE     #ME : NEE DHINAM KULICHAA POTHUMAE, VERA YATHUVUM VENDAAMAE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>test_img_22.jpg</td>\n",
       "      <td>VATHIYAR : CURRENT POGUM BODHU MUTHALA YEDHAI YEDUPINGA? TORCH LIGHT, EMERGENCY LIGHT?  RENDUMILLAI SIR, APPA PACKETLA ERUKKIRA PANATHAI YEDUPPEN SIR...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>test_img_23.jpg</td>\n",
       "      <td>DEI NAMMA ANDREA DA..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>test_img_24.jpg</td>\n",
       "      <td>#2K KIDS : INNAKKI VENNUMNA NEENGA ORE VEDIYILA 7 SHOT VEDIKKALAM......  #90 KIDS : AANA NANGA ANNAIKE 7 BIJILI VEDIYA ONNA THIRICHU ORE VEDIYILA 7SHOT VEDICHAVUKAPU!!!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>test_img_25.jpg</td>\n",
       "      <td>100CR... 125CR... NEENGA EDHACHU FEEL PANRINGALA??  *SIMBU : CHA CHA NAAN RAJA AH VANTHA ASSAULT AH 150 CR EDUPANGA..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>test_img_26.jpg</td>\n",
       "      <td>DECENTA KENJUFYING MY SISTER .. PROGRESS REPORT LA APPA KITTA SIGN VAANGI KUDUKARIYAA PLEASE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>test_img_27.jpg</td>\n",
       "      <td>THONDARGAL    *CAR OOTUNAR   *VEETUKARAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>test_img_28.jpg</td>\n",
       "      <td>WHEN 90S KIDS WATCHING THIS SERIAL...  ~90'S KIDS  : NAMAKU VEELAIKKAARI AMAIYULA... VEELAIKKARI VAIKURA ALAVUKKU PANACASITHYUM AMAIYULS...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>test_img_29.jpg</td>\n",
       "      <td>COLLEGE AMAIVATHU ELLAM +2 KODUTHA VARAM... OFFICE AMAIVATHU ELLAM AVAN AVAN SEITHA VINAI... NAMAKU COLLEGE'UM OZHUNGA AMAIYALA...OFFICE'UM OZHUNGA AMAIYALA INTHA PAGE'AAVATHU MAINTAIN PANNUVOM..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>637</th>\n",
       "      <td>test_img_637.jpg</td>\n",
       "      <td>IF YOU ARE BAD  I'M YOUR DAD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>638</th>\n",
       "      <td>test_img_638.jpg</td>\n",
       "      <td>MEERA: NAMMA HUMAN BEING NAMALUKU IRUKKURA MAARI POWER ETHUMAE KEDAIYAATHU ATHA SIXTH SENSE.. ORUTHAR NAMBA PESUMBOTHAE ATHA ENNA 'NU FEEL PANNA MUDIUM.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>639</th>\n",
       "      <td>test_img_639.jpg</td>\n",
       "      <td>GYM TRAINER TO NEW YEAR MOTIVATIONAL GUYS  UNNA KADAISIYA PONA JANUARY 2 DHETHY PATHATHU</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>640</th>\n",
       "      <td>test_img_640.jpg</td>\n",
       "      <td>NAA POREN JAANNU   DEII RAM.. VENAM DA SONNA KELU DA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>641</th>\n",
       "      <td>test_img_641.jpg</td>\n",
       "      <td>FOUR TYPES OF BIRTHDAY TREATS  ON EVERY GANG</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>642</th>\n",
       "      <td>test_img_642.jpg</td>\n",
       "      <td>90S KIDS DURING INTERVALS  *ME: HEIGHT SOLLADHA .. AVAN KITTA UNDERTAKER IRUKKU!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>643</th>\n",
       "      <td>test_img_643.jpg</td>\n",
       "      <td>THANI ORUVAN UNGA EDAM UNGA AALUNGA.. ENAKU BAYAM ILLA, ATHUKU ITHAAN SAMPLE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>644</th>\n",
       "      <td>test_img_644.jpg</td>\n",
       "      <td>LETS TAKE A SELFIE PULA   *SELFIE KUMAR: SHROOOVVV</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>645</th>\n",
       "      <td>test_img_645.jpg</td>\n",
       "      <td>~IN KAATRU VELIYIDAI MOVIE   IDE MAARI COLOUR FULL ANA SONG AH ENGAYO NAMMA PATHURUKAMEY....!!?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>646</th>\n",
       "      <td>test_img_646.jpg</td>\n",
       "      <td>5) TAMIL MATHANGAL KURIPU ELUDHUGA:  1) CHITTRAI  2}VAIKASI 3) AANI 4) AAVADI 5) PURATTASI</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>647</th>\n",
       "      <td>test_img_647.jpg</td>\n",
       "      <td>NO COMMENTS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>648</th>\n",
       "      <td>test_img_648.jpg</td>\n",
       "      <td>KAILA VERA PATHUPAISA ILLA IVAN VERA NAMBALA LOVE PANDRENNU SOLDRAN IVAN KUDA SENDHU APPADIYE VANDIYA OOTRIVENDI DHAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>649</th>\n",
       "      <td>test_img_649.jpg</td>\n",
       "      <td>ENERGY LOW AAGUM VARAI VELAIYADIYADHU ANDHA KALAM  BATTERY LOW AAGUM VARAI VELIYADHUVADHU INDHA KALAM.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>650</th>\n",
       "      <td>test_img_650.jpg</td>\n",
       "      <td>ayyey vaaya paaru</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>651</th>\n",
       "      <td>test_img_651.jpg</td>\n",
       "      <td>no comments</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>652</th>\n",
       "      <td>test_img_652.jpg</td>\n",
       "      <td>MEETS YOU BEFORE EXAM ASKS YOU QUESTIONS YOU'VE NEVER HEARD</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>653</th>\n",
       "      <td>test_img_653.jpg</td>\n",
       "      <td>CURRENT THOTTAA SAADHARANA MANUSHANUKU THAAN SHOCK ADIKUM NAAN NARASIMMA.. ENNA THOTTAA ANDHA CURRENTUKE SHOCK ADIKKUM    CAPTAIN Y U NO PRODUCE CURRENT FOR TAMILNADU POWERCUT?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>654</th>\n",
       "      <td>test_img_654.jpg</td>\n",
       "      <td>DAI NANACHUM PADAM MATTUM THA RELEASE PANREN... YENNA NAMBI NEENGA PADAM PAKALAMDA...  AANA IVANAILAM ORU AALA MATHICHU IVANUGALA NAMBI PADATHUKU POVATHINGADA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>655</th>\n",
       "      <td>test_img_655.jpg</td>\n",
       "      <td>ONGI ADICHA ONDRA TON WEIGHT AH??  AIYO, ATHU SATHIYAMA NAAN ILLAINGA NA.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>656</th>\n",
       "      <td>test_img_656.jpg</td>\n",
       "      <td>YENDA UNGA APPA PEYARAI FRIDGEKULLA YELUDHI VACHURUKA.? AVARDHANDA SONNARU EN PEYAR KETTUPOGAMMA PAARTHUKONNU....!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>657</th>\n",
       "      <td>test_img_657.jpg</td>\n",
       "      <td>WATCHING OTHER LANGUAGE MOVIE IN THEATRE SOOOPER APPU!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>658</th>\n",
       "      <td>test_img_658.jpg</td>\n",
       "      <td>ENAKU PIDITHA YENNALME THOLAI THURATHULA ANDRU NILLA.. INDRU NEE..</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>659</th>\n",
       "      <td>test_img_659.jpg</td>\n",
       "      <td>EVERUONE IS A GANGSTER  UNTIL THE REAL GANSTER ARRIVES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>660</th>\n",
       "      <td>test_img_660.jpg</td>\n",
       "      <td>*TN GIRLS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>661</th>\n",
       "      <td>test_img_661.jpg</td>\n",
       "      <td>CLIMAX SCENE LA MAGALA KOOTITU VANDHU ALUDHA ELLARUM SUNA PANA AAGIDA MUDIYUMA? DA</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>662</th>\n",
       "      <td>test_img_662.jpg</td>\n",
       "      <td>DURING SCHOOL DAYS  ME &amp; FRNDS: ANNA MAIL CHECK PANNANUM LAST SYSTEM KODUNGA  BROWSING CENTER ANNA:</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>663</th>\n",
       "      <td>test_img_663.jpg</td>\n",
       "      <td>REMEMBER THIS KID    THIS KID IS RIGHT NOW</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>664</th>\n",
       "      <td>test_img_664.jpg</td>\n",
       "      <td>*SINGLES: IVAN VERA ENGALA ROMBA TORCHER PANDRANDA...!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665</th>\n",
       "      <td>test_img_665.jpg</td>\n",
       "      <td>ENAIKI PUBG LA CHICKEN DINNER VANGURA PAYALUGALAM YARUUU  ORU KALATHULA DUCK KA SHOOT PANNA MUDIYAMA ENDHA NAI KITAA HE HE HE VANGUNA PAYALUGA THANN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>666</th>\n",
       "      <td>test_img_666.jpg</td>\n",
       "      <td>LATER THAT DAY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>667 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            imagename  \\\n",
       "0      test_img_0.jpg   \n",
       "1      test_img_1.jpg   \n",
       "2      test_img_2.jpg   \n",
       "3      test_img_3.jpg   \n",
       "4      test_img_4.jpg   \n",
       "5      test_img_5.jpg   \n",
       "6      test_img_6.jpg   \n",
       "7      test_img_7.jpg   \n",
       "8      test_img_8.jpg   \n",
       "9      test_img_9.jpg   \n",
       "10    test_img_10.jpg   \n",
       "11    test_img_11.jpg   \n",
       "12    test_img_12.jpg   \n",
       "13    test_img_13.jpg   \n",
       "14    test_img_14.jpg   \n",
       "15    test_img_15.jpg   \n",
       "16    test_img_16.jpg   \n",
       "17    test_img_17.jpg   \n",
       "18    test_img_18.jpg   \n",
       "19    test_img_19.jpg   \n",
       "20    test_img_20.jpg   \n",
       "21    test_img_21.jpg   \n",
       "22    test_img_22.jpg   \n",
       "23    test_img_23.jpg   \n",
       "24    test_img_24.jpg   \n",
       "25    test_img_25.jpg   \n",
