02-train_SIMPLE_CNN_RNN.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "606fa719",
   "metadata": {},
   "source": [
    "# Train SIMPLE_CNN_RNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a8a5bd39",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.nn.utils import clip_grad_norm_\n",
    "\n",
    "from tqdm import tqdm\n",
    "\n",
    "from rouge_score import rouge_scorer\n",
    "\n",
    "from utils.GloveMgr import GloveMgr\n",
    "from utils.Dataset import Dataset\n",
    "from utils.DataLoader import DataLoader\n",
    "from utils.preprocess_df import preprocess_df\n",
    "from utils.accuracy_nb_sent_per_doc import accuracy_nb_sent_per_doc_fn\n",
    "from utils.accuracy_prop_sent_per_doc import accuracy_prop_sent_per_doc_fn\n",
    "from utils.split_all_docs import split_all_docs\n",
    "\n",
    "from models.SIMPLE_CNN_RNN import SIMPLE_CNN_RNN\n",
    "\n",
    "from time import time\n",
    "\n",
    "import json\n",
    "import argparse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "818410ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "def is_notebook() -> bool:\n",
    "    try:\n",
    "        shell = get_ipython().__class__.__name__\n",
    "        if shell == 'ZMQInteractiveShell':\n",
    "            return True   # Jupyter notebook or qtconsole\n",
    "        elif shell == 'TerminalInteractiveShell':\n",
    "            return False  # Terminal running IPython\n",
    "        else:\n",
    "            return False  # Other type (?)\n",
    "    except NameError:\n",
    "        return False      # Probably standard Python interpreter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "690fa032",
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_columns', None)\n",
    "pd.set_option('display.max_rows', None)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "b2636e0b",
   "metadata": {},
   "source": [
    "## (Hyper-)parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6e7aad39",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Parse args if script mode\n",
    "parser = argparse.ArgumentParser(description='extractive summary')\n",
    "\n",
    "parser.add_argument('-is_graphic',type=int,default=1,choices=[0,1])\n",
    "parser.add_argument('-gpu_num',type=int,default=0)\n",
    "parser.add_argument('-batch_size',type=int,default=32)\n",
    "parser.add_argument('-epochs',type=int,default=100)\n",
    "\n",
    "args = None\n",
    "\n",
    "if is_notebook():\n",
    "    args = parser.parse_args(\"\")\n",
    "else:\n",
    "    args = parser.parse_args()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "824eacf5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "parse:\n",
      "is_graphic: True\n",
      "cuda_num: 0\n",
      "epochs 100\n",
      "batch_size 32\n"
     ]
    }
   ],
   "source": [
    "# parameters\n",
    "is_graphic = args.is_graphic != 0\n",
    "cuda_num = args.gpu_num\n",
    "embed_name = \"glove.6B.100\"\n",
    "\n",
    "# hyper-parameters\n",
    "vocab_size = 150000\n",
    "batch_size = args.batch_size\n",
    "epochs = args.epochs\n",
    "learning_rate = 1e-3\n",
    "early_stopping = 3\n",
    "model_name = \"02-train_SIMPLE_CNN_RNN\"\n",
    "sub_folder_name = \"model_name__{}__time__{}__embed_name__{}__lr__{}__batch_size__{}__vocab_size__{}__cuda_num__{}\".format(model_name, time(), embed_name, learning_rate, batch_size, vocab_size, cuda_num)\n",
    "checkpoints_folder = \"./checkpoints/\" + sub_folder_name\n",
    "average_number_of_sentences_per_document = 3\n",
    "\n",
    "# print\n",
    "print(\"parse:\")\n",
    "print(\"is_graphic:\", is_graphic)\n",
    "print(\"cuda_num:\", cuda_num)\n",
    "print(\"epochs\", epochs)\n",
    "print(\"batch_size\", batch_size)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "c8ddd4d5",
   "metadata": {},
   "source": [
    "## PyTorch initialisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d09bc3ed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of available GPUs: 6\n",
      "GPU 0: NVIDIA GeForce GTX 1080 Ti\n",
      "GPU 1: NVIDIA GeForce GTX 1080 Ti\n",
      "GPU 2: NVIDIA GeForce GTX 1080 Ti\n",
      "GPU 3: NVIDIA GeForce GTX 1080\n",
      "GPU 4: NVIDIA GeForce GTX 1080\n",
      "GPU 5: NVIDIA GeForce GTX 1080\n"
     ]
    }
   ],
   "source": [
    "# Check if a GPU is available\n",
    "if torch.cuda.is_available():\n",
    "    # Display the number of available GPUs\n",
    "    print(f\"Number of available GPUs: {torch.cuda.device_count()}\")\n",
