layoutlm-document-sequence-labeling-model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "reverse-terrorist",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:00:58.300165Z",
     "iopub.status.busy": "2021-05-30T14:00:58.298556Z",
     "iopub.status.idle": "2021-05-30T14:01:00.063705Z",
     "shell.execute_reply": "2021-05-30T14:01:00.064164Z",
     "shell.execute_reply.started": "2021-05-30T13:23:18.838749Z"
    },
    "papermill": {
     "duration": 1.783104,
     "end_time": "2021-05-30T14:01:00.064427",
     "exception": false,
     "start_time": "2021-05-30T14:00:58.281323",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import subprocess\n",
    "import custom_preprocess\n",
    "from IPython.display import Image, display"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aquatic-moral",
   "metadata": {
    "papermill": {
     "duration": 0.011708,
     "end_time": "2021-05-30T14:01:00.088895",
     "exception": false,
     "start_time": "2021-05-30T14:01:00.077187",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Setting up the Environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "determined-missouri",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_kg_hide-input": true,
    "_kg_hide-output": true,
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2021-05-30T14:01:00.120973Z",
     "iopub.status.busy": "2021-05-30T14:01:00.120391Z",
     "iopub.status.idle": "2021-05-30T14:01:20.510156Z",
     "shell.execute_reply": "2021-05-30T14:01:20.509582Z",
     "shell.execute_reply.started": "2021-05-30T05:36:54.794116Z"
    },
    "papermill": {
     "duration": 20.40951,
     "end_time": "2021-05-30T14:01:20.510328",
     "exception": false,
     "start_time": "2021-05-30T14:01:00.100818",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing /kaggle/working/unilm/layoutlm\n",
      "Collecting transformers==2.9.0\n",
      "  Downloading transformers-2.9.0-py3-none-any.whl (635 kB)\n",
      "Collecting tensorboardX==2.0\n",
      "  Downloading tensorboardX-2.0-py2.py3-none-any.whl (195 kB)\n",
      "Collecting lxml==4.5.1\n",
      "  Downloading lxml-4.5.1-cp37-cp37m-manylinux1_x86_64.whl (5.5 MB)\n",
      "Collecting seqeval==0.0.12\n",
      "  Downloading seqeval-0.0.12.tar.gz (21 kB)\n",
      "Collecting Pillow==7.1.2\n",
      "  Downloading Pillow-7.1.2-cp37-cp37m-manylinux1_x86_64.whl (2.1 MB)\n",
      "Requirement already satisfied: numpy>=1.14.0 in /opt/conda/lib/python3.7/site-packages (from seqeval==0.0.12->layoutlm==0.0) (1.19.5)\n",
      "Requirement already satisfied: Keras>=2.2.4 in /opt/conda/lib/python3.7/site-packages (from seqeval==0.0.12->layoutlm==0.0) (2.4.3)\n",
      "Requirement already satisfied: protobuf>=3.8.0 in /opt/conda/lib/python3.7/site-packages (from tensorboardX==2.0->layoutlm==0.0) (3.15.8)\n",
      "Requirement already satisfied: six in /opt/conda/lib/python3.7/site-packages (from tensorboardX==2.0->layoutlm==0.0) (1.15.0)\n",
      "Collecting tokenizers==0.7.0\n",
      "  Downloading tokenizers-0.7.0-cp37-cp37m-manylinux1_x86_64.whl (5.6 MB)\n",
      "Requirement already satisfied: requests in /opt/conda/lib/python3.7/site-packages (from transformers==2.9.0->layoutlm==0.0) (2.25.1)\n",
      "Requirement already satisfied: tqdm>=4.27 in /opt/conda/lib/python3.7/site-packages (from transformers==2.9.0->layoutlm==0.0) (4.59.0)\n",
      "Requirement already satisfied: regex!=2019.12.17 in /opt/conda/lib/python3.7/site-packages (from transformers==2.9.0->layoutlm==0.0) (2021.3.17)\n",
      "Requirement already satisfied: sacremoses in /opt/conda/lib/python3.7/site-packages (from transformers==2.9.0->layoutlm==0.0) (0.0.45)\n",
      "Requirement already satisfied: sentencepiece in /opt/conda/lib/python3.7/site-packages (from transformers==2.9.0->layoutlm==0.0) (0.1.95)\n",
      "Requirement already satisfied: filelock in /opt/conda/lib/python3.7/site-packages (from transformers==2.9.0->layoutlm==0.0) (3.0.12)\n",
      "Requirement already satisfied: h5py in /opt/conda/lib/python3.7/site-packages (from Keras>=2.2.4->seqeval==0.0.12->layoutlm==0.0) (2.10.0)\n",
      "Requirement already satisfied: pyyaml in /opt/conda/lib/python3.7/site-packages (from Keras>=2.2.4->seqeval==0.0.12->layoutlm==0.0) (5.3.1)\n",
      "Requirement already satisfied: scipy>=0.14 in /opt/conda/lib/python3.7/site-packages (from Keras>=2.2.4->seqeval==0.0.12->layoutlm==0.0) (1.5.4)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.7/site-packages (from requests->transformers==2.9.0->layoutlm==0.0) (2020.12.5)\n",
      "Requirement already satisfied: idna<3,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests->transformers==2.9.0->layoutlm==0.0) (2.10)\n",
      "Requirement already satisfied: chardet<5,>=3.0.2 in /opt/conda/lib/python3.7/site-packages (from requests->transformers==2.9.0->layoutlm==0.0) (4.0.0)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests->transformers==2.9.0->layoutlm==0.0) (1.26.4)\n",
      "Requirement already satisfied: joblib in /opt/conda/lib/python3.7/site-packages (from sacremoses->transformers==2.9.0->layoutlm==0.0) (1.0.1)\n",
      "Requirement already satisfied: click in /opt/conda/lib/python3.7/site-packages (from sacremoses->transformers==2.9.0->layoutlm==0.0) (7.1.2)\n",
      "Building wheels for collected packages: layoutlm, seqeval\n",
      "  Building wheel for layoutlm (setup.py): started\n",
      "  Building wheel for layoutlm (setup.py): finished with status 'done'\n",
      "  Created wheel for layoutlm: filename=layoutlm-0.0-py3-none-any.whl size=11485 sha256=3274528e1c43f8d0c8d3328bdd0b24d6ab2a1dd5cc4ada2d1b4aecc92e0f348c\n",
