Model backlog/Train/15-tweet-train-distilbert-base-uncased-refactor.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_kg_hide-input": true,
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [],
   "source": [
    "from tweet_utility_scripts import *\n",
    "from transformers import TFDistilBertModel, DistilBertConfig\n",
    "from tokenizers import BertWordPieceTokenizer\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras import optimizers, metrics, losses\n",
    "from tensorflow.keras.callbacks import EarlyStopping, TensorBoard\n",
    "from tensorflow.keras.layers import Dense, Input, Dropout, GlobalAveragePooling1D, GlobalMaxPooling1D, Concatenate\n",
    "\n",
    "\n",
    "SEED = 0\n",
    "seed_everything(SEED)\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_kg_hide-input": true,
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>textID</th>\n",
       "      <th>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>text_len</th>\n",
       "      <th>selected_text_len</th>\n",
       "      <th>set</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a5ca70509c</td>\n",
       "      <td>cant stop playin` in my head  -- pussycat doll...</td>\n",
       "      <td>cant stop playin` in my head  -- pussycat doll...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>95</td>\n",
       "      <td>95</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>f18b75e863</td>\n",
       "      <td>i hate you</td>\n",
       "      <td>i hate you</td>\n",
       "      <td>negative</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>649e31adcc</td>\n",
       "      <td>starbucks i`m lovin` it</td>\n",
       "      <td>starbucks i`m lovin` it</td>\n",
       "      <td>positive</td>\n",
       "      <td>23</td>\n",
       "      <td>23</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8891d08a8c</td>\n",
       "      <td>ben and jerry...yummmmy!!!</td>\n",
       "      <td>.yummmmy!</td>\n",
       "      <td>positive</td>\n",
       "      <td>27</td>\n",
       "      <td>9</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7fb24b4a56</td>\n",
       "      <td>wow.. purple leopard skin.  fieeerrceee..</td>\n",
       "      <td>wow.. purple leopard skin.  fieeerrceee..</td>\n",
       "      <td>neutral</td>\n",
       "      <td>42</td>\n",
       "      <td>41</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       textID                                               text  \\\n",
       "0  a5ca70509c  cant stop playin` in my head  -- pussycat doll...   \n",
       "1  f18b75e863                                         i hate you   \n",
       "2  649e31adcc                            starbucks i`m lovin` it   \n",
       "3  8891d08a8c                         ben and jerry...yummmmy!!!   \n",
       "4  7fb24b4a56          wow.. purple leopard skin.  fieeerrceee..   \n",
       "\n",
       "                                       selected_text sentiment  text_len  \\\n",
       "0  cant stop playin` in my head  -- pussycat doll...   neutral        95   \n",
       "1                                         i hate you  negative        11   \n",
       "2                            starbucks i`m lovin` it  positive        23   \n",
       "3                                          .yummmmy!  positive        27   \n",
       "4          wow.. purple leopard skin.  fieeerrceee..   neutral        42   \n",
       "\n",
       "   selected_text_len    set  \n",
       "0                 95  train  \n",
       "1                 10  train  \n",
       "2                 23  train  \n",
       "3                  9  train  \n",
       "4                 41  train  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hold_out/\r\n",
      "hold_out/y_valid.npy\r\n",
      "hold_out/x_train.npy\r\n",
      "hold_out/y_train.npy\r\n",
      "hold_out/x_valid.npy\r\n"
     ]
    }
   ],
   "source": [
    "database_base_path = '/kaggle/input/tweet-dataset-split-distilbert-uncased-128/'\n",
    "hold_out = pd.read_csv(database_base_path + 'hold-out.csv')\n",
    "train = hold_out[hold_out['set'] == 'train']\n",
    "validation = hold_out[hold_out['set'] == 'validation']\n",
    "display(hold_out.head())\n",
    "\n",
    "# Unzip files\n",
    "!tar -xvf /kaggle/input/tweet-dataset-split-distilbert-uncased-128/hold_out.tar.gz\n",
    "\n",
    "base_data_path = 'hold_out/'\n",