    "    # Display the name of each GPU\n",
    "    for i in range(torch.cuda.device_count()):\n",
    "        print(f\"GPU {i}: {torch.cuda.get_device_name(i)}\")\n",
    "else:\n",
    "    print(\"No GPU available.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b0c5786e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='cuda', index=0)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "if torch.cuda.is_available():  \n",
    "  dev = \"cuda:\" + str(cuda_num) \n",
    "else:  \n",
    "  dev = \"cpu\" \n",
    "\n",
    "device = torch.device(dev)\n",
    "device"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "3393fe4e",
   "metadata": {},
   "source": [
    "## Chargement des données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "97644745",
   "metadata": {},
   "outputs": [],
   "source": [
    "glovemgr = GloveMgr(\"./data/glove.6B/glove.6B.100d.txt\", vocab_size=vocab_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fe745c8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = pd.read_json(\"./data/cnn_dailymail/train.json\")\n",
    "train_dataset = Dataset(preprocess_df(df_train, glovemgr=glovemgr, is_sep_n=False, remove_stop_word=True, stemming=False, trunc_sent=50, padding_sent=50, trunc_doc=100, doc_column_name=\"article\", labels_column_name=\"labels\"))\n",
    "train_iter = DataLoader(dataset=train_dataset, batch_size=batch_size, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f8b39f3e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_val = pd.read_json(\"./data/cnn_dailymail/val.json\")\n",
    "val_dataset = Dataset(preprocess_df(df_val, glovemgr=glovemgr, is_sep_n=False, remove_stop_word=True, stemming=False, trunc_sent=50, padding_sent=50, trunc_doc=100, doc_column_name=\"article\", labels_column_name=\"labels\"))\n",
    "val_iter = DataLoader(dataset=val_dataset, batch_size=1, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0f086978",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test = pd.read_json(\"./data/cnn_dailymail/test.json\")\n",
    "test_dataset = Dataset(preprocess_df(df_test, glovemgr=glovemgr, is_sep_n=False, remove_stop_word=True, stemming=False, trunc_sent=50, padding_sent=50, trunc_doc=100, doc_column_name=\"article\", labels_column_name=\"labels\"))\n",
    "test_iter = DataLoader(dataset=test_dataset, batch_size=1, shuffle=False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "fc41eb07",
   "metadata": {},
   "source": [
    "## Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9683721f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "380f8ba1",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = SIMPLE_CNN_RNN(device=device, vocab_size=vocab_size, word_embed=glovemgr.getEmbeddings())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "2ae1a236",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SIMPLE_CNN_RNN(\n",
       "  (word_embedding): Embedding(150002, 100)\n",
       "  (word_convs): Sequential(\n",
       "    (0): Conv1d(100, 100, kernel_size=(4,), stride=(1,), padding=same)\n",
       "    (1): BatchNorm1d(100, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "    (2): ReLU(inplace=True)\n",
       "  )\n",
       "  (sent_GRU): GRU(100, 200, batch_first=True, bidirectional=True)\n",
       "  (rel_pos_emb): Embedding(11, 100)\n",
       "  (abs_pos_emb): Embedding(100, 100)\n",
       "  (Wdoc): Linear(in_features=400, out_features=400, bias=True)\n",
       "  (Wcontent): Linear(in_features=400, out_features=1, bias=False)\n",
       "  (Wsalience): Bilinear(in1_features=400, in2_features=400, out_features=1, bias=False)\n",
       "  (Wnovelty): Bilinear(in1_features=400, in2_features=400, out_features=1, bias=False)\n",
       "  (Wabs_pos): Linear(in_features=100, out_features=1, bias=False)\n",
       "  (Wrel_pos): Linear(in_features=100, out_features=1, bias=False)\n",
       ")"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4bc411ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_fn = nn.BCELoss()\n",
    "optimizer = torch.optim.Adamax(model.parameters(),lr=learning_rate)\n",
    "scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=1, gamma=0.5)\n",
    "mae_fn = nn.L1Loss()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "1852182a",
   "metadata": {},
   "source": [
    "## Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5c803c4f",
   "metadata": {},
   "outputs": [],
   "source": [
    "if not os.path.exists(checkpoints_folder):\n",
