emotion_from_text_clf.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from transformers import RobertaTokenizerFast, TFRobertaForSequenceClassification, pipeline\n",
    "from sklearn.metrics import confusion_matrix, accuracy_score\n",
    "from seaborn import heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model ready\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    classifier = pipeline('sentiment-analysis', model='distilroberta_emo')\n",
    "except FileNotFoundError as e:\n",
    "    print(e, 'Downloading weigths...')\n",
    "    classifier = pipeline(\"sentiment-analysis\",model=\"j-hartmann/emotion-english-distilroberta-base\")\n",
    "\n",
    "print('model ready')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = pd.read_csv('sth_senti.csv', sep='\\t', header=None)\n",
    "dataset.columns = ['id', 'content', 'sentiment', 'DROPIT']\n",
    "dataset.drop('DROPIT', axis=1, inplace=True)\n",
    "dataset.set_index('id', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = dataset.sample(100)\n",
    "output = [classifier(txt) for txt in sample.content.values.tolist()]\n",
    "sample['emo'] = [[p['label'] for p in pred][np.argmax([p['score'] for p in pred])] for pred in output]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.77"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(sample.sentiment, sample.emo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "label_map = dict([(k, v) for v, k in enumerate(sample.emo.value_counts().index.tolist())])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(sample.sentiment.map(label_map), sample.emo.map(label_map))\n",
    "heatmap(cm, annot=True, fmt='d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\anton\\AppData\\Local\\Temp\\ipykernel_17208\\1717034821.py:1: RuntimeWarning: invalid value encountered in true_divide\n",
      "  label_map, 'acc', cm.diagonal() / cm.sum(axis=1), 'prec', cm.diagonal() / cm.sum(axis=0)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "({'fear': 0,\n",
       "  'anger': 1,\n",
       "  'joy': 2,\n",
       "  'sadness': 3,\n",
       "  'neutral': 4,\n",
       "  'surprise': 5,\n",
       "  'disgust': 6},\n",
       " 'acc',\n",
       " array([0.83333333, 0.73076923, 0.8       , 0.68421053,        nan,\n",
       "               nan,        nan]),\n",
       " 'prec',\n",
       " array([0.71428571, 0.9047619 , 1.        , 0.76470588, 0.        ,\n",
       "        0.        , 0.        ]))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "label_map, 'acc', cm.diagonal() / cm.sum(axis=1), 'prec', cm.diagonal() / cm.sum(axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "4891ca856c9b0ce2e4575d3dbe3f7fea92e25931a8a9aa4ff64a73277cd5eb3c"
  },
  "kernelspec": {
   "display_name": "Python 3.9.7 64-bit ('hack': venv)",
   "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": "3.9.7"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}