    "x_train = np.load(base_data_path + 'x_train.npy')\n",
    "y_train = np.load(base_data_path + 'y_train.npy')\n",
    "x_valid = np.load(base_data_path + 'x_valid.npy')\n",
    "y_valid = np.load(base_data_path + 'y_valid.npy')\n",
    "\n",
    "# Delete data dir\n",
    "shutil.rmtree(base_data_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "MAX_LEN = 128\n",
    "BATCH_SIZE = 64\n",
    "EPOCHS = 20\n",
    "LEARNING_RATE = 1e-5\n",
    "ES_PATIENCE = 3\n",
    "question_size = 3\n",
    "\n",
    "tokenizer_path = database_base_path + 'vocab.txt'\n",
    "base_path = '/kaggle/input/qa-transformers/distilbert/'\n",
    "base_model_path = base_path + 'distilbert-base-uncased-distilled-squad-tf_model.h5'\n",
    "config_path = base_path + 'distilbert-base-uncased-distilled-squad-config.json'\n",
    "model_path = 'model.h5'\n",
    "log_path = './'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "attention_mask (InputLayer)     [(None, 128)]        0                                            \n",
      "__________________________________________________________________________________________________\n",
      "input_ids (InputLayer)          [(None, 128)]        0                                            \n",
      "__________________________________________________________________________________________________\n",
      "token_type_ids (InputLayer)     [(None, 128)]        0                                            \n",
      "__________________________________________________________________________________________________\n",
      "base_model (TFDistilBertModel)  ((None, 128, 768),)  66362880    attention_mask[0][0]             \n",
      "                                                                 input_ids[0][0]                  \n",
      "                                                                 token_type_ids[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "global_average_pooling1d (Globa (None, 768)          0           base_model[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "y_start (Dense)                 (None, 128)          98432       global_average_pooling1d[0][0]   \n",
      "__________________________________________________________________________________________________\n",
      "y_end (Dense)                   (None, 128)          98432       global_average_pooling1d[0][0]   \n",
      "==================================================================================================\n",
      "Total params: 66,559,744\n",
      "Trainable params: 66,559,744\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "module_config = DistilBertConfig.from_pretrained(config_path, output_hidden_states=False)\n",
    "\n",
    "def model_fn():\n",
    "    input_ids = Input(shape=(MAX_LEN,), dtype=tf.int32, name='input_ids')\n",
    "    attention_mask = Input(shape=(MAX_LEN,), dtype=tf.int32, name='attention_mask')\n",
    "    token_type_ids = Input(shape=(MAX_LEN,), dtype=tf.int32, name='token_type_ids')\n",
    "    \n",
    "    base_model = TFDistilBertModel.from_pretrained(base_model_path, config=module_config, name=\"base_model\")\n",
    "    sequence_output = base_model({'input_ids': input_ids, 'attention_mask': attention_mask, 'token_type_ids': token_type_ids})\n",
    "    last_state = sequence_output[0]\n",
    "    \n",
    "    x = GlobalAveragePooling1D()(last_state)\n",
    "    \n",
    "    y_start = Dense(MAX_LEN, activation='softmax', name='y_start')(x)\n",
    "    y_end = Dense(MAX_LEN, activation='softmax', name='y_end')(x)\n",
    "    \n",
    "    model = Model(inputs=[input_ids, attention_mask, token_type_ids], outputs=[y_start, y_end])\n",
    "    model.compile(optimizers.Adam(lr=LEARNING_RATE), \n",
    "                  loss=losses.CategoricalCrossentropy(), \n",
    "                  metrics=[metrics.CategoricalAccuracy()])\n",
    "    \n",
    "    return model\n",
    "\n",
    "model = model_fn()\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_kg_hide-input": false,
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 21984 samples, validate on 5496 samples\n",
      "Epoch 1/20\n",
      "21984/21984 - 181s - loss: 5.0825 - y_start_loss: 1.8813 - y_end_loss: 3.2012 - y_start_categorical_accuracy: 0.5736 - y_end_categorical_accuracy: 0.1187 - val_loss: 4.1048 - val_y_start_loss: 1.6303 - val_y_end_loss: 2.4740 - val_y_start_categorical_accuracy: 0.5806 - val_y_end_categorical_accuracy: 0.2691\n",