      "  Stored in directory: /tmp/pip-ephem-wheel-cache-po04qkm1/wheels/20/8c/c3/1b8351c14a0c100501699b82d43277172453291475eac2534f\n",
      "  Building wheel for seqeval (setup.py): started\n",
      "  Building wheel for seqeval (setup.py): finished with status 'done'\n",
      "  Created wheel for seqeval: filename=seqeval-0.0.12-py3-none-any.whl size=7422 sha256=98436c794e4bb5b030f9ffd0261fc0f24485088264fb45d8061f6236f20c28b4\n",
      "  Stored in directory: /root/.cache/pip/wheels/dc/cc/62/a3b81f92d35a80e39eb9b2a9d8b31abac54c02b21b2d466edc\n",
      "Successfully built layoutlm seqeval\n",
      "Installing collected packages: tokenizers, transformers, tensorboardX, seqeval, Pillow, lxml, layoutlm\n",
      "  Attempting uninstall: tokenizers\n",
      "    Found existing installation: tokenizers 0.10.2\n",
      "    Uninstalling tokenizers-0.10.2:\n",
      "      Successfully uninstalled tokenizers-0.10.2\n",
      "  Attempting uninstall: transformers\n",
      "    Found existing installation: transformers 4.5.1\n",
      "    Uninstalling transformers-4.5.1:\n",
      "      Successfully uninstalled transformers-4.5.1\n",
      "  Attempting uninstall: tensorboardX\n",
      "    Found existing installation: tensorboardX 2.2\n",
      "    Uninstalling tensorboardX-2.2:\n",
      "      Successfully uninstalled tensorboardX-2.2\n",
      "  Attempting uninstall: Pillow\n",
      "    Found existing installation: Pillow 7.2.0\n",
      "    Uninstalling Pillow-7.2.0:\n",
      "      Successfully uninstalled Pillow-7.2.0\n",
      "  Attempting uninstall: lxml\n",
      "    Found existing installation: lxml 4.6.3\n",
      "    Uninstalling lxml-4.6.3:\n",
      "      Successfully uninstalled lxml-4.6.3\n",
      "Successfully installed Pillow-7.1.2 layoutlm-0.0 lxml-4.5.1 seqeval-0.0.12 tensorboardX-2.0 tokenizers-0.7.0 transformers-2.9.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Cloning into 'unilm'...\n",
      "ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
      "catalyst 21.4.1 requires tensorboardX>=2.1.0, but you have tensorboardx 2.0 which is incompatible.\n",
      "allennlp 2.3.0 requires transformers<4.6,>=4.1, but you have transformers 2.9.0 which is incompatible.\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "\n",
    "cd /kaggle/working\n",
    "git clone https://github.com/microsoft/unilm.git\n",
    "cd unilm/layoutlm\n",
    "pip install ."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "heard-housing",
   "metadata": {
    "papermill": {
     "duration": 0.012521,
     "end_time": "2021-05-30T14:01:20.536485",
     "exception": false,
     "start_time": "2021-05-30T14:01:20.523964",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "According to github repo, \"First, we need to preprocess the JSON file into txt. We can run the preprocessing scripts funsd_preprocess.py in the scripts directory. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "instant-victim",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:01:20.570183Z",
     "iopub.status.busy": "2021-05-30T14:01:20.569444Z",
     "iopub.status.idle": "2021-05-30T14:01:42.491377Z",
     "shell.execute_reply": "2021-05-30T14:01:42.490940Z",
     "shell.execute_reply.started": "2021-05-30T05:37:50.710429Z"
    },
    "papermill": {
     "duration": 21.942111,
     "end_time": "2021-05-30T14:01:42.491523",
     "exception": false,
     "start_time": "2021-05-30T14:01:20.549412",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CompletedProcess(args='./preprocess.sh', returncode=0)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os.chdir('/kaggle/working/unilm/layoutlm/examples/seq_labeling')\n",
    "subprocess.run('./preprocess.sh') "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "diagnostic-bunch",
   "metadata": {
    "papermill": {
     "duration": 0.012891,
     "end_time": "2021-05-30T14:01:42.518853",
     "exception": false,
     "start_time": "2021-05-30T14:01:42.505962",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Model \n",
    "\n",
    "The LayoutLM model has been pretrained on 11M documents using BERT uncased models (base and large) along with the token coordinates. We can use the pretrained model to label parts of other document sets with only minimal tuning.  We can use the FUNSD dataset to tune the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "honey-awareness",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:01:42.549808Z",
     "iopub.status.busy": "2021-05-30T14:01:42.549060Z",
     "iopub.status.idle": "2021-05-30T14:01:42.568194Z",
     "shell.execute_reply": "2021-05-30T14:01:42.567412Z",
     "shell.execute_reply.started": "2021-05-25T06:47:40.741788Z"
    },
    "papermill": {
     "duration": 0.036392,
     "end_time": "2021-05-30T14:01:42.568321",
     "exception": false,
     "start_time": "2021-05-30T14:01:42.531929",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "labels.txt\n",
      "test.txt\n",
      "test.txt.tmp\n",
      "test_box.txt\n",
      "test_box.txt.tmp\n",
      "test_image.txt\n",
      "test_image.txt.tmp\n",
      "testing_data\n",
      "train.txt\n",
      "train.txt.tmp\n",
      "train_box.txt\n",
      "train_box.txt.tmp\n",
      "train_image.txt\n",
      "train_image.txt.tmp\n",
      "training_data\n",
      "\n",
      "B-ANSWER\n",
      "B-HEADER\n",
      "B-QUESTION\n",
      "E-ANSWER\n",
      "E-HEADER\n",
      "E-QUESTION\n",
      "I-ANSWER\n",
      "I-HEADER\n",
      "I-QUESTION\n",
      "O\n",
      "S-ANSWER\n",
      "S-HEADER\n",
      "S-QUESTION\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "\n",
    "ls data\n",
