ensemble_wassa_EMO_binary.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 232,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "metadata": {},
   "outputs": [],
   "source": [
    "emotions = [\n",
    "    'fear',\n",
    "    'anger',\n",
    "    'surprise',\n",
    "    'sadness',\n",
    "    'disgust',\n",
    "    'joy',\n",
    "    'hope',\n",
    "    'neutral'\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = {}\n",
    "for emotion in emotions:\n",
    "\tpreds[emotion] = pd.read_csv(f'./dataset_binary/predictions/{emotion}.tsv', sep='\\t', header=None)\n",
    "n_samples = len(preds['sadness'])\n",
    "\n",
    "dev_lbl = pd.read_csv('./datasets/WASSA23_essay_level_dev_preproc.tsv', sep='\\t')['emotion'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " '',\n",
       " '',\n",
       " '',\n",
       " 'neutral',\n",
       " 'anger',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'disgust',\n",
       " '',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " '',\n",
       " '',\n",
       " 'sadness',\n",
       " 'anger',\n",
       " '',\n",
       " 'fear',\n",
       " 'joy',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'anger',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'disgust',\n",
       " '',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'disgust',\n",
       " 'fear/neutral',\n",
       " 'anger',\n",
       " 'sadness',\n",
       " 'fear',\n",
       " 'sadness',\n",
       " '',\n",
       " 'anger',\n",
       " 'anger/sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'neutral',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'anger',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " '',\n",
       " 'anger',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " 'disgust',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'disgust',\n",
       " 'disgust',\n",
       " 'fear/sadness',\n",
       " 'fear/sadness',\n",
       " 'sadness',\n",
       " 'anger',\n",
       " 'sadness',\n",
       " '',\n",
       " 'anger',\n",
       " 'disgust',\n",
       " 'neutral',\n",
       " '',\n",
       " 'anger/disgust',\n",
       " '',\n",
       " 'fear',\n",
       " 'anger',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " 'disgust',\n",
       " 'anger',\n",
       " '',\n",
       " '',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " '',\n",
       " 'anger',\n",
       " '',\n",
       " '',\n",
       " 'neutral',\n",
       " 'surprise',\n",
       " 'anger',\n",
       " 'joy',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'neutral',\n",
       " 'disgust',\n",
       " 'disgust',\n",
       " '',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'anger/neutral',\n",
       " '',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'anger/neutral',\n",
       " 'anger/disgust',\n",
       " '',\n",
       " 'disgust',\n",
       " 'fear/surprise',\n",
       " 'surprise',\n",
       " 'fear',\n",
       " 'neutral',\n",
       " 'disgust',\n",
       " '',\n",
       " 'sadness',\n",
       " '',\n",
       " '',\n",
       " '',\n",
       " 'sadness',\n",
       " '',\n",
       " 'surprise',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " 'fear/disgust',\n",
       " 'sadness',\n",
       " 'anger/neutral',\n",
       " '',\n",
       " 'fear/surprise',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness',\n",
       " 'anger',\n",
       " '',\n",
       " '',\n",
       " 'neutral',\n",
       " 'neutral',\n",
       " 'sadness',\n",
       " '',\n",
       " 'sadness/disgust',\n",
       " 'fear/hope',\n",
       " 'fear',\n",
       " 'sadness',\n",
       " 'anger/neutral',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'anger',\n",
       " 'anger/neutral',\n",
       " 'anger',\n",
       " 'anger',\n",
       " 'anger',\n",
       " 'neutral',\n",
       " 'fear/sadness',\n",
       " 'sadness/hope',\n",
       " '',\n",
       " 'anger/disgust',\n",
       " 'neutral',\n",
       " 'anger/neutral',\n",
       " 'fear',\n",
       " 'sadness',\n",
       " 'sadness',\n",
       " '',\n",
       " 'fear',\n",
       " 'sadness']"
      ]
     },
     "execution_count": 235,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preds_list = []\n",
    "for i in range(n_samples):\n",
    "    emos = []\n",
    "    for emotion in emotions:\n",
    "        if preds[emotion][0].loc[i]==emotion:\n",
    "            emos.append(emotion)\n",
    "    preds_list.append(emos)\n",
    "\n",
    "preds_str=[]\n",
    "for pred in preds_list:\n",
    "    str=''\n",
    "    for i, emo in enumerate(pred):\n",
    "        if i==len(pred)-1:\n",
    "            str+=emo\n",
    "        else:\n",
    "            str+=emo+'/'\n",
    "    preds_str.append(str)\n",
    "preds_str\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20\n"