      "Epoch 2/20\n",
      "21984/21984 - 173s - loss: 3.6037 - y_start_loss: 1.4966 - y_end_loss: 2.1071 - y_start_categorical_accuracy: 0.5829 - y_end_categorical_accuracy: 0.3971 - val_loss: 3.1474 - val_y_start_loss: 1.3896 - val_y_end_loss: 1.7571 - val_y_start_categorical_accuracy: 0.5986 - val_y_end_categorical_accuracy: 0.5055\n",
      "Epoch 3/20\n",
      "21984/21984 - 172s - loss: 2.9200 - y_start_loss: 1.3199 - y_end_loss: 1.6001 - y_start_categorical_accuracy: 0.6007 - y_end_categorical_accuracy: 0.5460 - val_loss: 2.7949 - val_y_start_loss: 1.3209 - val_y_end_loss: 1.4728 - val_y_start_categorical_accuracy: 0.6028 - val_y_end_categorical_accuracy: 0.5979\n",
      "Epoch 4/20\n",
      "21984/21984 - 172s - loss: 2.5246 - y_start_loss: 1.2042 - y_end_loss: 1.3204 - y_start_categorical_accuracy: 0.6232 - y_end_categorical_accuracy: 0.6240 - val_loss: 2.6348 - val_y_start_loss: 1.2850 - val_y_end_loss: 1.3489 - val_y_start_categorical_accuracy: 0.6050 - val_y_end_categorical_accuracy: 0.6359\n",
      "Epoch 5/20\n",
      "21984/21984 - 172s - loss: 2.2882 - y_start_loss: 1.1118 - y_end_loss: 1.1765 - y_start_categorical_accuracy: 0.6447 - y_end_categorical_accuracy: 0.6563 - val_loss: 2.5574 - val_y_start_loss: 1.2449 - val_y_end_loss: 1.3115 - val_y_start_categorical_accuracy: 0.6225 - val_y_end_categorical_accuracy: 0.6339\n",
      "Epoch 6/20\n",
      "21984/21984 - 172s - loss: 2.1073 - y_start_loss: 1.0372 - y_end_loss: 1.0701 - y_start_categorical_accuracy: 0.6613 - y_end_categorical_accuracy: 0.6803 - val_loss: 2.5731 - val_y_start_loss: 1.2582 - val_y_end_loss: 1.3137 - val_y_start_categorical_accuracy: 0.6206 - val_y_end_categorical_accuracy: 0.6401\n",
      "Epoch 7/20\n",
      "21984/21984 - 172s - loss: 1.9550 - y_start_loss: 0.9667 - y_end_loss: 0.9883 - y_start_categorical_accuracy: 0.6840 - y_end_categorical_accuracy: 0.7025 - val_loss: 2.5647 - val_y_start_loss: 1.2526 - val_y_end_loss: 1.3110 - val_y_start_categorical_accuracy: 0.6172 - val_y_end_categorical_accuracy: 0.6434\n",
      "Epoch 8/20\n",
      "Restoring model weights from the end of the best epoch.\n",
      "21984/21984 - 172s - loss: 1.8045 - y_start_loss: 0.8967 - y_end_loss: 0.9079 - y_start_categorical_accuracy: 0.7041 - y_end_categorical_accuracy: 0.7225 - val_loss: 2.5900 - val_y_start_loss: 1.2683 - val_y_end_loss: 1.3206 - val_y_start_categorical_accuracy: 0.6163 - val_y_end_categorical_accuracy: 0.6363\n",
      "Epoch 00008: early stopping\n"
     ]
    }
   ],
   "source": [
    "tb_callback = TensorBoard(log_dir=log_path)\n",
    "es = EarlyStopping(monitor='val_loss', mode='min', patience=ES_PATIENCE, \n",
    "                   restore_best_weights=True, verbose=1)\n",
    "\n",
    "history = model.fit(list(x_train), list(y_train),\n",
    "                    validation_data=(list(x_valid), list(y_valid)),\n",
    "                    callbacks=[es, tb_callback],\n",
    "                    epochs=EPOCHS, \n",
    "                    verbose=2).history\n",
    "\n",
    "model.save_weights(model_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "_kg_hide-input": true,
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train/\r\n",
      "train/events.out.tfevents.1586484101.7f1dde097818.13.5413.v2\r\n",
      "train/plugins/\r\n",
      "train/plugins/profile/\r\n",
      "train/plugins/profile/2020-04-10_02-01-48/\r\n",
      "train/plugins/profile/2020-04-10_02-01-48/local.trace\r\n",
      "train/events.out.tfevents.1586484108.7f1dde097818.profile-empty\r\n",
      "validation/\r\n",
      "validation/events.out.tfevents.1586484282.7f1dde097818.13.24606.v2\r\n"
     ]
    }
   ],
   "source": [
    "# Compress logs dir\n",
    "!tar -cvzf train.tar.gz train\n",
    "!tar -cvzf validation.tar.gz validation\n",
    "\n",
    "# Delete logs dir\n",
    "if os.path.exists('/kaggle/working/train/'):\n",
    "    shutil.rmtree('/kaggle/working/train/')\n",
    "if os.path.exists('/kaggle/working/validation/'):\n",
    "    shutil.rmtree('/kaggle/working/validation/')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model loss graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1800 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"whitegrid\")\n",