    "echo \"\"\n",
    "cat data/labels.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "racial-substitute",
   "metadata": {
    "papermill": {
     "duration": 0.013907,
     "end_time": "2021-05-30T14:01:42.596094",
     "exception": false,
     "start_time": "2021-05-30T14:01:42.582187",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Train the model with the command given in the github repo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "presidential-terminal",
   "metadata": {
    "_kg_hide-output": true,
    "execution": {
     "iopub.execute_input": "2021-05-30T14:01:42.639210Z",
     "iopub.status.busy": "2021-05-30T14:01:42.638441Z",
     "iopub.status.idle": "2021-05-30T14:10:24.483521Z",
     "shell.execute_reply": "2021-05-30T14:10:24.482609Z",
     "shell.execute_reply.started": "2021-05-30T05:38:21.577861Z"
    },
    "papermill": {
     "duration": 521.873653,
     "end_time": "2021-05-30T14:10:24.483665",
     "exception": false,
     "start_time": "2021-05-30T14:01:42.610012",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2021-05-30 14:01:46.728334: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0\r\n",
      "Epoch:   0%|                                             | 0/60 [00:00<?, ?it/s]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A/opt/conda/lib/python3.7/site-packages/transformers/optimization.py:155: UserWarning: This overload of add_ is deprecated:\r\n",
      "\tadd_(Number alpha, Tensor other)\r\n",
      "Consider using one of the following signatures instead:\r\n",
      "\tadd_(Tensor other, *, Number alpha) (Triggered internally at  /opt/conda/conda-bld/pytorch_1603729047590/work/torch/csrc/utils/python_arg_parser.cpp:882.)\r\n",
      "  exp_avg.mul_(beta1).add_(1.0 - beta1, grad)\r\n",
      "\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:01<00:15,  1.77s/it]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:02<00:09,  1.24s/it]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:03<00:07,  1.08s/it]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:04<00:05,  1.00it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:05<00:04,  1.05it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:06<00:03,  1.07it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:07<00:02,  1.09it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.11it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:08<00:00,  1.12it/s]\u001b[A/opt/conda/lib/python3.7/site-packages/torch/optim/lr_scheduler.py:247: UserWarning: To get the last learning rate computed by the scheduler, please use `get_last_lr()`.\r\n",
      "  warnings.warn(\"To get the last learning rate computed by the scheduler, \"\r\n",
      "\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:09<00:00,  1.09it/s]\r\n",
      "Epoch:   2%|▌                                    | 1/60 [00:09<09:00,  9.17s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:   3%|█▏                                   | 2/60 [00:17<08:20,  8.63s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:   5%|█▊                                   | 3/60 [00:25<08:01,  8.45s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:   7%|██▍                                  | 4/60 [00:33<07:49,  8.38s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.13it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.20it/s]\r\n",
      "Epoch:   8%|███                                  | 5/60 [00:42<07:39,  8.35s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  10%|███▋                                 | 6/60 [00:50<07:29,  8.32s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.13it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.13it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  12%|████▎                                | 7/60 [00:58<07:20,  8.31s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  13%|████▉                                | 8/60 [01:07<07:11,  8.29s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.13it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  15%|█████▌                               | 9/60 [01:15<07:02,  8.29s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  17%|██████                              | 10/60 [01:23<06:53,  8.27s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  18%|██████▌                             | 11/60 [01:31<06:44,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  20%|███████▏                            | 12/60 [01:40<06:36,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.13it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  22%|███████▊                            | 13/60 [01:48<06:28,  8.27s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  23%|████████▍                           | 14/60 [01:56<06:19,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  25%|█████████                           | 15/60 [02:04<06:11,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  27%|█████████▌                          | 16/60 [02:13<06:03,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:08,  1.12it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.12it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.13it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  28%|██████████▏                         | 17/60 [02:21<05:55,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.13it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  30%|██████████▊                         | 18/60 [02:29<05:46,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  32%|███████████▍                        | 19/60 [02:37<05:38,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.13it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  33%|████████████                        | 20/60 [02:46<05:30,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  35%|████████████▌                       | 21/60 [02:54<05:21,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  37%|█████████████▏                      | 22/60 [03:02<05:13,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  38%|█████████████▊                      | 23/60 [03:10<05:04,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.16it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.12it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.20it/s]\r\n",
