perpetrator.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import and clean the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 9914 entries, 0 to 9913\n",
      "Data columns (total 2 columns):\n",
      " #   Column            Non-Null Count  Dtype \n",
      "---  ------            --------------  ----- \n",
      " 0   perpetrator       9914 non-null   object\n",
      " 1   incident_summary  9914 non-null   object\n",
      "dtypes: object(2)\n",
      "memory usage: 155.0+ KB\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>perpetrator</th>\n",
       "      <th>incident_summary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Security</td>\n",
       "      <td>An alleged arms supplier to the Communist Part...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Maoist</td>\n",
       "      <td>A Kamareddy dalam (squad) member belonging to ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Security</td>\n",
       "      <td>Senior CPI-Maoist 'Polit Bureau' and 'central ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Maoist</td>\n",
       "      <td>A TDP leader and former Sarpanch of Jerrela Gr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Maoist</td>\n",
       "      <td>The CPI-Maoist cadres blasted coffee pulping u...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  perpetrator                                   incident_summary\n",
       "0    Security  An alleged arms supplier to the Communist Part...\n",
       "1      Maoist  A Kamareddy dalam (squad) member belonging to ...\n",
       "2    Security  Senior CPI-Maoist 'Polit Bureau' and 'central ...\n",
       "3      Maoist  A TDP leader and former Sarpanch of Jerrela Gr...\n",
       "4      Maoist  The CPI-Maoist cadres blasted coffee pulping u..."
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import pandas\n",
    "import pandas as pd\n",
    "\n",
    "# import data\n",
    "perp_df = pd.read_csv('data/perpetrator.csv')\n",
    "\n",
    "# view dataset info\n",
    "perp_df.info()\n",
    "perp_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Encode labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 9914 entries, 0 to 9913\n",
      "Data columns (total 3 columns):\n",
      " #   Column            Non-Null Count  Dtype \n",
      "---  ------            --------------  ----- \n",
      " 0   perpetrator       9914 non-null   object\n",
      " 1   incident_summary  9914 non-null   object\n",
      " 2   labels            9914 non-null   int64 \n",
      "dtypes: int64(1), object(2)\n",
      "memory usage: 232.5+ KB\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>perpetrator</th>\n",
       "      <th>incident_summary</th>\n",
       "      <th>labels</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Security</td>\n",
       "      <td>An alleged arms supplier to the Communist Part...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Maoist</td>\n",
       "      <td>A Kamareddy dalam (squad) member belonging to ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Security</td>\n",
       "      <td>Senior CPI-Maoist 'Polit Bureau' and 'central ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Maoist</td>\n",
       "      <td>A TDP leader and former Sarpanch of Jerrela Gr...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Maoist</td>\n",
       "      <td>The CPI-Maoist cadres blasted coffee pulping u...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  perpetrator                                   incident_summary  labels\n",
       "0    Security  An alleged arms supplier to the Communist Part...       1\n",
       "1      Maoist  A Kamareddy dalam (squad) member belonging to ...       0\n",
       "2    Security  Senior CPI-Maoist 'Polit Bureau' and 'central ...       1\n",
       "3      Maoist  A TDP leader and former Sarpanch of Jerrela Gr...       0\n",
       "4      Maoist  The CPI-Maoist cadres blasted coffee pulping u...       0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import label encoder\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "# Initialize the label encoder\n",
    "label_encoder = LabelEncoder()\n",
    "\n",
    "# Fit the encoder to the labels and transform them into numeric ids\n",
    "perp_df['labels'] = label_encoder.fit_transform(perp_df['perpetrator'])\n",
    "\n",
    "# view dataset info\n",
    "perp_df.info()\n",
    "perp_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 2])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# explore labels\n",