    "    os.makedirs(checkpoints_folder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c6618af5",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_val[\"doc_splitted\"] = split_all_docs(df_val[\"article\"], is_sep_n = False)\n",
    "val_set = df_val\n",
    "scorer = rouge_scorer.RougeScorer(['rouge1', 'rouge2', 'rougeL'], use_stemmer=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0991d3c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 1:   0%|          | 0/4 [00:00<?, ?batch/s]/home/raoufdine/.local/lib/python3.9/site-packages/torch/nn/modules/conv.py:309: UserWarning: Using padding='same' with even kernel lengths and odd dilation may require a zero-padded copy of the input be created (Triggered internally at ../aten/src/ATen/native/Convolution.cpp:895.)\n",
      "  return F.conv1d(input, weight, bias, self.stride,\n",
      "Epoch 1: 100%|██████████| 4/4 [00:01<00:00,  2.08batch/s, accuracy=0.687, loss=0.667, mae=0.282]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1 : val loss = 0.630, val mae = 0.428, val accuracy = 0.691, r1 = 0.260, r2 = 0.061, rL = 0.157\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 2: 100%|██████████| 4/4 [00:00<00:00,  5.50batch/s, accuracy=0.707, loss=0.695, mae=0.42] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 2 : val loss = 0.593, val mae = 0.407, val accuracy = 0.693, r1 = 0.267, r2 = 0.064, rL = 0.161\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 3: 100%|██████████| 4/4 [00:00<00:00,  5.81batch/s, accuracy=0.755, loss=0.451, mae=0.253]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3 : val loss = 0.562, val mae = 0.386, val accuracy = 0.702, r1 = 0.291, r2 = 0.082, rL = 0.180\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 4: 100%|██████████| 4/4 [00:00<00:00,  5.67batch/s, accuracy=0.757, loss=0.452, mae=0.243]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 4 : val loss = 0.553, val mae = 0.381, val accuracy = 0.715, r1 = 0.312, r2 = 0.100, rL = 0.198\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 5: 100%|██████████| 4/4 [00:00<00:00,  5.52batch/s, accuracy=0.755, loss=0.446, mae=0.248]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 5 : val loss = 0.545, val mae = 0.376, val accuracy = 0.730, r1 = 0.329, r2 = 0.115, rL = 0.206\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 6: 100%|██████████| 4/4 [00:00<00:00,  5.60batch/s, accuracy=0.757, loss=0.441, mae=0.253]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6 : val loss = 0.533, val mae = 0.368, val accuracy = 0.754, r1 = 0.352, r2 = 0.129, rL = 0.219\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 7: 100%|██████████| 4/4 [00:00<00:00,  5.81batch/s, accuracy=0.762, loss=0.422, mae=0.251]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 7 : val loss = 0.518, val mae = 0.357, val accuracy = 0.780, r1 = 0.371, r2 = 0.141, rL = 0.229\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 8: 100%|██████████| 4/4 [00:00<00:00,  5.61batch/s, accuracy=0.76, loss=0.436, mae=0.254] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 8 : val loss = 0.501, val mae = 0.344, val accuracy = 0.807, r1 = 0.408, r2 = 0.176, rL = 0.255\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 9: 100%|██████████| 4/4 [00:00<00:00,  5.64batch/s, accuracy=0.762, loss=0.427, mae=0.252]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 9 : val loss = 0.485, val mae = 0.331, val accuracy = 0.819, r1 = 0.420, r2 = 0.181, rL = 0.260\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 10: 100%|██████████| 4/4 [00:00<00:00,  5.53batch/s, accuracy=0.762, loss=0.435, mae=0.255]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 10 : val loss = 0.472, val mae = 0.319, val accuracy = 0.831, r1 = 0.422, r2 = 0.185, rL = 0.262\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 11: 100%|██████████| 4/4 [00:00<00:00,  5.83batch/s, accuracy=0.76, loss=0.428, mae=0.256] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 11 : val loss = 0.461, val mae = 0.309, val accuracy = 0.835, r1 = 0.426, r2 = 0.188, rL = 0.266\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 12: 100%|██████████| 4/4 [00:00<00:00,  5.74batch/s, accuracy=0.759, loss=0.432, mae=0.256]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 12 : val loss = 0.453, val mae = 0.300, val accuracy = 0.839, r1 = 0.426, r2 = 0.188, rL = 0.266\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 13: 100%|██████████| 4/4 [00:00<00:00,  5.81batch/s, accuracy=0.762, loss=0.441, mae=0.26] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 13 : val loss = 0.448, val mae = 0.295, val accuracy = 0.839, r1 = 0.427, r2 = 0.188, rL = 0.265\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 14: 100%|██████████| 4/4 [00:00<00:00,  5.49batch/s, accuracy=0.755, loss=0.434, mae=0.255]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 14 : val loss = 0.444, val mae = 0.290, val accuracy = 0.841, r1 = 0.431, r2 = 0.189, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 15: 100%|██████████| 4/4 [00:00<00:00,  5.93batch/s, accuracy=0.765, loss=0.424, mae=0.254]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 15 : val loss = 0.441, val mae = 0.286, val accuracy = 0.841, r1 = 0.431, r2 = 0.189, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 16: 100%|██████████| 4/4 [00:00<00:00,  5.84batch/s, accuracy=0.752, loss=0.446, mae=0.26] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 16 : val loss = 0.439, val mae = 0.284, val accuracy = 0.841, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 17: 100%|██████████| 4/4 [00:00<00:00,  5.61batch/s, accuracy=0.754, loss=0.443, mae=0.26] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17 : val loss = 0.438, val mae = 0.282, val accuracy = 0.841, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 18: 100%|██████████| 4/4 [00:00<00:00,  5.77batch/s, accuracy=0.76, loss=0.432, mae=0.255] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 18 : val loss = 0.437, val mae = 0.282, val accuracy = 0.841, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 19: 100%|██████████| 4/4 [00:00<00:00,  5.76batch/s, accuracy=0.752, loss=0.44, mae=0.257] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 19 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 20: 100%|██████████| 4/4 [00:00<00:00,  5.72batch/s, accuracy=0.762, loss=0.428, mae=0.255]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 20 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 21: 100%|██████████| 4/4 [00:00<00:00,  5.81batch/s, accuracy=0.763, loss=0.441, mae=0.26] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 21 : val loss = 0.437, val mae = 0.281, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 22: 100%|██████████| 4/4 [00:00<00:00,  5.52batch/s, accuracy=0.762, loss=0.422, mae=0.248]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 22 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 23: 100%|██████████| 4/4 [00:00<00:00,  5.74batch/s, accuracy=0.761, loss=0.427, mae=0.253]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 23 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 24: 100%|██████████| 4/4 [00:00<00:00,  5.74batch/s, accuracy=0.759, loss=0.431, mae=0.257]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 24 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 25: 100%|██████████| 4/4 [00:00<00:00,  5.70batch/s, accuracy=0.756, loss=0.435, mae=0.257]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 25 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 26: 100%|██████████| 4/4 [00:00<00:00,  5.76batch/s, accuracy=0.759, loss=0.43, mae=0.257] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 26 : val loss = 0.436, val mae = 0.280, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 27: 100%|██████████| 4/4 [00:00<00:00,  5.50batch/s, accuracy=0.757, loss=0.434, mae=0.253]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 27 : val loss = 0.436, val mae = 0.279, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 28: 100%|██████████| 4/4 [00:00<00:00,  5.48batch/s, accuracy=0.763, loss=0.419, mae=0.249]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 28 : val loss = 0.435, val mae = 0.279, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 29: 100%|██████████| 4/4 [00:00<00:00,  5.49batch/s, accuracy=0.763, loss=0.426, mae=0.252]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 29 : val loss = 0.436, val mae = 0.279, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 30: 100%|██████████| 4/4 [00:00<00:00,  5.55batch/s, accuracy=0.76, loss=0.429, mae=0.253] \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 30 : val loss = 0.436, val mae = 0.279, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 31: 100%|██████████| 4/4 [00:00<00:00,  5.48batch/s, accuracy=0.763, loss=0.426, mae=0.254]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 31 : val loss = 0.436, val mae = 0.279, val accuracy = 0.842, r1 = 0.432, r2 = 0.190, rL = 0.267\n",
      "Training duration = 78.97104358673096\n"
     ]
    }
   ],
   "source": [
    "arr_train_loss = []\n",
    "arr_train_mae = []\n",
    "arr_train_acc = []\n",
    "arr_val_loss = []\n",
    "arr_val_mae = []\n",