      "Epoch:  40%|██████████████▍                     | 24/60 [03:19<04:57,  8.27s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  42%|███████████████                     | 25/60 [03:27<04:48,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  43%|███████████████▌                    | 26/60 [03:35<04:40,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  45%|████████████████▏                   | 27/60 [03:43<04:32,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.13it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  47%|████████████████▊                   | 28/60 [03:52<04:23,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  48%|█████████████████▍                  | 29/60 [04:00<04:15,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  50%|██████████████████                  | 30/60 [04:08<04:06,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  52%|██████████████████▌                 | 31/60 [04:16<03:58,  8.22s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.16it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.13it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.13it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.13it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.12it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.20it/s]\r\n",
      "Epoch:  53%|███████████████████▏                | 32/60 [04:25<03:50,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  55%|███████████████████▊                | 33/60 [04:33<03:42,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  57%|████████████████████▍               | 34/60 [04:41<03:34,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  58%|█████████████████████               | 35/60 [04:49<03:25,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  60%|█████████████████████▌              | 36/60 [04:57<03:18,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  62%|██████████████████████▏             | 37/60 [05:06<03:09,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  63%|██████████████████████▊             | 38/60 [05:14<03:01,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  65%|███████████████████████▍            | 39/60 [05:22<02:52,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  67%|████████████████████████            | 40/60 [05:30<02:44,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  68%|████████████████████████▌           | 41/60 [05:39<02:36,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  70%|█████████████████████████▏          | 42/60 [05:47<02:28,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  72%|█████████████████████████▊          | 43/60 [05:55<02:19,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.13it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  73%|██████████████████████████▍         | 44/60 [06:03<02:11,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  75%|███████████████████████████         | 45/60 [06:12<02:03,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  77%|███████████████████████████▌        | 46/60 [06:20<01:55,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  78%|████████████████████████████▏       | 47/60 [06:28<01:47,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  80%|████████████████████████████▊       | 48/60 [06:36<01:38,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  82%|█████████████████████████████▍      | 49/60 [06:44<01:30,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  83%|██████████████████████████████      | 50/60 [06:53<01:22,  8.23s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  85%|██████████████████████████████▌     | 51/60 [07:01<01:14,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:08,  1.11it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:07,  1.13it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.13it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  87%|███████████████████████████████▏    | 52/60 [07:09<01:06,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.14it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  88%|███████████████████████████████▊    | 53/60 [07:17<00:57,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  90%|████████████████████████████████▍   | 54/60 [07:26<00:49,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.13it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.13it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.20it/s]\r\n",
      "Epoch:  92%|█████████████████████████████████   | 55/60 [07:34<00:41,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.13it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.13it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch:  93%|█████████████████████████████████▌  | 56/60 [07:42<00:33,  8.27s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  95%|██████████████████████████████████▏ | 57/60 [07:50<00:24,  8.25s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.16it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.14it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.14it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.14it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.14it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.15it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.22it/s]\r\n",