    "plot_metrics(history, metric_list=['loss', 'y_start_loss', 'y_end_loss', 'y_start_categorical_accuracy', 'y_end_categorical_accuracy'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenizer = BertWordPieceTokenizer(tokenizer_path , lowercase=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Train</th>\n",
       "      <th>Validation</th>\n",
       "      <th>Train samples</th>\n",
       "      <th>Validation samples</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Metric/Jaccard</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Overall</th>\n",
       "      <td>0.705218</td>\n",
       "      <td>0.645261</td>\n",
       "      <td>21984</td>\n",
       "      <td>5496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sentiment neutral</th>\n",
       "      <td>0.975719</td>\n",
       "      <td>0.971899</td>\n",
       "      <td>8842</td>\n",
       "      <td>2275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sentiment negative</th>\n",
       "      <td>0.517952</td>\n",
       "      <td>0.426003</td>\n",
       "      <td>6258</td>\n",
       "      <td>1523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sentiment positive</th>\n",
       "      <td>0.528016</td>\n",
       "      <td>0.404289</td>\n",
       "      <td>6884</td>\n",
       "      <td>1698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0 &lt;= text &lt; 10</th>\n",
       "      <td>0.973727</td>\n",
       "      <td>0.962500</td>\n",
       "      <td>203</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10 &lt;= text &lt; 50</th>\n",
       "      <td>0.781281</td>\n",
       "      <td>0.742510</td>\n",
       "      <td>7823</td>\n",
       "      <td>1945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50 &lt;= text &lt; 100</th>\n",
       "      <td>0.674321</td>\n",
       "      <td>0.611913</td>\n",
       "      <td>8858</td>\n",
       "      <td>2201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100 &lt;= text &lt; 200</th>\n",
       "      <td>0.631521</td>\n",
       "      <td>0.547215</td>\n",
       "      <td>5100</td>\n",
       "      <td>1310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0 &lt;= selected text &lt; 10</th>\n",
       "      <td>0.510267</td>\n",
       "      <td>0.374766</td>\n",
       "      <td>6326</td>\n",
       "      <td>1492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10 &lt;= selected text &lt; 50</th>\n",
       "      <td>0.706100</td>\n",
       "      <td>0.663700</td>\n",
       "      <td>9575</td>\n",
       "      <td>2455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50 &lt;= selected text &lt; 100</th>\n",
       "      <td>0.890874</td>\n",
       "      <td>0.867331</td>\n",
       "      <td>4075</td>\n",
       "      <td>1042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100 &lt;= selected text &lt; 200</th>\n",
       "      <td>0.938419</td>\n",
       "      <td>0.895586</td>\n",
       "      <td>2008</td>\n",
       "      <td>507</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                Train  Validation  Train samples  \\\n",
       "Metric/Jaccard                                                     \n",
       "Overall                      0.705218    0.645261          21984   \n",
       "Sentiment neutral            0.975719    0.971899           8842   \n",
       "Sentiment negative           0.517952    0.426003           6258   \n",
       "Sentiment positive           0.528016    0.404289           6884   \n",
       "0 <= text < 10               0.973727    0.962500            203   \n",
       "10 <= text < 50              0.781281    0.742510           7823   \n",
       "50 <= text < 100             0.674321    0.611913           8858   \n",
       "100 <= text < 200            0.631521    0.547215           5100   \n",
       "0 <= selected text < 10      0.510267    0.374766           6326   \n",
       "10 <= selected text < 50     0.706100    0.663700           9575   \n",
       "50 <= selected text < 100    0.890874    0.867331           4075   \n",
       "100 <= selected text < 200   0.938419    0.895586           2008   \n",
       "\n",
       "                             Validation samples  \n",
       "Metric/Jaccard                                   \n",
       "Overall                                    5496  \n",
       "Sentiment neutral                          2275  \n",