     ]
    }
   ],
   "source": [
    "ctr=0\n",
    "for pred in preds_list:\n",
    "    if len(pred)>1:\n",
    "        ctr+=1\n",
    "print(ctr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", neutral\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", neutral\n",
      "sadness, sadness\n",
      ", sadness\n",
      ", neutral\n",
      ", sadness\n",
      "neutral, neutral\n",
      "anger, anger\n",
      ", hope/neutral\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "disgust, disgust\n",
      ", sadness\n",
      "neutral, neutral\n",
      "sadness, fear/sadness\n",
      "sadness, hope/sadness\n",
      ", anger\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", disgust\n",
      "sadness, sadness\n",
      ", anger\n",
      ", anger/disgust\n",
      "sadness, sadness\n",
      "anger, anger/neutral\n",
      ", sadness\n",
      "fear, neutral\n",
      "joy, neutral\n",
      "sadness, hope/sadness\n",
      "sadness, sadness\n",
      ", disgust\n",
      "anger, neutral\n",
      "sadness, neutral\n",
      ", fear\n",
      "sadness, neutral/sadness\n",
      "sadness, sadness/surprise\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", anger\n",
      "disgust, anger/disgust\n",
      ", neutral\n",
      "sadness, sadness\n",
      ", neutral\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "disgust, anger/disgust\n",
      "fear/neutral, neutral\n",
      "anger, neutral\n",
      "sadness, sadness\n",
      "fear, sadness\n",
      "sadness, hope\n",
      ", hope/sadness\n",
      "anger, anger\n",
      "anger/sadness, sadness\n",
      "sadness, sadness\n",
      ", joy\n",
      "sadness, sadness\n",
      "neutral, neutral\n",
      "neutral, neutral\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "anger, anger/disgust\n",
      "sadness, anger/sadness\n",
      "sadness, sadness\n",
      ", neutral/sadness\n",
      ", anger\n",
      "anger, anger/sadness\n",
      "neutral, neutral\n",
      "sadness, anger/sadness\n",
      "disgust, disgust\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "disgust, disgust\n",
      "disgust, disgust/sadness\n",
      "fear/sadness, sadness\n",
      "fear/sadness, sadness\n",
      "sadness, sadness\n",
      "anger, anger\n",
      "sadness, sadness\n",
      ", neutral\n",
      "anger, anger\n",
      "disgust, disgust\n",
      "neutral, neutral\n",
      ", fear\n",
      "anger/disgust, anger\n",
      ", sadness\n",
      "fear, sadness\n",
      "anger, anger/disgust\n",
      "neutral, neutral\n",
      "sadness, sadness\n",
      "disgust, disgust\n",
      "anger, neutral\n",
      ", anger\n",
      ", anger\n",
      "sadness, hope/sadness\n",
      ", neutral\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", neutral\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", neutral\n",
      "neutral, neutral\n",
      "sadness, sadness\n",
      ", neutral\n",
      "anger, anger\n",
      ", hope\n",
      ", anger\n",
      "neutral, neutral\n",
      "surprise, neutral\n",
      "anger, anger/disgust\n",
      "joy, hope\n",
      "sadness, sadness\n",
      ", sadness/surprise\n",
      "sadness, sadness\n",
      "sadness, anger/sadness\n",
      "sadness, neutral\n",
      "neutral, sadness\n",
      "disgust, disgust/hope\n",
      "disgust, disgust\n",
      ", joy\n",
      ", anger\n",
      "sadness, sadness\n",
      "sadness, anger/sadness\n",
      ", sadness\n",
      "sadness, sadness\n",
      "anger/neutral, anger/disgust\n",
      ", neutral\n",
      "neutral, sadness\n",
      "sadness, hope/sadness\n",
      "sadness, sadness\n",
      ", neutral\n",
      "anger/neutral, neutral\n",
      "anger/disgust, disgust\n",
      ", disgust/sadness\n",
      "disgust, sadness\n",
      "fear/surprise, neutral\n",
      "surprise, neutral/surprise\n",
      "fear, fear\n",
      "neutral, neutral\n",
      "disgust, disgust\n",
      ", hope\n",
      "sadness, sadness\n",
      ", anger\n",
      ", neutral\n",
      ", sadness\n",
      "sadness, disgust/sadness\n",
      ", neutral\n",
      "surprise, sadness\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      "fear/disgust, fear\n",
      "sadness, sadness\n",
      "anger/neutral, neutral\n",
      ", neutral\n",
      "fear/surprise, neutral\n",
      "neutral, anger\n",
      "sadness, sadness\n",
      "sadness, sadness\n",
      ", hope\n",
      "sadness, hope/sadness\n",
      "sadness, sadness\n",
      ", hope/neutral\n",
      "sadness, anger/sadness\n",
      "anger, neutral\n",
      ", neutral/sadness\n",
      ", neutral\n",
      "neutral, neutral\n",
      "neutral, hope\n",
      "sadness, sadness\n",
      ", anger\n",
      "sadness/disgust, disgust/sadness\n",
      "fear/hope, sadness\n",
      "fear, neutral\n",