    "arr_val_acc = []\n",
    "\n",
    "t1 = time()\n",
    "for epoch in range(1, epochs+1):\n",
    "    # train 1 epoch\n",
    "    model.train()\n",
    "    nb_batch_train = 0\n",
    "    total_train_loss = 0\n",
    "    total_train_mae = 0\n",
    "    total_train_acc = 0\n",
    "    with tqdm(train_iter, unit=\"batch\", total=len(train_iter)) as tepoch:\n",
    "        for batch in tepoch:\n",
    "            tepoch.set_description(f\"Epoch {epoch}\")\n",
    "            features = []\n",
    "            doc_lens = []\n",
    "\n",
    "            for j in range(batch_size):\n",
    "                doc_lens.append(len(batch[j][\"docs\"]))\n",
    "                features = features + batch[j][\"docs\"]\n",
    "            \n",
    "            features = torch.tensor(features, dtype=torch.long).to(device)\n",
    "\n",
    "            targets = [torch.tensor(batch[j][\"labels\"], dtype=torch.float) for j in range(batch_size)]\n",
    "            targets = torch.cat(targets)\n",
    "            targets = targets.to(device)\n",
    "            \n",
    "            probs = model(features, doc_lens)\n",
    "            \n",
    "            loss = loss_fn(probs, targets)\n",
    "            optimizer.zero_grad()\n",
    "            loss.backward()\n",
    "            clip_grad_norm_(model.parameters(), 1.0)\n",
    "            optimizer.step()\n",
    "            nb_batch_train += 1\n",
    "            total_train_loss += loss.item()\n",
    "            total_train_mae += mae_fn(probs, targets).item()\n",
    "            total_train_acc += accuracy_prop_sent_per_doc_fn(probs=probs.cpu().detach().numpy(), targets=targets.cpu().detach().numpy(), doc_lens=doc_lens)\n",
    "            tepoch.set_postfix(loss=total_train_loss/nb_batch_train, mae=total_train_mae/nb_batch_train, accuracy=total_train_acc/nb_batch_train)\n",
    "\n",
    "    # Save model\n",
    "    model.save(checkpoints_folder + \"/\" + model_name + \"-\" + str(epoch) + \".pt\")\n",
    "\n",
    "    # Show train and val score\n",
    "    model.eval()\n",
    "    nb_batch_val = 0\n",
    "    total_val_loss = 0\n",
    "    total_val_mae = 0\n",
    "    total_val_acc = 0\n",
    "\n",
    "    total_r1 = 0\n",
    "    total_r2 = 0\n",
    "    total_rl = 0\n",
    "\n",
    "    for i, batch in enumerate(val_iter):\n",
    "        features = []\n",
    "        doc_lens = []\n",
    "\n",
    "        for j in range(1):\n",
    "            doc_lens.append(len(batch[j][\"docs\"]))\n",
    "            features = features + batch[j][\"docs\"]\n",
    "            \n",
    "        features = torch.tensor(features, dtype=torch.long).to(device)\n",
    "\n",
    "        targets = [torch.tensor(batch[j][\"labels\"], dtype=torch.float) for j in range(1)]\n",
    "        targets = torch.cat(targets)\n",
    "        targets = targets.to(device)\n",
    "        \n",
    "        y_pred = model(features, doc_lens)\n",
    "\n",
    "        loss = loss_fn(y_pred, targets)\n",
    "\n",
    "        nb_batch_val += 1\n",
    "        total_val_loss += loss.item()\n",
    "\n",
    "        doc = val_set[\"doc_splitted\"].iloc[i]\n",
    "        summaries = val_set[\"highlights\"].iloc[i]\n",
    "\n",
    "        indices = torch.argsort(y_pred, descending=True)\n",
    "\n",
    "        y_pred_thresh = []\n",
    "\n",
    "        txt = \"\"\n",
    "\n",
    "        for j in range(min(len(doc), y_pred.shape[0])):\n",
    "            txt = txt + \". \" + doc[indices[j]]\n",
    "            y_pred_thresh.append(indices[j])\n",
    "            if len(txt) >= len(summaries):\n",
    "                break\n",
    "\n",
    "        y_pred_thresh.sort()\n",
    "\n",
    "        txt = \"\"\n",
    "\n",
    "        for j in y_pred_thresh:\n",
    "            txt = txt + \". \" + doc[j]\n",
    "\n",
    "        n = min(len(txt), len(summaries))\n",
    "\n",
    "        while n < len(txt) and txt[n].isalnum():\n",
    "            n += 1\n",
    "\n",
    "        txt = txt[:n]\n",
    "\n",
    "        #assert len(txt) - len(summaries) <= 20\n",
    "\n",
    "        scores = scorer.score(summaries, txt)\n",
    "        total_r1 += scores[\"rouge1\"].fmeasure\n",
    "        total_r2 += scores[\"rouge2\"].fmeasure\n",
    "        total_rl += scores[\"rougeL\"].fmeasure\n",
    "\n",
    "        probs = y_pred.tolist() # compute_probs(y_pred)\n",
    "        probs_tensor = y_pred #torch.tensor(probs, dtype=torch.float).to(device)\n",
    "\n",