       "26    test_img_26.jpg   \n",
       "27    test_img_27.jpg   \n",
       "28    test_img_28.jpg   \n",
       "29    test_img_29.jpg   \n",
       "..                ...   \n",
       "637  test_img_637.jpg   \n",
       "638  test_img_638.jpg   \n",
       "639  test_img_639.jpg   \n",
       "640  test_img_640.jpg   \n",
       "641  test_img_641.jpg   \n",
       "642  test_img_642.jpg   \n",
       "643  test_img_643.jpg   \n",
       "644  test_img_644.jpg   \n",
       "645  test_img_645.jpg   \n",
       "646  test_img_646.jpg   \n",
       "647  test_img_647.jpg   \n",
       "648  test_img_648.jpg   \n",
       "649  test_img_649.jpg   \n",
       "650  test_img_650.jpg   \n",
       "651  test_img_651.jpg   \n",
       "652  test_img_652.jpg   \n",
       "653  test_img_653.jpg   \n",
       "654  test_img_654.jpg   \n",
       "655  test_img_655.jpg   \n",
       "656  test_img_656.jpg   \n",
       "657  test_img_657.jpg   \n",
       "658  test_img_658.jpg   \n",
       "659  test_img_659.jpg   \n",
       "660  test_img_660.jpg   \n",
       "661  test_img_661.jpg   \n",
       "662  test_img_662.jpg   \n",
       "663  test_img_663.jpg   \n",
       "664  test_img_664.jpg   \n",
       "665  test_img_665.jpg   \n",
       "666  test_img_666.jpg   \n",
       "\n",
       "                                                                                                                                                                                                                                         captions  \n",
       "0                                                                                                                                                                                   sugarkaga nadandhava vida figuregaga nadandhavan dhan adhigam  \n",
       "1                                                                                                                                    i have come for my stones      stones thaane.. nalla cheap aana price la namma kalyan jewellers la kedaikkum  \n",
       "2                                                                                                                          \"special porotta\" nu pottuierukke spacial kum sadhavukkum yenna vidhiyasam..?     rendukkum oru naal dhan vidhiyasam..  \n",
       "3                                                                                                                                                                                        *we : amma .. cooker 3 whistle vanthuchu off panniten...  \n",
       "4                                               creating whatsapp group - 1st day vaanga ji.. vaanga ji.. ellarayum add pannunga ji TEAM WORK PANROM...ENJOY PANROMM..   AFTER ONE WEEK ... ENNAYA..? IVLO AMAITHIYA IRUKEENGA? ETHACHU PESUNGAYA  \n",
       "5                                                                                                                                                                       *FOODIE : YAARUM PAARKKAMAL ENNAI PAARKIREN ENNAI ARIYAMAL UNNAI PARKIREN  \n",
       "6                                                                                                                                                                                                             MARGALI MAADHA EFFECT YENNA KULLURU  \n",
       "7                                                                                                                                                                                                      YENNAMMA FEEL PANNI YELUTHI ERUKRANDA AVAN  \n",
       "8                                                                                                                                                                                          TAMIL SERIALS BE LIKE... INI AVARUKKU BADHIL..... IVAR  \n",
       "9                                                                                                                                                                                                                  Mama ponnu Expectation reality  \n",
       "10                                                                                                      pondatti yennai thittum bodhu... eppavum unga amma pechai dhan kepingala.,,? en maganai thittum bodhu... amma pechai ketkave mattiiya...?  \n",
       "11                                                                                                                               mattai adakuvatharku en mutha ponna kattikoduppen... yoo maams kannukuttiya adakitten ilaya ponnu ennakkudhan...  \n",
       "12                                                                                                                                                                                                                               ada paavi payale  \n",
       "13   *(me)ENGINEER : ennoda maasa  sambalam 30aayiram... unga ponna nalla pathukka idhu podhadha?   *mamanar : en ponnukku naan kodukkura pocket maneyye 25aayiram oru masathukku       sorry mama...adhaiyum sethudha 30aayiram endru kurinen...  \n",
       "14                                                                                                                             #10years challenge..... *2008 : NET RECHARGE PANNA 30RS KUDU AATHA... *2018 :  NET RECHARGE PANNA 300RS KUDU AATHA  \n",
       "15                                                                                                                                                              IDHUYENA PRAMADHAM ELECTION VARTTUM IDHAVIDA SPECIAL AACHIYANGALLAN KATHUERUKKU..  \n",
       "16                                                                                                                      *PEOPLE: MATRAM MUNNETRAM ENRIRE ADHARKU YENNA ARTHAM  *JINGLE BELL : PETTIGALAI KAI MATRI NAAN MUNNERUVEN  YENDRU ARTHAM  \n",
       "17                                                                                                                                                                                                               *TN PEOPLE : ADA ARAIBODHA NAAYE  \n",
       "18                                                                                                                                                   *SIMBU : NAMMA MASSU ENNA NU KATANUM! ENAKU ANDAL PAAL UTHUNGA!     YENGA VEETULA ANDA ILLA!  \n",
       "19                                                                                                                                                                  *MOM : VEETULA IRUNDHA ANDAA VAYUM PAAL PACKET UM KANOM      *BANNER JEE FANS  \n",
       "20                                                                                                                                                                              ONLY FOR 90S KIDS SCHOOL DAYS   *PANAKARA VEETU PULLAIYA ERUPPARO  \n",
       "21                                                                                                                                    BEAUTY CARE  #OTHER GIRLS    FOFR MY BESTIE     #ME : NEE DHINAM KULICHAA POTHUMAE, VERA YATHUVUM VENDAAMAE  \n",
       "22                                                                                       VATHIYAR : CURRENT POGUM BODHU MUTHALA YEDHAI YEDUPINGA? TORCH LIGHT, EMERGENCY LIGHT?  RENDUMILLAI SIR, APPA PACKETLA ERUKKIRA PANATHAI YEDUPPEN SIR...  \n",
       "23                                                                                                                                                                                                                          DEI NAMMA ANDREA DA..  \n",
       "24                                                                       #2K KIDS : INNAKKI VENNUMNA NEENGA ORE VEDIYILA 7 SHOT VEDIKKALAM......  #90 KIDS : AANA NANGA ANNAIKE 7 BIJILI VEDIYA ONNA THIRICHU ORE VEDIYILA 7SHOT VEDICHAVUKAPU!!!  \n",
       "25                                                                                                                          100CR... 125CR... NEENGA EDHACHU FEEL PANRINGALA??  *SIMBU : CHA CHA NAAN RAJA AH VANTHA ASSAULT AH 150 CR EDUPANGA..  \n",
       "26                                                                                                                                                   DECENTA KENJUFYING MY SISTER .. PROGRESS REPORT LA APPA KITTA SIGN VAANGI KUDUKARIYAA PLEASE  \n",
       "27                                                                                                                                                                                                       THONDARGAL    *CAR OOTUNAR   *VEETUKARAR  \n",
       "28                                                                                                    WHEN 90S KIDS WATCHING THIS SERIAL...  ~90'S KIDS  : NAMAKU VEELAIKKAARI AMAIYULA... VEELAIKKARI VAIKURA ALAVUKKU PANACASITHYUM AMAIYULS...  \n",