       "Sentiment negative                         1523  \n",
       "Sentiment positive                         1698  \n",
       "0 <= text < 10                               40  \n",
       "10 <= text < 50                            1945  \n",
       "50 <= text < 100                           2201  \n",
       "100 <= text < 200                          1310  \n",
       "0 <= selected text < 10                    1492  \n",
       "10 <= selected text < 50                   2455  \n",
       "50 <= selected text < 100                  1042  \n",
       "100 <= selected text < 200                  507  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_preds = model.predict(list(x_train))\n",
    "valid_preds = model.predict(list(x_valid))\n",
    "\n",
    "train['start'] = train_preds[0].argmax(axis=-1)\n",
    "train['end'] = train_preds[1].argmax(axis=-1)\n",
    "train['prediction'] = train.apply(lambda x: decode(x['start'], x['end'], x['text'], question_size, tokenizer), axis=1)\n",
    "train[\"prediction\"] = train[\"prediction\"].apply(lambda x: '.' if x.strip() == '' else x)\n",
    "\n",
    "validation['start'] = valid_preds[0].argmax(axis=-1)\n",
    "validation['end'] = valid_preds[1].argmax(axis=-1)\n",
    "validation['prediction'] = validation.apply(lambda x: decode(x['start'], x['end'], x['text'], question_size, tokenizer), axis=1)\n",
    "validation[\"prediction\"] = validation[\"prediction\"].apply(lambda x: '.' if x.strip() == '' else x)\n",
    "  \n",
    "display(evaluate_model(train, validation))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualize predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train set\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>textID</th>\n",
       "      <th>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>text_len</th>\n",
       "      <th>selected_text_len</th>\n",
       "      <th>set</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>prediction</th>\n",
       "      <th>jaccard</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a5ca70509c</td>\n",
       "      <td>cant stop playin` in my head  -- pussycat doll...</td>\n",
       "      <td>cant stop playin` in my head  -- pussycat doll...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>95</td>\n",
       "      <td>95</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>39</td>\n",
       "      <td>cant stop playin` in my head  -- pussycat doll...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>f18b75e863</td>\n",
       "      <td>i hate you</td>\n",
       "      <td>i hate you</td>\n",
       "      <td>negative</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>i hate you</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>649e31adcc</td>\n",
       "      <td>starbucks i`m lovin` it</td>\n",
       "      <td>starbucks i`m lovin` it</td>\n",
       "      <td>positive</td>\n",
       "      <td>23</td>\n",
       "      <td>23</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>11</td>\n",
       "      <td>starbucks i`m lovin` it</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8891d08a8c</td>\n",
       "      <td>ben and jerry...yummmmy!!!</td>\n",
       "      <td>.yummmmy!</td>\n",
       "      <td>positive</td>\n",
       "      <td>27</td>\n",
       "      <td>9</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>.yumm</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7fb24b4a56</td>\n",
       "      <td>wow.. purple leopard skin.  fieeerrceee..</td>\n",
       "      <td>wow.. purple leopard skin.  fieeerrceee..</td>\n",
       "      <td>neutral</td>\n",
       "      <td>42</td>\n",
       "      <td>41</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>17</td>\n",
       "      <td>wow.. purple leopard skin.  fieeerrceee..</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>cb280b3adb</td>\n",
       "      <td>_ kk`s  have fun eating lol, remember shaun lo...</td>\n",
       "      <td>fun</td>\n",
       "      <td>positive</td>\n",
       "      <td>58</td>\n",
       "      <td>3</td>\n",
       "      <td>train</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>have fun</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>8dd41515ff</td>\n",
       "      <td>definitley miley</td>\n",
       "      <td>definitley miley</td>\n",
       "      <td>neutral</td>\n",