    "perp_df['labels'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>perpetrator</th>\n",
       "      <th>incident_summary</th>\n",
       "      <th>labels</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>Andhra Pradesh Police killed two CPI-Maoist ca...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>The Nalgonda District Police killed four CPI-M...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>A former Maoist female cadre, Veeramalla Pushp...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>A senior cadre and District committee secretar...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Unknown</td>\n",
       "      <td>Two CPI-Maoist cadres were killed in an encoun...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   perpetrator                                   incident_summary  labels\n",
       "15     Unknown  Andhra Pradesh Police killed two CPI-Maoist ca...       2\n",
       "19     Unknown  The Nalgonda District Police killed four CPI-M...       2\n",
       "21     Unknown  A former Maoist female cadre, Veeramalla Pushp...       2\n",
       "24     Unknown  A senior cadre and District committee secretar...       2\n",
       "29     Unknown  Two CPI-Maoist cadres were killed in an encoun...       2"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make sure \"unknown\" category is right\n",
    "perp_df.loc[perp_df['labels'] == 2].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Apply train-test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# separate text summaries and labels into X and y\n",
    "X = perp_df.drop(['labels'], axis=1)  # text summaries\n",
    "y = perp_df['labels']  # labels\n",
    "\n",
    "# separate the test set\n",
    "X_train_val, X_test, y_train_val, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)\n",
    "\n",
    "# separate the validation set from the training set\n",
    "X_train, X_val, y_train, y_val = train_test_split(X_train_val, y_train_val, test_size=0.25, random_state=42, stratify=y_train_val)\n",
    "\n",
    "# Note: Adjusting `test_size` in the second split to 0.25 means that\n",
    "# 25% of the training+validation set becomes the validation set,\n",
    "# and 75% remains as the training set.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Recombine data frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Recombine the features and labels into pandas DataFrames\n",
    "train_df = pd.concat([X_train, y_train], axis=1)\n",
    "val_df = pd.concat([X_val, y_val], axis=1)\n",
    "test_df = pd.concat([X_test, y_test], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Label distribution in training:\n",
      "labels\n",
      "0    0.523537\n",
      "1    0.429892\n",
      "2    0.046570\n",
      "Name: proportion, dtype: float64\n",
      "\n",
      "\n",
      "Label distribution in validation:\n",
      "labels\n",
      "0    0.523449\n",
      "1    0.430156\n",
      "2    0.046394\n",
      "Name: proportion, dtype: float64\n",
      "\n",
      "\n",
      "Label distribution in test:\n",
      "labels\n",
      "0    0.523449\n",
      "1    0.430156\n",
      "2    0.046394\n",
      "Name: proportion, dtype: float64\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# see if labels are distributed properly across three datasets \n",
    "\n",
    "# Function to calculate and print label distribution\n",
    "def print_label_distribution(df, dataset_name):\n",
    "    label_counts = df['labels'].value_counts(normalize=True)  # Get normalized counts\n",
    "    print(f\"Label distribution in {dataset_name}:\")\n",
    "    print(label_counts)\n",
    "    print(\"\\n\")\n",
    "\n",
    "# Calculate and print the label distribution for each dataset\n",
    "print_label_distribution(train_df, \"training\")\n",
    "print_label_distribution(val_df, \"validation\")\n",
    "print_label_distribution(test_df, \"test\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tokenize textual data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ejt/anaconda3/envs/code-satp/lib/python3.9/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n",
      "Map: 100%|██████████| 5948/5948 [00:00<00:00, 9036.68 examples/s] \n",
      "Map: 100%|██████████| 1983/1983 [00:00<00:00, 11219.68 examples/s]\n",
      "Map: 100%|██████████| 1983/1983 [00:00<00:00, 10596.41 examples/s]\n"
     ]
    }
   ],
   "source": [
    "from transformers import AutoTokenizer\n",
    "from datasets import Dataset\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained('distilbert-base-uncased')\n",