       "29                                            COLLEGE AMAIVATHU ELLAM +2 KODUTHA VARAM... OFFICE AMAIVATHU ELLAM AVAN AVAN SEITHA VINAI... NAMAKU COLLEGE'UM OZHUNGA AMAIYALA...OFFICE'UM OZHUNGA AMAIYALA INTHA PAGE'AAVATHU MAINTAIN PANNUVOM..  \n",
       "..                                                                                                                                                                                                                                            ...  \n",
       "637                                                                                                                                                                                                                  IF YOU ARE BAD  I'M YOUR DAD  \n",
       "638                                                                                      MEERA: NAMMA HUMAN BEING NAMALUKU IRUKKURA MAARI POWER ETHUMAE KEDAIYAATHU ATHA SIXTH SENSE.. ORUTHAR NAMBA PESUMBOTHAE ATHA ENNA 'NU FEEL PANNA MUDIUM.  \n",
       "639                                                                                                                                                      GYM TRAINER TO NEW YEAR MOTIVATIONAL GUYS  UNNA KADAISIYA PONA JANUARY 2 DHETHY PATHATHU  \n",
       "640                                                                                                                                                                                          NAA POREN JAANNU   DEII RAM.. VENAM DA SONNA KELU DA  \n",
       "641                                                                                                                                                                                                  FOUR TYPES OF BIRTHDAY TREATS  ON EVERY GANG  \n",
       "642                                                                                                                                                              90S KIDS DURING INTERVALS  *ME: HEIGHT SOLLADHA .. AVAN KITTA UNDERTAKER IRUKKU!  \n",
       "643                                                                                                                                                                  THANI ORUVAN UNGA EDAM UNGA AALUNGA.. ENAKU BAYAM ILLA, ATHUKU ITHAAN SAMPLE  \n",
       "644                                                                                                                                                                                            LETS TAKE A SELFIE PULA   *SELFIE KUMAR: SHROOOVVV  \n",
       "645                                                                                                                                               ~IN KAATRU VELIYIDAI MOVIE   IDE MAARI COLOUR FULL ANA SONG AH ENGAYO NAMMA PATHURUKAMEY....!!?  \n",
       "646                                                                                                                                                    5) TAMIL MATHANGAL KURIPU ELUDHUGA:  1) CHITTRAI  2}VAIKASI 3) AANI 4) AAVADI 5) PURATTASI  \n",
       "647                                                                                                                                                                                                                                   NO COMMENTS  \n",
       "648                                                                                                                         KAILA VERA PATHUPAISA ILLA IVAN VERA NAMBALA LOVE PANDRENNU SOLDRAN IVAN KUDA SENDHU APPADIYE VANDIYA OOTRIVENDI DHAN  \n",
       "649                                                                                                                                        ENERGY LOW AAGUM VARAI VELAIYADIYADHU ANDHA KALAM  BATTERY LOW AAGUM VARAI VELIYADHUVADHU INDHA KALAM.  \n",
       "650                                                                                                                                                                                                                             ayyey vaaya paaru  \n",
       "651                                                                                                                                                                                                                                   no comments  \n",
       "652                                                                                                                                                                                   MEETS YOU BEFORE EXAM ASKS YOU QUESTIONS YOU'VE NEVER HEARD  \n",
       "653                                                              CURRENT THOTTAA SAADHARANA MANUSHANUKU THAAN SHOCK ADIKUM NAAN NARASIMMA.. ENNA THOTTAA ANDHA CURRENTUKE SHOCK ADIKKUM    CAPTAIN Y U NO PRODUCE CURRENT FOR TAMILNADU POWERCUT?  \n",
       "654                                                                             DAI NANACHUM PADAM MATTUM THA RELEASE PANREN... YENNA NAMBI NEENGA PADAM PAKALAMDA...  AANA IVANAILAM ORU AALA MATHICHU IVANUGALA NAMBI PADATHUKU POVATHINGADA...  \n",
       "655                                                                                                                                                                     ONGI ADICHA ONDRA TON WEIGHT AH??  AIYO, ATHU SATHIYAMA NAAN ILLAINGA NA.  \n",
       "656                                                                                                                            YENDA UNGA APPA PEYARAI FRIDGEKULLA YELUDHI VACHURUKA.? AVARDHANDA SONNARU EN PEYAR KETTUPOGAMMA PAARTHUKONNU....!  \n",
       "657                                                                                                                                                                                        WATCHING OTHER LANGUAGE MOVIE IN THEATRE SOOOPER APPU!  \n",
       "658                                                                                                                                                                            ENAKU PIDITHA YENNALME THOLAI THURATHULA ANDRU NILLA.. INDRU NEE..  \n",
       "659                                                                                                                                                                                        EVERUONE IS A GANGSTER  UNTIL THE REAL GANSTER ARRIVES  \n",
       "660                                                                                                                                                                                                                                     *TN GIRLS  \n",
       "661                                                                                                                                                            CLIMAX SCENE LA MAGALA KOOTITU VANDHU ALUDHA ELLARUM SUNA PANA AAGIDA MUDIYUMA? DA  \n",
       "662                                                                                                                                           DURING SCHOOL DAYS  ME & FRNDS: ANNA MAIL CHECK PANNANUM LAST SYSTEM KODUNGA  BROWSING CENTER ANNA:  \n",
       "663                                                                                                                                                                                                    REMEMBER THIS KID    THIS KID IS RIGHT NOW  \n",
       "664                                                                                                                                                                                        *SINGLES: IVAN VERA ENGALA ROMBA TORCHER PANDRANDA...!  \n",
       "665                                                                                          ENAIKI PUBG LA CHICKEN DINNER VANGURA PAYALUGALAM YARUUU  ORU KALATHULA DUCK KA SHOOT PANNA MUDIYAMA ENDHA NAI KITAA HE HE HE VANGUNA PAYALUGA THANN  \n",
       "666                                                                                                                                                                                                                                LATER THAT DAY  \n",
       "\n",
       "[667 rows x 2 columns]"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>imagename</th>\n",
       "      <th>captions</th>\n",
       "      <th>not_troll</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Not_troll_0.jpg</td>\n",
       "      <td>Ada pikkalipayalugala</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Not_troll_1.jpg</td>\n",
       "      <td>Etho sambavam nadandhirukkum pola</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Not_troll_10.jpg</td>\n",
       "      <td>Vunnayellam frienda vechirukken paaru</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Not_troll_100.jpg</td>\n",
       "      <td>Idho! Ivan dhaan naan nasama ponadhukku kaaranam</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Not_troll_1000.jpg</td>\n",