       "      <td>17</td>\n",
       "      <td>16</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>definitley miley</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>60fa8ea66e</td>\n",
       "      <td>just going to shower. brb in 10 mins or so</td>\n",
       "      <td>just going to shower. brb in 10 mins or so</td>\n",
       "      <td>neutral</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>16</td>\n",
       "      <td>just going to shower. brb in 10 mins or so</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>bb4ac6eb31</td>\n",
       "      <td>back to work.. with no distractions..</td>\n",
       "      <td>back to work.. with no distractions..</td>\n",
       "      <td>positive</td>\n",
       "      <td>37</td>\n",
       "      <td>37</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>14</td>\n",
       "      <td>back to work.. with no distractions..</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2f858c20c3</td>\n",
       "      <td>cool, i used to live 5 houses to the left of ...</td>\n",
       "      <td>cool,</td>\n",
       "      <td>positive</td>\n",
       "      <td>86</td>\n",
       "      <td>5</td>\n",
       "      <td>train</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>cool,</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       textID                                               text  \\\n",
       "0  a5ca70509c  cant stop playin` in my head  -- pussycat doll...   \n",
       "1  f18b75e863                                         i hate you   \n",
       "2  649e31adcc                            starbucks i`m lovin` it   \n",
       "3  8891d08a8c                         ben and jerry...yummmmy!!!   \n",
       "4  7fb24b4a56          wow.. purple leopard skin.  fieeerrceee..   \n",
       "5  cb280b3adb  _ kk`s  have fun eating lol, remember shaun lo...   \n",
       "6  8dd41515ff                                   definitley miley   \n",
       "7  60fa8ea66e         just going to shower. brb in 10 mins or so   \n",
       "8  bb4ac6eb31              back to work.. with no distractions..   \n",
       "9  2f858c20c3   cool, i used to live 5 houses to the left of ...   \n",
       "\n",
       "                                       selected_text sentiment  text_len  \\\n",
       "0  cant stop playin` in my head  -- pussycat doll...   neutral        95   \n",
       "1                                         i hate you  negative        11   \n",
       "2                            starbucks i`m lovin` it  positive        23   \n",
       "3                                          .yummmmy!  positive        27   \n",
       "4          wow.. purple leopard skin.  fieeerrceee..   neutral        42   \n",
       "5                                                fun  positive        58   \n",
       "6                                   definitley miley   neutral        17   \n",
       "7         just going to shower. brb in 10 mins or so   neutral        42   \n",
       "8              back to work.. with no distractions..  positive        37   \n",
       "9                                              cool,  positive        86   \n",
       "\n",
       "   selected_text_len    set  start  end  \\\n",
       "0                 95  train      4   39   \n",
       "1                 10  train      4    6   \n",
       "2                 23  train      4   11   \n",
       "3                  9  train      9   11   \n",
       "4                 41  train      4   17   \n",
       "5                  3  train      9   10   \n",
       "6                 16  train      4    8   \n",
       "7                 42  train      4   16   \n",
       "8                 37  train      4   14   \n",
       "9                  5  train      4    5   \n",
       "\n",
       "                                          prediction  jaccard  \n",
       "0  cant stop playin` in my head  -- pussycat doll...      1.0  \n",
       "1                                         i hate you      1.0  \n",
       "2                            starbucks i`m lovin` it      1.0  \n",
       "3                                              .yumm      0.0  \n",
       "4          wow.. purple leopard skin.  fieeerrceee..      1.0  \n",
       "5                                           have fun      0.5  \n",
       "6                                   definitley miley      1.0  \n",