    "\n",
    "def tokenize_function(examples):\n",
    "    return tokenizer(examples['incident_summary'], padding=\"max_length\", truncation=True)\n",
    "\n",
    "train_dataset = Dataset.from_pandas(train_df).map(tokenize_function, batched=True)\n",
    "val_dataset = Dataset.from_pandas(val_df).map(tokenize_function, batched=True)\n",
    "test_dataset = Dataset.from_pandas(test_df).map(tokenize_function, batched=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Inspect training, validation and test datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'perpetrator': 'Security', 'incident_summary': 'Aurangabad Police of Bihar arrested three wanted CPI-Maoist cadres from different localities under Daudnagar Sub-Division of the same district. Daudnagar SDPO Sanjay Kumar said on the basis of intelligence inputs, a Police team from Goh, Khudowa and Uphara Police Stations carried out raids at several hideouts and arrested three Maoists who have been identified as Sambhu Yadav, Dharmdeo Yadav and Raja Uadav.', 'labels': 1, '__index_level_0__': 1291, 'input_ids': [101, 15240, 13807, 9024, 2610, 1997, 16178, 4727, 2093, 2359, 28780, 1011, 15158, 2923, 28353, 6072, 2013, 2367, 19664, 2104, 4830, 6784, 14346, 4942, 1011, 2407, 1997, 1996, 2168, 2212, 1012, 4830, 6784, 14346, 17371, 6873, 29590, 9600, 2056, 2006, 1996, 3978, 1997, 4454, 20407, 1010, 1037, 2610, 2136, 2013, 2175, 2232, 1010, 1047, 6979, 3527, 4213, 1998, 2039, 11077, 2610, 3703, 3344, 2041, 11217, 2012, 2195, 29588, 2015, 1998, 4727, 2093, 15158, 5130, 2040, 2031, 2042, 4453, 2004, 3520, 23706, 2226, 8038, 29045, 1010, 28144, 27292, 3207, 2080, 8038, 29045, 1998, 10164, 25423, 29045, 1012, 102, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]}\n"
     ]
    }
   ],
   "source": [
    "# Inspect an example of the training set\n",
    "import torch\n",
    "print(train_dataset[1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'perpetrator': 'Maoist', 'incident_summary': 'CPI-Maoist cadres killed two persons, Gopiram Koreti and Rajendra Koreti, in Khargaon village in Rajnandgaon District. The victims were abducted from their village earlier this week.', 'labels': 0, '__index_level_0__': 4190, 'input_ids': [101, 28780, 1011, 15158, 2923, 28353, 6072, 2730, 2048, 5381, 1010, 2175, 8197, 6444, 12849, 13465, 2072, 1998, 11948, 19524, 12849, 13465, 2072, 1010, 1999, 1047, 8167, 27073, 2352, 1999, 11948, 7229, 2094, 27073, 2212, 1012, 1996, 5694, 2020, 20361, 2013, 2037, 2352, 3041, 2023, 2733, 1012, 102, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]}\n"
     ]
    }
   ],
   "source": [
    "# Inspect an example of the validation set\n",
    "print(val_dataset[1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'perpetrator': 'Security', 'incident_summary': \"Ravindra Kumar, a BJP member in Koderma District, arrested for having links with LWEs. Ravindra Kumar alias Raju Mehta, is the proprietor of 'Sarang Explosive' and said to be an active member of the BJP at Phulwaria of Domchanch in Koderma District\", 'labels': 1, '__index_level_0__': 6069, 'input_ids': [101, 16806, 17670, 9600, 1010, 1037, 24954, 2266, 1999, 12849, 4063, 2863, 2212, 1010, 4727, 2005, 2383, 6971, 2007, 1048, 18192, 1012, 16806, 17670, 9600, 14593, 25098, 2033, 22893, 1010, 2003, 1996, 21584, 1997, 1005, 7354, 3070, 11355, 1005, 1998, 2056, 2000, 2022, 2019, 3161, 2266, 1997, 1996, 24954, 2012, 6887, 5313, 20031, 2050, 1997, 14383, 14856, 2818, 1999, 12849, 4063, 2863, 2212, 102, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]}\n"
     ]
    }
   ],
   "source": [
    "# Inspect an example of the test set\n",
    "print(test_dataset[1])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of DistilBertForSequenceClassification were not initialized from the model checkpoint at distilbert-base-uncased and are newly initialized: ['classifier.bias', 'classifier.weight', 'pre_classifier.bias', 'pre_classifier.weight']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
      "/Users/ejt/anaconda3/envs/code-satp/lib/python3.9/site-packages/accelerate/accelerator.py:432: FutureWarning: Passing the following arguments to `Accelerator` is deprecated and will be removed in version 1.0 of Accelerate: dict_keys(['dispatch_batches', 'split_batches', 'even_batches', 'use_seedable_sampler']). Please pass an `accelerate.DataLoaderConfiguration` instead: \n",
      "dataloader_config = DataLoaderConfiguration(dispatch_batches=None, split_batches=False, even_batches=True, use_seedable_sampler=True)\n",
      "  warnings.warn(\n",