       "      <td>Ennada lusu thanama pesikittu irukka lusu payale</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Not_troll_1001.jpg</td>\n",
       "      <td>Vuhvuhvuhvuhvuh........</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Not_troll_1002.jpg</td>\n",
       "      <td>Yenna ore pattasu sathama kekkudhu..\\nOhhoo.. inaikku New year ah.. sari naama Namma velaiya paapom</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Not_troll_1003.jpg</td>\n",
       "      <td>Inimel ipidi post pottal vailaye vettuven</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Not_troll_1004.jpg</td>\n",
       "      <td>Ivaingakittarundhu yepidi thapikka porenanu theriyalaye</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Not_troll_1005.jpg</td>\n",
       "      <td>(no caption)</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Not_troll_1006.jpg</td>\n",
       "      <td>(no caption)</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Not_troll_1007.jpg</td>\n",
       "      <td>Inaikku mattum match win panniten naan jaichiten..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Not_troll_1008.jpg</td>\n",
       "      <td>Karupu vulaipaligalin vannam.. theriuma vunakku?</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Not_troll_1009.jpg</td>\n",
       "      <td>Andha group yellam poiyacha???</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Not_troll_101.jpg</td>\n",
       "      <td>Childhood Memories ☺️☺️☺️</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Not_troll_1010.jpg</td>\n",
       "      <td>Deii.. naanpatukku sivanennu dhanada irundhen..\\nYaaru vambu thumbukkavadhu ponena..?</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Not_troll_1011.jpg</td>\n",
       "      <td>Idhellam yenga vuruppudapogudhu..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Not_troll_1012.jpg</td>\n",
       "      <td>Avasarapattu vettidadhinganey..  naan verum like mattum dhaan potten.. post pottadhu vera yaaro..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Not_troll_1013.jpg</td>\n",
       "      <td>Ivan oruthan....  Neram,kaalam theriyama comedy pannuvaan...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Not_troll_1014.jpg</td>\n",
       "      <td>Anaivarukkum iniya kaalai vanakkam..\\nAama ivaru kaalai vanakkam sollalana yaarukku poludhu vidiyadhu paaru..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Not_troll_1015.jpg</td>\n",
       "      <td>Pattasai konjam thalli vechi vedikkurela pakkathula yen vandi nikkidhu..\\nPadadhu padadhu..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Not_troll_1016.jpg</td>\n",
       "      <td>Ennaiya vambilukkuradhe vunakku polappa pochi..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Not_troll_1017.jpg</td>\n",
       "      <td>Iru vadivelugalum govaluvai thediyapoludhu..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Not_troll_102.jpg</td>\n",
       "      <td>Podhum itthoda niruthikka..</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Not_troll_103.jpg</td>\n",
       "      <td>(no caption)</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Not_troll_104.jpg</td>\n",
       "      <td>Podhuma oottunadhu..\\nReel andhupochi...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Not_troll_105.jpg</td>\n",
       "      <td>Arumaiyaana karutthu...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Not_troll_106.jpg</td>\n",
       "      <td>Vittula oru velaium seiuradhilla! Facebookla vandhu adutthavangalukku advice panna vendiyadhu...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Not_troll_107.jpg</td>\n",
       "      <td>Idhoda niruthikka! Idhellam kavidhainu solladhinga naan aludhuruven😭😭😭</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Not_troll_108.jpg</td>\n",
       "      <td>Yenga thaikku yevalo dhillu paathiyaa..??</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2270</th>\n",
       "      <td>troll_416.png</td>\n",
       "      <td>FLIPKART  Ivanungalae Big Billion day nu onna aaramipaangalam  Adhula Out of stock products ah just oru rooba ku Sold out nu potu pilem kaatuvangalaam</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2271</th>\n",
       "      <td>troll_417.jpg</td>\n",
       "      <td>Whenever vanitha opens her mouth chai sariyana lambadi pombalaiya irupa pola ivala pethaingla ila....</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2272</th>\n",
       "      <td>troll_418.jpg</td>\n",
       "      <td>No Captions</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2273</th>\n",
       "      <td>troll_419.jpg</td>\n",
       "      <td>idhil yethanai vulladhu 3 alla 4 ??     yennada idhu... yenakku therinja konja nenja kannukuda marandhudum pola... yenanga padhil</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2274</th>\n",
       "      <td>troll_42.jpg</td>\n",
       "      <td>Evalo Prachana Irundhalum  Nalla Mooku mutta Saptom na  Prachana Theerudho illayo...  Naama Prachanaya Marandhruvom !!!</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2275</th>\n",
       "      <td>troll_420.jpg</td>\n",
       "      <td>No Captions</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2276</th>\n",
       "      <td>troll_421.jpg</td>\n",
       "      <td>HOW VIJAY TV CELEBRATE BODI   PALAYA PATIENT KUMKI, RAJA RANI,  SAATAI, BAAHUBALI 2 AH DISCHARGE PANUNGA   PODHU PATIENT PARIYERUM PERUMAL, VADA CHENNAI, SAAMY 2 , CCV AH ADMIT PANUNGA</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2277</th>\n",
       "      <td>troll_422.png</td>\n",
       "      <td>ANDHA 1Re PENDRIVE IRUKKA ENNA PENDRIVE IRUKKA  ANDHA 900 RS LUMIA IRUKKA ENNA LUMIA IRUKKA   SERI VIDUPA, ANDHA 1300 RS  SAMSUNG TAB AACHUM IRUKKA  ENNA TAB AACHUM IRUKKA</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2278</th>\n",
       "      <td>troll_423.jpg</td>\n",
       "      <td>WHAT YOU ORDERED...  WHAT THEY DELEVERED...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2279</th>\n",
       "      <td>troll_424.jpg</td>\n",
       "      <td>No Captions</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2280</th>\n",
       "      <td>troll_425.png</td>\n",
       "      <td>unayum vitu veikalay antha vjd</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2281</th>\n",
       "      <td>troll_426.jpg</td>\n",
       "      <td>in boys movie    thuniya thokkikodi ninanjida pogudhu   appo ni thokkalana thuni ninanjidukkum apidithana....</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2282</th>\n",
       "      <td>troll_427.jpg</td>\n",
       "      <td>That NALLAVAN Moment   While watching a Movie with Family and when a KISSING Scene Comes</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2283</th>\n",
       "      <td>troll_428.jpg</td>\n",
       "      <td>No Captions</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2284</th>\n",
       "      <td>troll_429.jpg</td>\n",
       "      <td>idhukku kalaila padichi irukalam idea illatha pasanga!!</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2285</th>\n",
       "      <td>troll_43.jpg</td>\n",
       "      <td>largest family in the world  ithu yellam nambura mari ah ga irruku</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2286</th>\n",
       "      <td>troll_430.jpg</td>\n",
       "      <td>ponnukku 100 kilo periya vengayamum....  50 kilo sinna vengayamum kudutthudunga kalyanam pesi pudichidalam...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2287</th>\n",
       "      <td>troll_431.jpg</td>\n",
       "      <td>poratasi effect     enga thala zebra ku evlo dhillu paathiya..?!!</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2288</th>\n",
       "      <td>troll_432.png</td>\n",
       "      <td>THAT AWESOME MOMENT WHEN U FIND MONEY FROM UR OLD DRESS AT MONTH END</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2289</th>\n",
       "      <td>troll_433.jpg</td>\n",
       "      <td>IN GIRLS VERSION  IN BOYS VERSION</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2290</th>\n",
       "      <td>troll_434.jpg</td>\n",
       "      <td>perumbettil paraparappu  valibarkaidhu..!    kaidhu seium naal: 23.02.2015    kaidhu seium idam : P.D.S  thirumana mandabam      kaidhu  anavar   A. Rajesh  kaidhu seidhavar  C. Sarmila    kuttram:   pennin  manadhai thirudivittar   thirpu:   mundru  mudichu podudhal    mukkiya satchigal     G. selvam   N. purushothamman   mattrum    perumbedu kiramam  nanbargal</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2291</th>\n",
       "      <td>troll_435.jpg</td>\n",
       "      <td>My bra 2000 rupees  branded    ethu antha roattu kadaila edutha athuva..</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2292</th>\n",
       "      <td>troll_436.png</td>\n",
       "      <td>INDIAN CRICKET TEAM  in sub continents   in  OVERSEAS</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2293</th>\n",