    "        total_val_mae += mae_fn(probs_tensor, targets).item()\n",
    "        total_val_acc += accuracy_nb_sent_per_doc_fn(probs=probs, targets=targets.cpu().detach().numpy(), doc_lens=doc_lens, average_number_of_sentences_per_document=average_number_of_sentences_per_document)\n",
    "\n",
    "    arr_train_loss.append(total_train_loss / nb_batch_train)\n",
    "    arr_train_mae.append(total_train_mae / nb_batch_train)\n",
    "    arr_train_acc.append(total_train_acc / nb_batch_train)\n",
    "\n",
    "    arr_val_loss.append(total_val_loss / nb_batch_val)\n",
    "    arr_val_mae.append(total_val_mae / nb_batch_val)\n",
    "    arr_val_acc.append(total_val_acc / nb_batch_val)\n",
    "    \n",
    "    print(\"Epoch {} : val loss = {:.3f}, val mae = {:.3f}, val accuracy = {:.3f}, r1 = {:.3f}, r2 = {:.3f}, rL = {:.3f}\".format(epoch, total_val_loss / nb_batch_val, total_val_mae / nb_batch_val, total_val_acc / nb_batch_val, total_r1 / nb_batch_val, total_r2 / nb_batch_val, total_rl / nb_batch_val))\n",
    "\n",
    "    if len(arr_val_loss) >= early_stopping+1:\n",
    "        if min(arr_val_loss[-early_stopping:]) >= arr_val_loss[-(early_stopping+1)]:\n",
    "            break\n",
    "\n",
    "    # update lr\n",
    "    scheduler.step()\n",
    "\n",
    "t2 = time()\n",
    "print(\"Training duration =\", t2-t1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2f39ceed",
   "metadata": {},
   "outputs": [],
   "source": [
    "training_metrics = {}\n",
    "training_metrics[\"duration\"]   = t2 - t1\n",
    "training_metrics[\"train_loss\"] = arr_train_loss\n",
    "training_metrics[\"train_mae\"]  = arr_train_mae\n",
    "training_metrics[\"train_acc\"]  = arr_train_acc\n",
    "training_metrics[\"val_loss\"]   = arr_val_loss\n",
    "training_metrics[\"val_mae\"]    = arr_val_mae\n",
    "training_metrics[\"val_acc\"]    = arr_val_acc\n",
    "\n",
    "# Save to file in JSON format\n",
    "\n",
    "with open(checkpoints_folder + \"/training_metrics.json\", 'w') as fp:\n",
    "    json.dump(training_metrics, fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8cd8fbee",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# draw Losses\n",
    "if is_graphic:\n",
    "    plt.plot(list(range(1, len(arr_train_loss) + 1)), arr_train_loss, label=\"train\")\n",
    "    plt.plot(list(range(1, len(arr_val_loss) + 1)), arr_val_loss, label=\"val\")\n",
    "\n",
    "    plt.title(\"Loss\")\n",
    "    plt.xlabel(\"epoch\")\n",
    "    plt.ylabel(\"loss\")\n",
    "    plt.legend(loc=\"upper left\")\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "8bda887e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# draw mae\n",
    "if is_graphic:\n",
    "    plt.plot(list(range(1, len(arr_train_mae) + 1)), arr_train_mae, label=\"train\")\n",
    "    plt.plot(list(range(1, len(arr_val_mae) + 1)), arr_val_mae, label=\"val\")\n",
    "\n",
    "    plt.title(\"MAE\")\n",
    "    plt.xlabel(\"epoch\")\n",
    "    plt.ylabel(\"mae\")\n",
    "    plt.legend(loc=\"upper right\")\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "1cc21154",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# draw Accuracies\n",
    "if is_graphic:\n",
    "    plt.plot(list(range(1, len(arr_train_acc) + 1)), arr_train_acc, label=\"train\")\n",
    "    plt.plot(list(range(1, len(arr_val_acc) + 1)), arr_val_acc, label=\"val\")\n",
    "\n",
    "    plt.title(\"Accuracy\")\n",
    "    plt.xlabel(\"epoch\")\n",
    "    plt.ylabel(\"accuracy\")\n",
    "    plt.legend(loc=\"upper left\")\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "0133ff45",
   "metadata": {},
   "source": [
    "# Eval model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e095b271",
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils.split_all_docs import split_all_docs\n",
    "from rouge_score import rouge_scorer\n",
    "from utils.threshold_probs import threshold_probs_by_prop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "273c6eb2",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = SIMPLE_CNN_RNN(device=device, vocab_size=vocab_size, word_embed=glovemgr.getEmbeddings())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "4fc41994",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "28"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_epoch = np.argmin(np.array(arr_val_loss)) + 1\n",
    "model.load(checkpoints_folder + \"/\" + model_name + \"-\" + str(best_epoch) + \".pt\")\n",
    "model.to(device)\n",
    "model.eval()\n",