       "7         just going to shower. brb in 10 mins or so      1.0  \n",
       "8              back to work.. with no distractions..      1.0  \n",
       "9                                              cool,      1.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validation set\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>textID</th>\n",
       "      <th>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>text_len</th>\n",
       "      <th>selected_text_len</th>\n",
       "      <th>set</th>\n",
       "      <th>start</th>\n",
       "      <th>end</th>\n",
       "      <th>prediction</th>\n",
       "      <th>jaccard</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>21984</th>\n",
       "      <td>3391184efc</td>\n",
       "      <td>loves the nice weather and 7:30 exams</td>\n",
       "      <td>loves</td>\n",
       "      <td>positive</td>\n",
       "      <td>37</td>\n",
       "      <td>5</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>loves</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21985</th>\n",
       "      <td>b35daf9677</td>\n",
       "      <td>okay, this weather isn`t 'cute sundress' frien...</td>\n",
       "      <td>okay, this weather isn`t 'cute sundress' frien...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>130</td>\n",
       "      <td>130</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>38</td>\n",
       "      <td>okay, this weather isn`t 'cute sundress' frien...</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21986</th>\n",
       "      <td>06e5249859</td>\n",
       "      <td>woo hoo!! congratulations</td>\n",
       "      <td>congratulations</td>\n",
       "      <td>positive</td>\n",
       "      <td>26</td>\n",
       "      <td>15</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>woo hoo!! congratulations</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21987</th>\n",
       "      <td>3cd4960670</td>\n",
       "      <td>thanks   got a hold of someone there who knew...</td>\n",
       "      <td>thanks</td>\n",
       "      <td>positive</td>\n",
       "      <td>68</td>\n",
       "      <td>6</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>thanks</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21988</th>\n",
       "      <td>92b75314ca</td>\n",
       "      <td>got back and putting in the laundry. we got in...</td>\n",
       "      <td>got back and putting in the laundry. we got in...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>95</td>\n",
       "      <td>95</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>29</td>\n",
       "      <td>got back and putting in the laundry. we got in...</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21989</th>\n",
       "      <td>7438c9c09a</td>\n",
       "      <td>congrats on graduating college!</td>\n",
       "      <td>congrats</td>\n",
       "      <td>positive</td>\n",
       "      <td>32</td>\n",
       "      <td>8</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>congrat</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21990</th>\n",
       "      <td>fa042d9ad5</td>\n",
       "      <td>can you get me a sub from subway when ur on yo...</td>\n",
       "      <td>can you get me a sub from subway when ur on yo...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>58</td>\n",
       "      <td>58</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>18</td>\n",
       "      <td>can you get me a sub from subway when ur on yo...</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21991</th>\n",
       "      <td>a4ede54987</td>\n",
       "      <td>it`s the weekend but 9 year old is grounded wh...</td>\n",
       "      <td>it`s the weekend but 9 year old is grounded wh...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>71</td>\n",
       "      <td>71</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>19</td>\n",
       "      <td>it`s the weekend but 9 year old is grounded wh...</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21992</th>\n",
       "      <td>35124a7481</td>\n",
       "      <td>it`s been fine for me for like, a week, but n...</td>\n",
       "      <td>it`s been fine for me for like, a week, but no...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>68</td>\n",
       "      <td>67</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>25</td>\n",