       "      <td>troll_437.jpg</td>\n",
       "      <td>No Captions</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2294</th>\n",
       "      <td>troll_438.png</td>\n",
       "      <td>GET A 1300 GRE SCORE RETAKES IT</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2295</th>\n",
       "      <td>troll_439.jpg</td>\n",
       "      <td>amma yaaro orutharu... pattu paaduraaruma... amma nadikkuraruma... kattuthanama dancelam aaduraru ivar dhaaan STRah ma?</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2296</th>\n",
       "      <td>troll_44.jpg</td>\n",
       "      <td>Phone  n Twitter hacked please don't respond  any random messages.. My back end team is working on getting things in control.    HACKED PAE!</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2297</th>\n",
       "      <td>troll_440.jpg</td>\n",
       "      <td>Respected deat  @narendramodi ji hearty corigratulations...  You made it !!!  God bless..</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2298</th>\n",
       "      <td>troll_441.jpg</td>\n",
       "      <td>Engu irutha nan mnnil piranthidum bothu   vera enga irunthan andrea va love panitu irunthan</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2299</th>\n",
       "      <td>troll_442.jpg</td>\n",
       "      <td>When STR fan hears about milk..     Paal ice... cup ice... paal ice...     Yen da...  Unakku vera paatu edhuvum kadaikalaya?    Enapa... Patta balama iruku... Balamurugam kovil ka?   Murugar kovil nu sollalam la.. Yen da balamurugan solura?   Ball kadakudhu paathu va pa..  Baal ball..   Adhuku pandhu nu sollalam la... Yen ball nu solra... Sodikkadhinga da ennaya..</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2300 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               imagename  \\\n",
       "0        Not_troll_0.jpg   \n",
       "1        Not_troll_1.jpg   \n",
       "2       Not_troll_10.jpg   \n",
       "3      Not_troll_100.jpg   \n",
       "4     Not_troll_1000.jpg   \n",
       "5     Not_troll_1001.jpg   \n",
       "6     Not_troll_1002.jpg   \n",
       "7     Not_troll_1003.jpg   \n",
       "8     Not_troll_1004.jpg   \n",
       "9     Not_troll_1005.jpg   \n",
       "10    Not_troll_1006.jpg   \n",
       "11    Not_troll_1007.jpg   \n",
       "12    Not_troll_1008.jpg   \n",
       "13    Not_troll_1009.jpg   \n",
       "14     Not_troll_101.jpg   \n",
       "15    Not_troll_1010.jpg   \n",
       "16    Not_troll_1011.jpg   \n",
       "17    Not_troll_1012.jpg   \n",
       "18    Not_troll_1013.jpg   \n",
       "19    Not_troll_1014.jpg   \n",
       "20    Not_troll_1015.jpg   \n",
       "21    Not_troll_1016.jpg   \n",
       "22    Not_troll_1017.jpg   \n",
       "23     Not_troll_102.jpg   \n",
       "24     Not_troll_103.jpg   \n",
       "25     Not_troll_104.jpg   \n",
       "26     Not_troll_105.jpg   \n",
       "27     Not_troll_106.jpg   \n",
       "28     Not_troll_107.jpg   \n",
       "29     Not_troll_108.jpg   \n",
       "...                  ...   \n",
       "2270       troll_416.png   \n",
       "2271       troll_417.jpg   \n",
       "2272       troll_418.jpg   \n",
       "2273       troll_419.jpg   \n",
       "2274        troll_42.jpg   \n",
       "2275       troll_420.jpg   \n",
       "2276       troll_421.jpg   \n",
       "2277       troll_422.png   \n",
       "2278       troll_423.jpg   \n",
       "2279       troll_424.jpg   \n",
       "2280       troll_425.png   \n",
       "2281       troll_426.jpg   \n",
       "2282       troll_427.jpg   \n",
       "2283       troll_428.jpg   \n",
       "2284       troll_429.jpg   \n",
       "2285        troll_43.jpg   \n",
       "2286       troll_430.jpg   \n",
       "2287       troll_431.jpg   \n",
       "2288       troll_432.png   \n",
       "2289       troll_433.jpg   \n",
       "2290       troll_434.jpg   \n",
       "2291       troll_435.jpg   \n",
       "2292       troll_436.png   \n",
       "2293       troll_437.jpg   \n",
       "2294       troll_438.png   \n",
       "2295       troll_439.jpg   \n",
       "2296        troll_44.jpg   \n",
       "2297       troll_440.jpg   \n",
       "2298       troll_441.jpg   \n",
       "2299       troll_442.jpg   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                               captions  \\\n",
       "0                                                                                                                                                                                                                                                                                                                                                                 Ada pikkalipayalugala   \n",
       "1                                                                                                                                                                                                                                                                                                                                                     Etho sambavam nadandhirukkum pola   \n",
       "2                                                                                                                                                                                                                                                                                                                                                 Vunnayellam frienda vechirukken paaru   \n",
       "3                                                                                                                                                                                                                                                                                                                                      Idho! Ivan dhaan naan nasama ponadhukku kaaranam   \n",
       "4                                                                                                                                                                                                                                                                                                                                      Ennada lusu thanama pesikittu irukka lusu payale   \n",
       "5                                                                                                                                                                                                                                                                                                                                                               Vuhvuhvuhvuhvuh........   \n",
       "6                                                                                                                                                                                                                                                                                   Yenna ore pattasu sathama kekkudhu..\\nOhhoo.. inaikku New year ah.. sari naama Namma velaiya paapom   \n",
       "7                                                                                                                                                                                                                                                                                                                                             Inimel ipidi post pottal vailaye vettuven   \n",
       "8                                                                                                                                                                                                                                                                                                                               Ivaingakittarundhu yepidi thapikka porenanu theriyalaye   \n",
       "9                                                                                                                                                                                                                                                                                                                                                                          (no caption)   \n",
       "10                                                                                                                                                                                                                                                                                                                                                                         (no caption)   \n",
       "11                                                                                                                                                                                                                                                                                                                                   Inaikku mattum match win panniten naan jaichiten..   \n",