    "best_epoch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "3edbae7a",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_set = df_test\n",
    "test_set[\"doc_splited\"] = split_all_docs(test_set[\"article\"], False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "bb8ab07a",
   "metadata": {},
   "outputs": [],
   "source": [
    "scorer = rouge_scorer.RougeScorer(['rouge1', 'rouge2', 'rougeL'], use_stemmer=True)\n",
    "criterion = nn.BCELoss()\n",
    "mae_fn = nn.L1Loss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "67659b8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Eval model: 100%|██████████| 100/100 [00:01<00:00, 55.51batch/s, accuracy=0.815, mae=0.302, rouge1=0.404, rouge2=0.167, rougeL=0.254]\n"
     ]
    }
   ],
   "source": [
    "arr_rouge1 = []\n",
    "arr_rouge2 = []\n",
    "arr_rougeL = []\n",
    "arr_loss = []\n",
    "mae = []\n",
    "accuracy = []\n",
    "\n",
    "idx = 0\n",
    "with tqdm(test_iter, unit=\"batch\", total=len(test_iter)) as tepoch:\n",
    "    for batch in tepoch:\n",
    "        tepoch.set_description(\"Eval model\")\n",
    "        features = []\n",
    "        doc_lens = []\n",
    "\n",
    "        for j in range(1):\n",
    "            doc_lens.append(len(batch[j][\"docs\"]))\n",
    "            features = features + batch[j][\"docs\"]\n",
    "                \n",
    "        features = torch.tensor(features, dtype=torch.long).to(device)\n",
    "\n",
    "        targets = [torch.tensor(batch[j][\"labels\"], dtype=torch.float) for j in range(1)]\n",
    "        targets = torch.cat(targets)\n",
    "        targets = targets.to(device)\n",
    "\n",
    "        y_pred = model(features, doc_lens)\n",
    "    \n",
    "        loss = criterion(y_pred, targets)\n",
    "\n",
    "        arr_loss.append(loss.item())\n",
    "\n",
    "        doc = test_set[\"doc_splited\"].iloc[idx]\n",
    "        summaries = test_set[\"highlights\"].iloc[idx]\n",
    "\n",
    "        probs = np.array(y_pred.tolist()) # compute_probs(y_pred)\n",
    "        probs_tensor = y_pred #torch.tensor(probs, dtype=torch.float).to(device)\n",
    "\n",
    "        #probs = threshold_probs_by_nb(probs=probs, doc_lens=[probs.shape[0]], average_number_of_sentences_per_document=average_number_of_sentences_per_document)\n",
    "        #probs = threshold_probs_by_prop(probs=probs, doc_lens=[probs.shape[0]], average_proportion_of_sentences_per_document=average_proportion_of_sentences_per_document)\n",
    "        indices = torch.argsort(y_pred, descending=True)\n",
    "\n",
    "        y_pred_thresh = []\n",
    "\n",
    "        txt = \"\"\n",
    "\n",
    "        for i in range(min(len(doc), y_pred.shape[0])):\n",
    "            txt = txt + \". \" + doc[indices[i]]\n",
    "            y_pred_thresh.append(indices[i])\n",
    "            if len(txt) >= len(summaries):\n",
    "                break\n",
    "\n",
    "        y_pred_thresh.sort()\n",
    "\n",
    "        txt = \"\"\n",
    "\n",
    "        for i in y_pred_thresh:#range(min(len(doc), y_pred.shape[0])):\n",
    "            txt = txt + \". \" + doc[i]\n",
    "\n",
    "        n = min(len(txt), len(summaries))\n",
    "\n",
    "        while n < len(txt) and txt[n].isalnum():\n",
    "            n += 1\n",
    "\n",
    "        txt = txt[:n]\n",
    "\n",
    "        #assert len(txt) - len(summaries) <= 20\n",
    "\n",
    "        scores = scorer.score(summaries, txt)\n",
    "        arr_rouge1.append(scores[\"rouge1\"].fmeasure)\n",
    "        arr_rouge2.append(scores[\"rouge2\"].fmeasure)\n",
    "        arr_rougeL.append(scores[\"rougeL\"].fmeasure)\n",
    "\n",
    "        mae.append(mae_fn(probs_tensor, targets).item())\n",
    "\n",
    "        accuracy.append(accuracy_nb_sent_per_doc_fn(probs=probs, targets=targets.cpu().detach().numpy(), doc_lens=[len(probs)], average_number_of_sentences_per_document=average_number_of_sentences_per_document))\n",
    "        #accuracy.append(accuracy_prop_sent_per_doc_fn(probs=probs, targets=targets.cpu().detach().numpy(), doc_lens=[len(probs)], average_proportion_of_sentences_per_document=average_proportion_of_sentences_per_document))\n",
    "\n",
    "        tepoch.set_postfix(accuracy = np.mean(accuracy), mae = np.mean(mae), rouge1 = np.mean(arr_rouge1), rouge2 = np.mean(arr_rouge2), rougeL = np.mean(arr_rougeL))\n",
    "        idx += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "5c1fd15f",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_metrics = {}\n",