      "Epoch:  97%|██████████████████████████████████▊ | 58/60 [07:59<00:16,  8.24s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.13it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.14it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:07<00:01,  1.14it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.13it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.20it/s]\r\n",
      "Epoch:  98%|███████████████████████████████████▍| 59/60 [08:07<00:08,  8.26s/it]\r\n",
      "Iteration:   0%|                                         | 0/10 [00:00<?, ?it/s]\u001b[A\r\n",
      "Iteration:  10%|███▎                             | 1/10 [00:00<00:07,  1.15it/s]\u001b[A\r\n",
      "Iteration:  20%|██████▌                          | 2/10 [00:01<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  30%|█████████▉                       | 3/10 [00:02<00:06,  1.15it/s]\u001b[A\r\n",
      "Iteration:  40%|█████████████▏                   | 4/10 [00:03<00:05,  1.15it/s]\u001b[A\r\n",
      "Iteration:  50%|████████████████▌                | 5/10 [00:04<00:04,  1.15it/s]\u001b[A\r\n",
      "Iteration:  60%|███████████████████▊             | 6/10 [00:05<00:03,  1.15it/s]\u001b[A\r\n",
      "Iteration:  70%|███████████████████████          | 7/10 [00:06<00:02,  1.15it/s]\u001b[A\r\n",
      "Iteration:  80%|██████████████████████████▍      | 8/10 [00:06<00:01,  1.15it/s]\u001b[A\r\n",
      "Iteration:  90%|█████████████████████████████▋   | 9/10 [00:07<00:00,  1.14it/s]\u001b[A\r\n",
      "Iteration: 100%|████████████████████████████████| 10/10 [00:08<00:00,  1.21it/s]\r\n",
      "Epoch: 100%|████████████████████████████████████| 60/60 [08:15<00:00,  8.26s/it]\r\n"
     ]
    }
   ],
   "source": [
    "! python run_seq_labeling.py  --data_dir data \\\n",
    "                            --model_type layoutlm \\\n",
    "                            --model_name_or_path /kaggle/input/layoutlm/layoutlm-base-uncased \\\n",
    "                            --do_lower_case \\\n",
    "                            --max_seq_length 512 \\\n",
    "                            --do_train \\\n",
    "                            --num_train_epochs 60 \\\n",
    "                            --logging_steps 10 \\\n",
    "                            --save_steps -1 \\\n",
    "                            --overwrite_output_dir \\\n",
    "                            --output_dir output \\\n",
    "                            --labels data/labels.txt \\\n",
    "                            --per_gpu_train_batch_size 16 \\\n",
    "                            --per_gpu_eval_batch_size 16 \\\n",
    "                            --seed 12"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "crude-fabric",
   "metadata": {
    "papermill": {
     "duration": 0.165613,
     "end_time": "2021-05-30T14:10:24.816639",
     "exception": false,
     "start_time": "2021-05-30T14:10:24.651026",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Evaluating the model in FUNSD test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "transparent-extra",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:10:25.160926Z",
     "iopub.status.busy": "2021-05-30T14:10:25.155914Z",
     "iopub.status.idle": "2021-05-30T14:10:46.528527Z",
     "shell.execute_reply": "2021-05-30T14:10:46.528062Z",
     "shell.execute_reply.started": "2021-05-30T05:39:15.685407Z"
    },
    "papermill": {
     "duration": 21.546766,
     "end_time": "2021-05-30T14:10:46.528657",
     "exception": false,
     "start_time": "2021-05-30T14:10:24.981891",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2021-05-30 14:10:28.856624: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0\r\n",
      "Evaluating: 100%|█████████████████████████████████| 7/7 [00:01<00:00,  5.89it/s]\r\n"
     ]
    }
   ],
   "source": [
    "!python run_seq_labeling.py  --do_eval \\\n",
    "                            --data_dir data \\\n",
    "                            --model_type layoutlm \\\n",
    "                            --model_name_or_path output \\\n",
    "                            --do_lower_case \\\n",
    "                            --output_dir output \\\n",
    "                            --labels data/labels.txt \\\n",
    "                            --fp16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "directed-tractor",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:10:46.879855Z",
     "iopub.status.busy": "2021-05-30T14:10:46.878904Z",
     "iopub.status.idle": "2021-05-30T14:10:47.504431Z",
     "shell.execute_reply": "2021-05-30T14:10:47.503917Z",
     "shell.execute_reply.started": "2021-05-30T05:41:14.154451Z"
    },
    "papermill": {
     "duration": 0.804524,
     "end_time": "2021-05-30T14:10:47.504614",
     "exception": false,
     "start_time": "2021-05-30T14:10:46.700090",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1 = 0.7865765330758088\r\n",
      "loss = 1.5561955996922083\r\n",
      "precision = 0.7601493233784414\r\n",
      "recall = 0.8149074537268635\r\n"
     ]
    }
   ],
   "source": [
    "!cat output/eval_results.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "environmental-infection",
   "metadata": {
    "papermill": {
     "duration": 0.30633,
     "end_time": "2021-05-30T14:10:48.044156",
     "exception": false,