      "  1%|          | 53/4461 [00:20<25:35,  2.87it/s] "
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb Cell 19\u001b[0m line \u001b[0;36m4\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=34'>35</a>\u001b[0m trainer \u001b[39m=\u001b[39m Trainer(\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=35'>36</a>\u001b[0m     model\u001b[39m=\u001b[39mmodel,\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=36'>37</a>\u001b[0m     args\u001b[39m=\u001b[39mtraining_args,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=39'>40</a>\u001b[0m     compute_metrics\u001b[39m=\u001b[39mcompute_metrics,  \u001b[39m# Add this line to include evaluation metrics\u001b[39;00m\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=40'>41</a>\u001b[0m )\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=42'>43</a>\u001b[0m \u001b[39m# train the model\u001b[39;00m\n\u001b[0;32m---> <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=43'>44</a>\u001b[0m trainer\u001b[39m.\u001b[39;49mtrain()\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=45'>46</a>\u001b[0m \u001b[39m# evaluate the model on the validation set\u001b[39;00m\n\u001b[1;32m     <a href='vscode-notebook-cell:/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp/perpetrator.ipynb#X24sZmlsZQ%3D%3D?line=46'>47</a>\u001b[0m trainer\u001b[39m.\u001b[39mevaluate()\n",
      "File \u001b[0;32m~/anaconda3/envs/code-satp/lib/python3.9/site-packages/transformers/trainer.py:1624\u001b[0m, in \u001b[0;36mTrainer.train\u001b[0;34m(self, resume_from_checkpoint, trial, ignore_keys_for_eval, **kwargs)\u001b[0m\n\u001b[1;32m   1622\u001b[0m         hf_hub_utils\u001b[39m.\u001b[39menable_progress_bars()\n\u001b[1;32m   1623\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 1624\u001b[0m     \u001b[39mreturn\u001b[39;00m inner_training_loop(\n\u001b[1;32m   1625\u001b[0m         args\u001b[39m=\u001b[39;49margs,\n\u001b[1;32m   1626\u001b[0m         resume_from_checkpoint\u001b[39m=\u001b[39;49mresume_from_checkpoint,\n\u001b[1;32m   1627\u001b[0m         trial\u001b[39m=\u001b[39;49mtrial,\n\u001b[1;32m   1628\u001b[0m         ignore_keys_for_eval\u001b[39m=\u001b[39;49mignore_keys_for_eval,\n\u001b[1;32m   1629\u001b[0m     )\n",
      "File \u001b[0;32m~/anaconda3/envs/code-satp/lib/python3.9/site-packages/transformers/trainer.py:2003\u001b[0m, in \u001b[0;36mTrainer._inner_training_loop\u001b[0;34m(self, batch_size, args, resume_from_checkpoint, trial, ignore_keys_for_eval)\u001b[0m\n\u001b[1;32m   1998\u001b[0m     _grad_norm \u001b[39m=\u001b[39m nn\u001b[39m.\u001b[39mutils\u001b[39m.\u001b[39mclip_grad_norm_(\n\u001b[1;32m   1999\u001b[0m         amp\u001b[39m.\u001b[39mmaster_params(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39moptimizer),\n\u001b[1;32m   2000\u001b[0m         args\u001b[39m.\u001b[39mmax_grad_norm,\n\u001b[1;32m   2001\u001b[0m     )\n\u001b[1;32m   2002\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m-> 2003\u001b[0m     _grad_norm \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49maccelerator\u001b[39m.\u001b[39;49mclip_grad_norm_(\n\u001b[1;32m   2004\u001b[0m         model\u001b[39m.\u001b[39;49mparameters(),\n\u001b[1;32m   2005\u001b[0m         args\u001b[39m.\u001b[39;49mmax_grad_norm,\n\u001b[1;32m   2006\u001b[0m     )\n\u001b[1;32m   2008\u001b[0m \u001b[39mif\u001b[39;00m (\n\u001b[1;32m   2009\u001b[0m     is_accelerate_available()\n\u001b[1;32m   2010\u001b[0m     \u001b[39mand\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39maccelerator\u001b[39m.\u001b[39mdistributed_type \u001b[39m==\u001b[39m DistributedType\u001b[39m.\u001b[39mDEEPSPEED\n\u001b[1;32m   2011\u001b[0m ):\n\u001b[1;32m   2012\u001b[0m     grad_norm \u001b[39m=\u001b[39m model\u001b[39m.\u001b[39mget_global_grad_norm()\n",
      "File \u001b[0;32m~/anaconda3/envs/code-satp/lib/python3.9/site-packages/accelerate/accelerator.py:2146\u001b[0m, in \u001b[0;36mAccelerator.clip_grad_norm_\u001b[0;34m(self, parameters, max_norm, norm_type)\u001b[0m\n\u001b[1;32m   2144\u001b[0m             acc_opt\u001b[39m.\u001b[39mgradient_state\u001b[39m.\u001b[39mis_xla_gradients_synced \u001b[39m=\u001b[39m \u001b[39mTrue\u001b[39;00m\n\u001b[1;32m   2145\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39munscale_gradients()\n\u001b[0;32m-> 2146\u001b[0m \u001b[39mreturn\u001b[39;00m torch\u001b[39m.\u001b[39;49mnn\u001b[39m.\u001b[39;49mutils\u001b[39m.\u001b[39;49mclip_grad_norm_(parameters, max_norm, norm_type\u001b[39m=\u001b[39;49mnorm_type)\n",