       "12                                                                                                                                                                                                                                                                                                                                     Karupu vulaipaligalin vannam.. theriuma vunakku?   \n",
       "13                                                                                                                                                                                                                                                                                                                                                       Andha group yellam poiyacha???   \n",
       "14                                                                                                                                                                                                                                                                                                                                                            Childhood Memories ☺️☺️☺️   \n",
       "15                                                                                                                                                                                                                                                                                                Deii.. naanpatukku sivanennu dhanada irundhen..\\nYaaru vambu thumbukkavadhu ponena..?   \n",
       "16                                                                                                                                                                                                                                                                                                                                                    Idhellam yenga vuruppudapogudhu..   \n",
       "17                                                                                                                                                                                                                                                                                    Avasarapattu vettidadhinganey..  naan verum like mattum dhaan potten.. post pottadhu vera yaaro..   \n",
       "18                                                                                                                                                                                                                                                                                                                         Ivan oruthan....  Neram,kaalam theriyama comedy pannuvaan...   \n",
       "19                                                                                                                                                                                                                                                                        Anaivarukkum iniya kaalai vanakkam..\\nAama ivaru kaalai vanakkam sollalana yaarukku poludhu vidiyadhu paaru..   \n",
       "20                                                                                                                                                                                                                                                                                          Pattasai konjam thalli vechi vedikkurela pakkathula yen vandi nikkidhu..\\nPadadhu padadhu..   \n",
       "21                                                                                                                                                                                                                                                                                                                                      Ennaiya vambilukkuradhe vunakku polappa pochi..   \n",
       "22                                                                                                                                                                                                                                                                                                                                         Iru vadivelugalum govaluvai thediyapoludhu..   \n",
       "23                                                                                                                                                                                                                                                                                                                                                          Podhum itthoda niruthikka..   \n",
       "24                                                                                                                                                                                                                                                                                                                                                                         (no caption)   \n",
       "25                                                                                                                                                                                                                                                                                                                                             Podhuma oottunadhu..\\nReel andhupochi...   \n",
       "26                                                                                                                                                                                                                                                                                                                                                              Arumaiyaana karutthu...   \n",
       "27                                                                                                                                                                                                                                                                                     Vittula oru velaium seiuradhilla! Facebookla vandhu adutthavangalukku advice panna vendiyadhu...   \n",
       "28                                                                                                                                                                                                                                                                                                               Idhoda niruthikka! Idhellam kavidhainu solladhinga naan aludhuruven😭😭😭   \n",
       "29                                                                                                                                                                                                                                                                                                                                            Yenga thaikku yevalo dhillu paathiyaa..??   \n",
       "...                                                                                                                                                                                                                                                                                                                                                                                 ...   \n",
       "2270                                                                                                                                                                                                                             FLIPKART  Ivanungalae Big Billion day nu onna aaramipaangalam  Adhula Out of stock products ah just oru rooba ku Sold out nu potu pilem kaatuvangalaam   \n",
       "2271                                                                                                                                                                                                                                                                              Whenever vanitha opens her mouth chai sariyana lambadi pombalaiya irupa pola ivala pethaingla ila....   \n",
       "2272                                                                                                                                                                                                                                                                                                                                                                        No Captions   \n",
       "2273                                                                                                                                                                                                                                                  idhil yethanai vulladhu 3 alla 4 ??     yennada idhu... yenakku therinja konja nenja kannukuda marandhudum pola... yenanga padhil   \n",