     "start_time": "2021-05-30T14:10:47.737826",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Predicting in a New Image\n",
    "We can do evaluation or inference by replacing --do_train with --do_eval or --do_predict. Inorder to prepare our custom image for the preprocessing script, Pytessaract is used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "naval-locking",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:10:48.390986Z",
     "iopub.status.busy": "2021-05-30T14:10:48.390353Z",
     "iopub.status.idle": "2021-05-30T14:10:48.410781Z",
     "shell.execute_reply": "2021-05-30T14:10:48.410286Z",
     "shell.execute_reply.started": "2021-05-30T13:23:34.964369Z"
    },
    "papermill": {
     "duration": 0.195514,
     "end_time": "2021-05-30T14:10:48.410904",
     "exception": false,
     "start_time": "2021-05-30T14:10:48.215390",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(Image(filename='/kaggle/input/test-image/82252956_2958.png'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "planned-oxide",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:10:48.802826Z",
     "iopub.status.busy": "2021-05-30T14:10:48.801942Z",
     "iopub.status.idle": "2021-05-30T14:10:54.173639Z",
     "shell.execute_reply": "2021-05-30T14:10:54.173170Z",
     "shell.execute_reply.started": "2021-05-30T05:48:12.928021Z"
    },
    "papermill": {
     "duration": 5.56101,
     "end_time": "2021-05-30T14:10:54.173771",
     "exception": false,
     "start_time": "2021-05-30T14:10:48.612761",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d0d1a74a649e446e8539ea1242331b59",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading:   0%|          | 0.00/570 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c01a379a216a42b797615f3fde398c4a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading:   0%|          | 0.00/232k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "322661acec62405a9a2f5d9491cb26ae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading:   0%|          | 0.00/466k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3b043a8f45bb4e54b7e2e3f5b6fe2098",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading:   0%|          | 0.00/28.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "transform=custom_preprocess.custom_img_annotation_()\n",
    "ocr_df=transform.write_annoteFile('/kaggle/input/test-image/82252956_2958.png')\n",
    "transform.convert()\n",
    "transform.seg()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "satellite-bridal",
   "metadata": {
    "papermill": {
     "duration": 0.201258,
     "end_time": "2021-05-30T14:10:54.551503",
     "exception": false,
     "start_time": "2021-05-30T14:10:54.350245",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Pytessaract has done poorly to detect words, has ommitted many words and has read many words wrongly. Maybe, poor resolution might be a reason. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "technological-hughes",
   "metadata": {
    "papermill": {
     "duration": 0.240831,
     "end_time": "2021-05-30T14:10:55.084237",
     "exception": false,
     "start_time": "2021-05-30T14:10:54.843406",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "**Prediciting the sequence labeling in new Image**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "iraqi-malawi",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:10:55.445765Z",
     "iopub.status.busy": "2021-05-30T14:10:55.444866Z",
     "iopub.status.idle": "2021-05-30T14:11:15.514931Z",
     "shell.execute_reply": "2021-05-30T14:11:15.514049Z",
     "shell.execute_reply.started": "2021-05-30T05:48:22.723667Z"
    },
    "papermill": {
     "duration": 20.255942,
     "end_time": "2021-05-30T14:11:15.515062",
     "exception": false,
     "start_time": "2021-05-30T14:10:55.259120",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2021-05-30 14:10:59.086170: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0\r\n",
      "Evaluating: 100%|█████████████████████████████████| 7/7 [00:01<00:00,  5.94it/s]\r\n"
     ]
    }
   ],
   "source": [
    "!python run_seq_labeling.py  --do_predict \\\n",
    "                            --data_dir data \\\n",
    "                            --model_type layoutlm \\\n",
    "                            --model_name_or_path output \\\n",
    "                            --do_lower_case \\\n",
    "                            --output_dir output \\\n",
    "                            --labels data/labels.txt \\\n",
    "                            --fp16"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "resistant-willow",
   "metadata": {
    "papermill": {
     "duration": 0.178976,
     "end_time": "2021-05-30T14:11:15.876857",
     "exception": false,
     "start_time": "2021-05-30T14:11:15.697881",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Output are tagged in **BIESO format** (B-beginning, I-intermediate, E-end, S-single word\n",
    "entity and O-outside) format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "senior-chest",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-05-30T14:11:16.402086Z",
     "iopub.status.busy": "2021-05-30T14:11:16.399840Z",
     "iopub.status.idle": "2021-05-30T14:11:17.467945Z",
     "shell.execute_reply": "2021-05-30T14:11:17.467409Z",
     "shell.execute_reply.started": "2021-05-30T05:48:47.412836Z"
    },
    "papermill": {
     "duration": 1.364902,
     "end_time": "2021-05-30T14:11:17.468069",
     "exception": false,
     "start_time": "2021-05-30T14:11:16.103167",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KA, S-QUESTION\r\n",
      "\r\n",
      "Sparrow S-QUESTION\r\n",
      "\r\n",
      "FROM: S-QUESTION\r\n",
      "\r\n",