    "test_metrics[\"accuracy\"] = np.mean(accuracy)\n",
    "test_metrics[\"mae\"]      = np.mean(mae)\n",
    "test_metrics[\"rouge1\"]   = np.mean(arr_rouge1)\n",
    "test_metrics[\"rouge2\"]   = np.mean(arr_rouge2)\n",
    "test_metrics[\"rougeL\"]   = np.mean(arr_rougeL)\n",
    "\n",
    "# Save to file in JSON format\n",
    "\n",
    "with open(checkpoints_folder + \"/test_metrics.json\", 'w') as fp:\n",
    "    json.dump(test_metrics, fp)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "496646f2",
   "metadata": {},
   "source": [
    "### Lead-3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "4c96cdd0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Lead-3: 100%|██████████| 100/100 [00:00<00:00, 211.39batch/s, rouge1=0.4, rouge2=0.165, rougeL=0.249] \n"
     ]
    }
   ],
   "source": [
    "arr_rouge1 = []\n",
    "arr_rouge2 = []\n",
    "arr_rougeL = []\n",
    "mae = []\n",
    "accuracy = []\n",
    "\n",
    "idx = 0\n",
    "with tqdm(test_iter, unit=\"batch\", total=len(test_iter)) as tepoch:\n",
    "    for batch in tepoch:\n",
    "        tepoch.set_description(\"Lead-3\")\n",
    "        doc = test_set[\"doc_splited\"].iloc[idx]\n",
    "\n",
    "        txt = \"\"\n",
    "\n",
    "        for i in range(min(len(doc), 3)):\n",
    "            txt = txt + doc[i]\n",
    "\n",
    "        summaries = test_set[\"highlights\"].iloc[idx]\n",
    "\n",
    "        n = min(len(txt), len(summaries))\n",
    "\n",
    "        while n < len(txt) and txt[n].isalnum():\n",
    "            n += 1\n",
    "\n",
    "        txt = txt[:n]\n",
    "\n",
    "        scores = scorer.score(summaries, txt)\n",
    "        arr_rouge1.append(scores[\"rouge1\"].fmeasure)\n",
    "        arr_rouge2.append(scores[\"rouge2\"].fmeasure)\n",
    "        arr_rougeL.append(scores[\"rougeL\"].fmeasure)\n",
    "\n",
    "        idx += 1\n",
    "\n",
    "        tepoch.set_postfix(rouge1 = np.mean(arr_rouge1), rouge2 = np.mean(arr_rouge2), rougeL = np.mean(arr_rougeL))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78a3c5d9",
   "metadata": {},
   "source": [
    "### First n char"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "a4c96c70",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "First-n-char': 100%|██████████| 100/100 [00:00<00:00, 193.70batch/s, rouge1=0.406, rouge2=0.168, rougeL=0.252]\n"
     ]
    }
   ],
   "source": [
    "arr_rouge1 = []\n",
    "arr_rouge2 = []\n",
    "arr_rougeL = []\n",
    "mae = []\n",
    "accuracy = []\n",
    "\n",
    "idx = 0\n",
    "with tqdm(test_iter, unit=\"batch\", total=len(test_iter)) as tepoch:\n",
    "    for batch in tepoch:\n",
    "        tepoch.set_description(\"First-n-char'\")\n",
    "        doc = test_set[\"doc_splited\"].iloc[idx]\n",
    "\n",
    "        txt = \"\"\n",
    "\n",
    "        for i in range(len(doc)):\n",
    "            txt = txt + doc[i]\n",
    "\n",
    "        summaries = test_set[\"highlights\"].iloc[idx]\n",
    "\n",
    "        n = min(len(txt), len(summaries))\n",
    "\n",
    "        while n < len(txt) and txt[n].isalnum():\n",
    "            n += 1\n",
    "\n",
    "        txt = txt[:n]\n",
    "\n",
    "        scores = scorer.score(summaries, txt)\n",
    "        arr_rouge1.append(scores[\"rouge1\"].fmeasure)\n",
    "        arr_rouge2.append(scores[\"rouge2\"].fmeasure)\n",
    "        arr_rougeL.append(scores[\"rougeL\"].fmeasure)\n",
    "\n",
    "        idx += 1\n",
    "\n",
    "        tepoch.set_postfix(rouge1 = np.mean(arr_rouge1), rouge2 = np.mean(arr_rouge2), rougeL = np.mean(arr_rougeL))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "e4f4796c",
   "metadata": {},
   "outputs": [
    {
     "ename": "SystemExit",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "An exception has occurred, use %tb to see the full traceback.\n",
      "\u001b[0;31mSystemExit\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/raoufdine/anaconda3/lib/python3.9/site-packages/IPython/core/interactiveshell.py:3465: UserWarning: To exit: use 'exit', 'quit', or Ctrl-D.\n",
      "  warn(\"To exit: use 'exit', 'quit', or Ctrl-D.\", stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "# Does not excute this cell if you want to execute the following cells.\n",
    "import sys\n",
    "sys.exit()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "vscode": {
   "interpreter": {
    "hash": "4f810f0bde7d7b5dc90f9e6832822547e13e234c53731a97fc91254323942f40"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}