       "2274                                                                                                                                                                                                                                                            Evalo Prachana Irundhalum  Nalla Mooku mutta Saptom na  Prachana Theerudho illayo...  Naama Prachanaya Marandhruvom !!!   \n",
       "2275                                                                                                                                                                                                                                                                                                                                                                        No Captions   \n",
       "2276                                                                                                                                                                                           HOW VIJAY TV CELEBRATE BODI   PALAYA PATIENT KUMKI, RAJA RANI,  SAATAI, BAAHUBALI 2 AH DISCHARGE PANUNGA   PODHU PATIENT PARIYERUM PERUMAL, VADA CHENNAI, SAAMY 2 , CCV AH ADMIT PANUNGA   \n",
       "2277                                                                                                                                                                                                        ANDHA 1Re PENDRIVE IRUKKA ENNA PENDRIVE IRUKKA  ANDHA 900 RS LUMIA IRUKKA ENNA LUMIA IRUKKA   SERI VIDUPA, ANDHA 1300 RS  SAMSUNG TAB AACHUM IRUKKA  ENNA TAB AACHUM IRUKKA   \n",
       "2278                                                                                                                                                                                                                                                                                                                                        WHAT YOU ORDERED...  WHAT THEY DELEVERED...   \n",
       "2279                                                                                                                                                                                                                                                                                                                                                                        No Captions   \n",
       "2280                                                                                                                                                                                                                                                                                                                                                     unayum vitu veikalay antha vjd   \n",
       "2281                                                                                                                                                                                                                                                                      in boys movie    thuniya thokkikodi ninanjida pogudhu   appo ni thokkalana thuni ninanjidukkum apidithana....   \n",
       "2282                                                                                                                                                                                                                                                                                           That NALLAVAN Moment   While watching a Movie with Family and when a KISSING Scene Comes   \n",
       "2283                                                                                                                                                                                                                                                                                                                                                                        No Captions   \n",
       "2284                                                                                                                                                                                                                                                                                                                            idhukku kalaila padichi irukalam idea illatha pasanga!!   \n",
       "2285                                                                                                                                                                                                                                                                                                                 largest family in the world  ithu yellam nambura mari ah ga irruku   \n",
       "2286                                                                                                                                                                                                                                                                      ponnukku 100 kilo periya vengayamum....  50 kilo sinna vengayamum kudutthudunga kalyanam pesi pudichidalam...   \n",
       "2287                                                                                                                                                                                                                                                                                                                  poratasi effect     enga thala zebra ku evlo dhillu paathiya..?!!   \n",
       "2288                                                                                                                                                                                                                                                                                                               THAT AWESOME MOMENT WHEN U FIND MONEY FROM UR OLD DRESS AT MONTH END   \n",
       "2289                                                                                                                                                                                                                                                                                                                                                  IN GIRLS VERSION  IN BOYS VERSION   \n",
       "2290  perumbettil paraparappu  valibarkaidhu..!    kaidhu seium naal: 23.02.2015    kaidhu seium idam : P.D.S  thirumana mandabam      kaidhu  anavar   A. Rajesh  kaidhu seidhavar  C. Sarmila    kuttram:   pennin  manadhai thirudivittar   thirpu:   mundru  mudichu podudhal    mukkiya satchigal     G. selvam   N. purushothamman   mattrum    perumbedu kiramam  nanbargal        \n",
       "2291                                                                                                                                                                                                                                                                                                           My bra 2000 rupees  branded    ethu antha roattu kadaila edutha athuva..   \n",
       "2292                                                                                                                                                                                                                                                                                                                              INDIAN CRICKET TEAM  in sub continents   in  OVERSEAS   \n",
       "2293                                                                                                                                                                                                                                                                                                                                                                        No Captions   \n",
       "2294                                                                                                                                                                                                                                                                                                                                                    GET A 1300 GRE SCORE RETAKES IT   \n",
       "2295                                                                                                                                                                                                                                                            amma yaaro orutharu... pattu paaduraaruma... amma nadikkuraruma... kattuthanama dancelam aaduraru ivar dhaaan STRah ma?   \n",
       "2296                                                                                                                                                                                                                                       Phone  n Twitter hacked please don't respond  any random messages.. My back end team is working on getting things in control.    HACKED PAE!   \n",