      "D. S-QUESTION\r\n",
      "\r\n",
      "J. O\r\n",
      "\r\n",
      "Landro B-HEADER\r\n",
      "\r\n",
      "May I-HEADER\r\n",
      "\r\n",
      "4[ E-HEADER\r\n",
      "\r\n",
      "aus B-QUESTION\r\n",
      "\r\n",
      "4| I-QUESTION\r\n",
      "\r\n",
      "JUN E-QUESTION\r\n",
      "\r\n",
      "23) B-QUESTION\r\n",
      "\r\n",
      "SEP I-QUESTION\r\n",
      "\r\n",
      "45 E-QUESTION\r\n",
      "\r\n",
      "‘SUBJECT: B-QUESTION\r\n",
      "\r\n",
      "OLD E-QUESTION\r\n",
      "\r\n",
      "GOLD B-QUESTION\r\n",
      "\r\n",
      "MENTHOL E-QUESTION\r\n",
      "\r\n",
      "LIGHTS B-QUESTION\r\n",
      "\r\n",
      "& I-QUESTION\r\n",
      "\r\n",
      "ULTRA E-QUESTION\r\n",
      "\r\n",
      "LIGHTS B-QUESTION\r\n",
      "\r\n",
      "10'S E-QUESTION\r\n",
      "\r\n",
      "- B-QUESTION\r\n",
      "\r\n",
      "PROGRESS E-QUESTION\r\n",
      "\r\n",
      "REPORT S-QUESTION\r\n",
      "\r\n",
      "REGION: S-ANSWER\r\n",
      "\r\n",
      "(ONLY S-QUESTION\r\n",
      "\r\n",
      "IF S-QUESTION\r\n",
      "\r\n",
      "PARTIAL S-ANSWER\r\n",
      "\r\n",
      "REGION S-QUESTION\r\n",
      "\r\n",
      "CONTINUE S-QUESTION\r\n",
      "\r\n",
      "WITH S-QUESTION\r\n",
      "\r\n",
      "DIVISION(S) S-QUESTION\r\n",
      "\r\n",
      "SCOPE) S-QUESTION\r\n",
      "\r\n",
      "DIVISION: S-QUESTION\r\n",
      "\r\n",
      "DIVISION S-ANSWER\r\n",
      "\r\n",
      "NAME: S-ANSWER\r\n",
      "\r\n",
      "Milw. S-ANSWER\r\n",
      "\r\n",
      "South_ S-ANSWER\r\n",
      "\r\n",
      "DIVISION S-ANSWER\r\n",
      "\r\n",
      "NAME: S-ANSWER\r\n",
      "\r\n",
      "#REP S-ANSWER\r\n",
      "\r\n",
      "DIVISION O\r\n",
      "\r\n",
      "NAME: B-HEADER\r\n",
      "\r\n",
      "Milw. I-HEADER\r\n",
      "\r\n",
      "North O\r\n",
      "\r\n",
      "_ E-HEADER\r\n",
      "\r\n",
      "DIVISION O\r\n",
      "\r\n",
      "NAME: O\r\n",
      "\r\n",
      "#REP O\r\n",
      "\r\n",
      "z O\r\n",
      "\r\n",
      "DIVISION O\r\n",
      "\r\n",
      "NAME: O\r\n",
      "\r\n",
      "DIVISION O\r\n",
      "\r\n",
      "NAME: O\r\n",
      "\r\n",
      "#REPS O\r\n",
      "\r\n",
      "DIRECT O\r\n",
      "\r\n",
      "ACCOUNTS O\r\n",
      "\r\n",
      "AND O\r\n",
      "\r\n",
      "CHAINS O\r\n",
      "\r\n",
      "HEADQUARTERED O\r\n",
      "\r\n",
      "WITHIN O\r\n",
      "\r\n",
      "THE O\r\n",
      "\r\n",
      "REGION O\r\n",
      "\r\n",
      "(15 O\r\n",
      "\r\n",
      "+ B-QUESTION\r\n",
      "\r\n",
      "STORES) I-QUESTION\r\n",
      "\r\n",
      "STOCKING I-QUESTION\r\n",
      "\r\n",
      "NO I-QUESTION\r\n",
      "\r\n",
      "OLD I-QUESTION\r\n",
      "\r\n",
      "GOLD I-QUESTION\r\n",
      "\r\n",
      "MENTHOL I-QUESTION\r\n",
      "\r\n",
      "LIGHTS I-QUESTION\r\n",
      "\r\n",
      "OR I-QUESTION\r\n",
      "\r\n",
      "ULTRA E-QUESTION\r\n",
      "\r\n",
      "LIGHTS B-ANSWER\r\n",
      "\r\n",
      "100'S I-ANSWER\r\n",
      "\r\n",
      "[Walgreen I-ANSWER\r\n",
      "\r\n",
      "Drug I-ANSWER\r\n",
      "\r\n",
      "Inga E-ANSWER\r\n",
      "\r\n",
      "93] B-ANSWER\r\n",
      "\r\n",
      "otecoto E-ANSWER\r\n",
      "\r\n",
      "xis B-ANSWER\r\n",
      "\r\n",
      "Paget E-ANSWER\r\n",
      "\r\n",
      "of B-ANSWER\r\n",
      "\r\n",
      "3 I-ANSWER\r\n",
      "\r\n",
      "Pages I-ANSWER\r\n",
      "\r\n",
      "95675208 I-ANSWER\r\n",
      "\r\n"
     ]
    }
   ],
   "source": [
    "!cat output/test_predictions.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "different-choice",
   "metadata": {
    "papermill": {
     "duration": 0.184038,
     "end_time": "2021-05-30T14:11:17.840539",
     "exception": false,
     "start_time": "2021-05-30T14:11:17.656501",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "test_pridictions.txt isn't a linked output. **Hit a like if you liked the notebook.**"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 626.802936,
   "end_time": "2021-05-30T14:11:18.669784",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2021-05-30T14:00:51.866848",
   "version": "2.3.3"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "0828309abb614369b00d42d8e6f4b84a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "0cbc05ee46674235b92793ee7a4a0ddb": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "11865d48aa9c42ffbbb4f8d0d9fc1b1e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0828309abb614369b00d42d8e6f4b84a",
       "placeholder": "​",
       "style": "IPY_MODEL_963417ee7e0b4370a594b015edb4a001",
       "value": "Downloading: 100%"
      }
     },
     "14a00dda39a24eefa8a6b03c7ce7229d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_adc59763f5a44426930d21cb236e0a37",
       "placeholder": "​",
       "style": "IPY_MODEL_a5c55c02384b4b22a91cd1767797fb5f",
       "value": "Downloading: 100%"
      }
     },
     "19a61f9c275d410486af54a6849f294e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "1baef9bd731a4deeb6f3f9d81b9a6016": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_93a572a451ee4238aa5c706e36d783ca",
       "placeholder": "​",
       "style": "IPY_MODEL_6030fcdfcc514bcfad3e2c2297a9c3bc",
       "value": "Downloading: 100%"
      }
     },
     "21c8367ac05944c3a02780e7eb91e549": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_e6bcb89e138149ddb411402b83d2c0cc",
       "placeholder": "​",
       "style": "IPY_MODEL_a12309e9c2ca42e59daa6bc7fdbf95d5",
       "value": " 28.0/28.0 [00:00&lt;00:00, 843B/s]"
      }
     },