      "File \u001b[0;32m~/anaconda3/envs/code-satp/lib/python3.9/site-packages/torch/nn/utils/clip_grad.py:82\u001b[0m, in \u001b[0;36mclip_grad_norm_\u001b[0;34m(parameters, max_norm, norm_type, error_if_nonfinite, foreach)\u001b[0m\n\u001b[1;32m     80\u001b[0m         clip_coef_clamped_device \u001b[39m=\u001b[39m clip_coef_clamped\u001b[39m.\u001b[39mto(device)\n\u001b[1;32m     81\u001b[0m         \u001b[39mfor\u001b[39;00m g \u001b[39min\u001b[39;00m grads:\n\u001b[0;32m---> 82\u001b[0m             g\u001b[39m.\u001b[39;49mdetach()\u001b[39m.\u001b[39;49mmul_(clip_coef_clamped_device)\n\u001b[1;32m     84\u001b[0m \u001b[39mreturn\u001b[39;00m total_norm\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# import libraries\n",
    "from transformers import AutoModelForSequenceClassification, Trainer, TrainingArguments\n",
    "from sklearn.metrics import accuracy_score, precision_recall_fscore_support\n",
    "\n",
    "# define model\n",
    "model = AutoModelForSequenceClassification.from_pretrained('distilbert-base-uncased', num_labels=len(label_encoder.classes_))\n",
    "#model = DistilBertForSequenceClassification.from_pretrained('distilbert-base-uncased', num_labels=3)\n",
    "\n",
    "# add evaluation metrics\n",
    "def compute_metrics(pred):\n",
    "    labels = pred.label_ids\n",
    "    preds = pred.predictions.argmax(-1)\n",
    "    precision, recall, f1, _ = precision_recall_fscore_support(labels, preds, average='weighted')\n",
    "    acc = accuracy_score(labels, preds)\n",
    "    return {\n",
    "        'accuracy': acc,\n",
    "        'f1': f1,\n",
    "        'precision': precision,\n",
    "        'recall': recall\n",
    "    }\n",
    "\n",
    "# specify training arguments\n",
    "training_args = TrainingArguments(\n",
    "    output_dir='./results',\n",
    "    num_train_epochs=3,\n",
    "    per_device_train_batch_size=4,\n",
    "    per_device_eval_batch_size=4,\n",
    "    warmup_steps=500,\n",
    "    weight_decay=0.01,\n",
    "    logging_dir='./logs',\n",
    "    evaluation_strategy='epoch',  # Evaluate at the end of each epoch\n",
    ")\n",
    "\n",
    "# specify training args\n",
    "trainer = Trainer(\n",
    "    model=model,\n",
    "    args=training_args,\n",
    "    train_dataset=train_dataset,\n",
    "    eval_dataset=val_dataset,\n",
    "    compute_metrics=compute_metrics,  # Add this line to include evaluation metrics\n",
    ")\n",
    "\n",
    "# train the model\n",
    "trainer.train()\n",
    "\n",
    "# evaluate the model on the validation set\n",
    "trainer.evaluate()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the model on the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 496/496 [00:47<00:00, 10.37it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'eval_loss': 0.25290557742118835,\n",
       " 'eval_accuracy': 0.9515885022692889,\n",
       " 'eval_f1': 0.9518842404484158,\n",
       " 'eval_precision': 0.9522573284725719,\n",
       " 'eval_recall': 0.9515885022692889,\n",
       " 'eval_runtime': 48.0274,\n",
       " 'eval_samples_per_second': 41.289,\n",
       " 'eval_steps_per_second': 10.327,\n",
       " 'epoch': 3.0}"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trainer.evaluate(test_dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save the model..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator/tokenizer_config.json',\n",
       " '/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator/special_tokens_map.json',\n",
       " '/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator/vocab.txt',\n",
       " '/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator/added_tokens.json',\n",
       " '/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator/tokenizer.json')"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.save_pretrained('/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator')\n",
    "tokenizer.save_pretrained('/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## To use model again..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import AutoModelForSequenceClassification, AutoTokenizer\n",
    "\n",
    "model = AutoModelForSequenceClassification.from_pretrained('/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator')\n",
    "tokenizer = AutoTokenizer.from_pretrained('/Users/ejt/Library/CloudStorage/Dropbox/Projects/code-satp-models/perpetrator')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "code-satp",
   "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": "3.9.18"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}