       "      <td>it`s been fine for me for like, a week, but no...</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21993</th>\n",
       "      <td>e4b6c9a295</td>\n",
       "      <td>ntah, i really wanna be there</td>\n",
       "      <td>wanna</td>\n",
       "      <td>positive</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>validation</td>\n",
       "      <td>4</td>\n",
       "      <td>11</td>\n",
       "      <td>ntah, i really wanna be there</td>\n",
       "      <td>0.166667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           textID                                               text  \\\n",
       "21984  3391184efc              loves the nice weather and 7:30 exams   \n",
       "21985  b35daf9677  okay, this weather isn`t 'cute sundress' frien...   \n",
       "21986  06e5249859                          woo hoo!! congratulations   \n",
       "21987  3cd4960670   thanks   got a hold of someone there who knew...   \n",
       "21988  92b75314ca  got back and putting in the laundry. we got in...   \n",
       "21989  7438c9c09a                    congrats on graduating college!   \n",
       "21990  fa042d9ad5  can you get me a sub from subway when ur on yo...   \n",
       "21991  a4ede54987  it`s the weekend but 9 year old is grounded wh...   \n",
       "21992  35124a7481   it`s been fine for me for like, a week, but n...   \n",
       "21993  e4b6c9a295                      ntah, i really wanna be there   \n",
       "\n",
       "                                           selected_text sentiment  text_len  \\\n",
       "21984                                              loves  positive        37   \n",
       "21985  okay, this weather isn`t 'cute sundress' frien...   neutral       130   \n",
       "21986                                    congratulations  positive        26   \n",
       "21987                                             thanks  positive        68   \n",
       "21988  got back and putting in the laundry. we got in...   neutral        95   \n",
       "21989                                           congrats  positive        32   \n",
       "21990  can you get me a sub from subway when ur on yo...   neutral        58   \n",
       "21991  it`s the weekend but 9 year old is grounded wh...   neutral        71   \n",
       "21992  it`s been fine for me for like, a week, but no...   neutral        68   \n",
       "21993                                              wanna  positive        30   \n",
       "\n",
       "       selected_text_len         set  start  end  \\\n",
       "21984                  5  validation      4    4   \n",
       "21985                130  validation      4   38   \n",
       "21986                 15  validation      4    9   \n",
       "21987                  6  validation      4    4   \n",
       "21988                 95  validation      4   29   \n",
       "21989                  8  validation      4    5   \n",
       "21990                 58  validation      4   18   \n",
       "21991                 71  validation      4   19   \n",
       "21992                 67  validation      4   25   \n",
       "21993                  5  validation      4   11   \n",
       "\n",
       "                                              prediction   jaccard  \n",
       "21984                                              loves  1.000000  \n",
       "21985  okay, this weather isn`t 'cute sundress' frien...  1.000000  \n",
       "21986                          woo hoo!! congratulations  0.333333  \n",
       "21987                                             thanks  1.000000  \n",
       "21988  got back and putting in the laundry. we got in...  1.000000  \n",
       "21989                                            congrat  0.000000  \n",
       "21990  can you get me a sub from subway when ur on yo...  1.000000  \n",
       "21991  it`s the weekend but 9 year old is grounded wh...  1.000000  \n",
       "21992  it`s been fine for me for like, a week, but no...  1.000000  \n",
       "21993                      ntah, i really wanna be there  0.166667  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Train set')\n",
    "display(train.head(10))\n",
    "\n",
    "print('Validation set')\n",
    "display(validation.head(10))"
   ]
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}