       "2297                                                                                                                                                                                                                                                                                          Respected deat  @narendramodi ji hearty corigratulations...  You made it !!!  God bless..   \n",
       "2298                                                                                                                                                                                                                                                                                        Engu irutha nan mnnil piranthidum bothu   vera enga irunthan andrea va love panitu irunthan   \n",
       "2299     When STR fan hears about milk..     Paal ice... cup ice... paal ice...     Yen da...  Unakku vera paatu edhuvum kadaikalaya?    Enapa... Patta balama iruku... Balamurugam kovil ka?   Murugar kovil nu sollalam la.. Yen da balamurugan solura?   Ball kadakudhu paathu va pa..  Baal ball..   Adhuku pandhu nu sollalam la... Yen ball nu solra... Sodikkadhinga da ennaya..   \n",
       "\n",
       "      not_troll  label  \n",
       "0             1      0  \n",
       "1             1      0  \n",
       "2             1      0  \n",
       "3             1      0  \n",
       "4             1      0  \n",
       "5             1      0  \n",
       "6             1      0  \n",
       "7             1      0  \n",
       "8             1      0  \n",
       "9             1      0  \n",
       "10            1      0  \n",
       "11            1      0  \n",
       "12            1      0  \n",
       "13            1      0  \n",
       "14            1      0  \n",
       "15            1      0  \n",
       "16            1      0  \n",
       "17            1      0  \n",
       "18            1      0  \n",
       "19            1      0  \n",
       "20            1      0  \n",
       "21            1      0  \n",
       "22            1      0  \n",
       "23            1      0  \n",
       "24            1      0  \n",
       "25            1      0  \n",
       "26            1      0  \n",
       "27            1      0  \n",
       "28            1      0  \n",
       "29            1      0  \n",
       "...         ...    ...  \n",
       "2270          0      1  \n",
       "2271          0      1  \n",
       "2272          0      1  \n",
       "2273          0      1  \n",
       "2274          0      1  \n",
       "2275          0      1  \n",
       "2276          0      1  \n",
       "2277          0      1  \n",
       "2278          0      1  \n",
       "2279          0      1  \n",
       "2280          0      1  \n",
       "2281          0      1  \n",
       "2282          0      1  \n",
       "2283          0      1  \n",
       "2284          0      1  \n",
       "2285          0      1  \n",
       "2286          0      1  \n",
       "2287          0      1  \n",
       "2288          0      1  \n",
       "2289          0      1  \n",
       "2290          0      1  \n",
       "2291          0      1  \n",
       "2292          0      1  \n",
       "2293          0      1  \n",
       "2294          0      1  \n",
       "2295          0      1  \n",
       "2296          0      1  \n",
       "2297          0      1  \n",
       "2298          0      1  \n",
       "2299          0      1  \n",
       "\n",
       "[2300 rows x 4 columns]"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_path = \"meme-classification/data/uploaded_tamil_memes\"\n",
    "dataloader = ImageDataLoaders.from_df(df, train_path, item_tfms=Resize(128, 128), bs=_batch_size, batch_tfms= tfms, valid_pct=0, shuffle_train=False, label_col=3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "batch no. :  0\n",
      "TensorImage([[[[-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          ...,\n",
      "          [-2.1179, -2.1179, -1.4713,  ..., -0.2718, -0.3900, -0.5826],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -0.2002, -0.3137, -0.4954],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -0.1916, -0.2944, -0.4632]],\n",
      "\n",
      "         [[-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          ...,\n",
      "          [-2.0357, -2.0357, -1.5155,  ..., -0.7734, -0.8655, -1.0067],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -0.7492, -0.8471, -0.9563],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -0.6971, -0.8084, -0.9129]],\n",
      "\n",
      "         [[-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          ...,\n",
      "          [-1.8044, -1.8044, -1.3299,  ..., -0.7420, -0.8014, -0.8916],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -0.7024, -0.7792, -0.8509],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -0.6559, -0.7538, -0.8338]]],\n",
      "\n",
      "\n",
      "        [[[-2.1179, -2.1179, -2.1179,  ...,  2.1623,  2.1502,  2.1392],\n",
      "          [-2.1179, -2.1179, -2.1179,  ...,  2.1627,  2.1304,  2.0733],\n",
      "          [-2.1179, -2.1179, -2.1179,  ...,  2.1697,  2.1112,  2.0145],\n",
      "          ...,\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179]],\n",
      "\n",
      "         [[-2.0357, -2.0357, -2.0357,  ...,  2.0833,  1.9432,  1.7517],\n",
      "          [-2.0357, -2.0357, -2.0357,  ...,  2.1327,  1.9468,  1.6620],\n",
      "          [-2.0357, -2.0357, -2.0357,  ...,  2.2334,  2.0195,  1.6870],\n",
      "          ...,\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357]],\n",
      "\n",
      "         [[-1.8044, -1.8044, -1.8044,  ...,  1.3699,  1.1031,  0.8827],\n",
      "          [-1.8044, -1.8044, -1.8044,  ...,  1.5882,  1.3092,  0.9734],\n",
      "          [-1.8044, -1.8044, -1.8044,  ...,  1.8552,  1.6008,  1.2346],\n",
      "          ...,\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044]]],\n",
      "\n",
      "\n",
      "        [[[-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          ...,\n",
      "          [ 0.9429,  0.8234,  0.8648,  ..., -0.2800, -0.2599, -0.2424],\n",
      "          [ 0.9576,  0.7924,  0.8475,  ..., -0.3058, -0.2806, -0.2543],\n",
      "          [ 0.9598,  0.7771,  0.8419,  ..., -0.3434, -0.3249, -0.3128]],\n",
      "\n",
      "         [[-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          ...,\n",
      "          [ 0.9595,  0.8254,  0.8874,  ..., -0.9777, -0.9435, -0.9115],\n",
      "          [ 0.9370,  0.7631,  0.8346,  ..., -1.0228, -0.9849, -0.9452],\n",
      "          [ 0.9003,  0.7242,  0.7937,  ..., -1.0579, -1.0242, -0.9898]],\n",
      "\n",
      "         [[-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          ...,\n",
      "          [ 0.1140,  0.0063,  0.0620,  ..., -1.3183, -1.3055, -1.2888],\n",
      "          [ 0.1049, -0.0428,  0.0188,  ..., -1.3529, -1.3417, -1.3243],\n",
      "          [ 0.0832, -0.0699, -0.0069,  ..., -1.3593, -1.3534, -1.3466]]],\n",
      "\n",
      "\n",
      "        [[[ 0.2006,  0.4443,  0.5806,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-1.6271, -1.5950, -1.5828,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.0878, -2.0885, -2.0869,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          ...,\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179],\n",
      "          [-2.1179, -2.1179, -2.1179,  ..., -2.1179, -2.1179, -2.1179]],\n",
      "\n",
      "         [[ 0.3771,  0.6181,  0.7485,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-1.5385, -1.5076, -1.4970,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0111, -2.0117, -2.0123,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          ...,\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357],\n",
      "          [-2.0357, -2.0357, -2.0357,  ..., -2.0357, -2.0357, -2.0357]],\n",
      "\n",
      "         [[-1.7577, -1.7576, -1.7576,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.7576, -1.7576, -1.7576,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.7766, -1.7769, -1.7772,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          ...,\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044],\n",
      "          [-1.8044, -1.8044, -1.8044,  ..., -1.8044, -1.8044, -1.8044]]]],\n",
      "       device='cuda:0')\n",
      "TensorCategory([0, 1, 0, 0], device='cuda:0')\n"
     ]
    }
   ],
   "source": [
    "for idx, data in enumerate(dataloader.train.new(shuffle=False)):\n",
    "    # get the inputs\n",
    "    text_model.eval()\n",
    "    print(\"batch no. : \", idx)\n",
    "    inputs, labels = data\n",
    "    print(inputs)\n",
    "    print(labels)\n",
    "    break"
   ]
  },
  {
   "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}