     "225ba93f81a04ef38cf99758d157d48f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "2466ed7869034a09854287f6d89c7272": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_f1c9489f83814070b576fddfe573eaa7",
       "placeholder": "​",
       "style": "IPY_MODEL_3bfb3a8b645b416cb9793d089159f462",
       "value": " 232k/232k [00:00&lt;00:00, 1.44MB/s]"
      }
     },
     "28a5f8eaee52409a892acef2154db11e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_225ba93f81a04ef38cf99758d157d48f",
       "placeholder": "​",
       "style": "IPY_MODEL_0cbc05ee46674235b92793ee7a4a0ddb",
       "value": " 570/570 [00:00&lt;00:00, 19.5kB/s]"
      }
     },
     "322661acec62405a9a2f5d9491cb26ae": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_11865d48aa9c42ffbbb4f8d0d9fc1b1e",
        "IPY_MODEL_a5018ec4b1f44681b0080d471d5306d1",
        "IPY_MODEL_5d50eaefd46643af8861bc6dfb930299"
       ],
       "layout": "IPY_MODEL_a3cb8957de33458cae1e4ce3a42be200"
      }
     },
     "3b043a8f45bb4e54b7e2e3f5b6fe2098": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_14a00dda39a24eefa8a6b03c7ce7229d",
        "IPY_MODEL_6db74c0d2f304d9184942f344c6a59f3",
        "IPY_MODEL_21c8367ac05944c3a02780e7eb91e549"
       ],
       "layout": "IPY_MODEL_860f3e7fda874a3fafd6a5086f52628a"
      }
     },
     "3bfb3a8b645b416cb9793d089159f462": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "3c4d7a9672294a9b81dc6c14dd475ce0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "41c0ba1c2bbb423face04998790c52aa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "4ac8382684dd460fa175329fc20fdfc6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "5adfe6ed244f411bb724ffe5081b7d7a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_4ac8382684dd460fa175329fc20fdfc6",
       "max": 570.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_d4cb00c5181445ca9299f1dca09fc055",
       "value": 570.0
      }
     },
     "5d50eaefd46643af8861bc6dfb930299": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_aa5e4d1441fe4fca840aa3546428b2e8",
       "placeholder": "​",
       "style": "IPY_MODEL_19a61f9c275d410486af54a6849f294e",
       "value": " 466k/466k [00:00&lt;00:00, 1.42MB/s]"
      }
     },
     "6030fcdfcc514bcfad3e2c2297a9c3bc": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "6db74c0d2f304d9184942f344c6a59f3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_acb5b71383fc41b89b13e3c721cc765b",
       "max": 28.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_af12ec2a40484c09b08e17a4f43852d2",
       "value": 28.0
      }
     },
     "7c4b93a8b71144bc8fb1b214c4522034": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "860f3e7fda874a3fafd6a5086f52628a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "87dc0a4565e643efb1efb2dddfeed5a7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "93a572a451ee4238aa5c706e36d783ca": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "963417ee7e0b4370a594b015edb4a001": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "9a39cdb2a76b4ca1bffc83d62379cf83": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "a12309e9c2ca42e59daa6bc7fdbf95d5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "a3cb8957de33458cae1e4ce3a42be200": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "a5018ec4b1f44681b0080d471d5306d1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_9a39cdb2a76b4ca1bffc83d62379cf83",
       "max": 466062.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_c5952f285c794f72b3dd6d0854e1ba71",
       "value": 466062.0
      }
     },
     "a5c55c02384b4b22a91cd1767797fb5f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "a7b5e2eda3a846a19234a1123112cc0a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "aa5e4d1441fe4fca840aa3546428b2e8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "acb5b71383fc41b89b13e3c721cc765b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "adc59763f5a44426930d21cb236e0a37": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "af12ec2a40484c09b08e17a4f43852d2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "c01a379a216a42b797615f3fde398c4a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_1baef9bd731a4deeb6f3f9d81b9a6016",
        "IPY_MODEL_f2716ec28bf24075b2d070c30bc8e46e",
        "IPY_MODEL_2466ed7869034a09854287f6d89c7272"
       ],
       "layout": "IPY_MODEL_87dc0a4565e643efb1efb2dddfeed5a7"
      }
     },
     "c5952f285c794f72b3dd6d0854e1ba71": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "d0d1a74a649e446e8539ea1242331b59": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_dc3f136c46ec41a0aeff504ce7c9efe3",
        "IPY_MODEL_5adfe6ed244f411bb724ffe5081b7d7a",
        "IPY_MODEL_28a5f8eaee52409a892acef2154db11e"
       ],
       "layout": "IPY_MODEL_fc25e0a1a98a47828ce5a70f4653ff8d"
      }
     },
     "d4cb00c5181445ca9299f1dca09fc055": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "dc3f136c46ec41a0aeff504ce7c9efe3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_7c4b93a8b71144bc8fb1b214c4522034",
       "placeholder": "​",
       "style": "IPY_MODEL_a7b5e2eda3a846a19234a1123112cc0a",
       "value": "Downloading: 100%"
      }
     },
     "e6bcb89e138149ddb411402b83d2c0cc": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "f1c9489f83814070b576fddfe573eaa7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "f2716ec28bf24075b2d070c30bc8e46e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_3c4d7a9672294a9b81dc6c14dd475ce0",
       "max": 231508.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_41c0ba1c2bbb423face04998790c52aa",
       "value": 231508.0
      }
     },
     "fc25e0a1a98a47828ce5a70f4653ff8d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}