      "sadness, sadness\n",
      "anger/neutral, anger\n",
      "sadness, disgust/sadness\n",
      "sadness, hope/sadness\n",
      ", sadness\n",
      "anger, anger\n",
      "anger/neutral, neutral\n",
      "anger, anger\n",
      "anger, neutral\n",
      "anger, anger/sadness\n",
      "neutral, anger\n",
      "fear/sadness, sadness\n",
      "sadness/hope, sadness\n",
      ", disgust\n",
      "anger/disgust, anger\n",
      "neutral, neutral\n",
      "anger/neutral, neutral\n",
      "fear, fear\n",
      "sadness, fear/sadness\n",
      "sadness, sadness\n",
      ", neutral\n",
      "fear, fear\n",
      "sadness, anger\n"
     ]
    }
   ],
   "source": [
    "for pred, true in zip(preds_str, dev_lbl):\n",
    "\tprint(f'{pred}, {true}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\giuli\\anaconda3\\lib\\site-packages\\sklearn\\preprocessing\\_label.py:895: UserWarning: unknown class(es) [''] will be ignored\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'macro_f1': 0.3794,\n",
       " 'micro_f1': 0.6176,\n",
       " 'micro_jaccard': 0.4467,\n",
       " 'macro_precision': 0.4373,\n",
       " 'macro_recall': 0.3637,\n",
       " 'micro_precision': 0.7429,\n",
       " 'micro_recall': 0.5285}"
      ]
     },
     "execution_count": 238,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from utils import compute_metrics\n",
    "scores = compute_metrics(dev_lbl, preds_str, \"EMO\")\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "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>fear</th>\n",
       "      <th>anger</th>\n",
       "      <th>surprise</th>\n",
       "      <th>sadness</th>\n",
       "      <th>disgust</th>\n",
       "      <th>joy</th>\n",
       "      <th>hope</th>\n",
       "      <th>neutral</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.494536</td>\n",
       "      <td>-0.407202</td>\n",
       "      <td>-0.583641</td>\n",
       "      <td>-0.562369</td>\n",
       "      <td>-0.480437</td>\n",
       "      <td>-0.506987</td>\n",
       "      <td>-0.547754</td>\n",
       "      <td>-0.536931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.489842</td>\n",
       "      <td>-0.412878</td>\n",
       "      <td>-0.613541</td>\n",
       "      <td>-0.456550</td>\n",
       "      <td>-0.442765</td>\n",
       "      <td>-0.532474</td>\n",
       "      <td>-0.541169</td>\n",
       "      <td>-0.559709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.495759</td>\n",
       "      <td>-0.405055</td>\n",
       "      <td>-0.564010</td>\n",
       "      <td>-0.501943</td>\n",
       "      <td>-0.476336</td>\n",
       "      <td>-0.525681</td>\n",
       "      <td>-0.554068</td>\n",
       "      <td>-0.523637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.501696</td>\n",
       "      <td>-0.399005</td>\n",
       "      <td>-0.682127</td>\n",
       "      <td>-0.478255</td>\n",
       "      <td>-0.461079</td>\n",
       "      <td>-0.528920</td>\n",
       "      <td>-0.580923</td>\n",
       "      <td>-0.517579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.511346</td>\n",
       "      <td>-0.405602</td>\n",
       "      <td>-0.560721</td>\n",
       "      <td>-0.558665</td>\n",
       "      <td>-0.478959</td>\n",
       "      <td>-0.550218</td>\n",
       "      <td>-0.541957</td>\n",
       "      <td>-0.528266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>-0.431560</td>\n",
       "      <td>-0.410137</td>\n",
       "      <td>-0.548627</td>\n",
       "      <td>-0.522740</td>\n",
       "      <td>-0.450733</td>\n",
       "      <td>-0.558613</td>\n",
       "      <td>-0.519598</td>\n",
       "      <td>-0.577776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>-0.508904</td>\n",
       "      <td>-0.404790</td>\n",
       "      <td>-0.564122</td>\n",
       "      <td>-0.602938</td>\n",
       "      <td>-0.482949</td>\n",
       "      <td>-0.540833</td>\n",
       "      <td>-0.555277</td>\n",
       "      <td>-0.527882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205</th>\n",
       "      <td>-0.534702</td>\n",
       "      <td>-0.405044</td>\n",
       "      <td>-0.603539</td>\n",
       "      <td>-0.476945</td>\n",
       "      <td>-0.501007</td>\n",
       "      <td>-0.537009</td>\n",
       "      <td>-0.550371</td>\n",
       "      <td>-0.402117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>206</th>\n",
       "      <td>-0.500915</td>\n",
       "      <td>-0.407830</td>\n",
       "      <td>-0.563245</td>\n",
       "      <td>-0.529514</td>\n",
       "      <td>-0.498139</td>\n",
       "      <td>-0.516272</td>\n",
       "      <td>-0.520820</td>\n",
       "      <td>-0.546265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>207</th>\n",
       "      <td>-0.490236</td>\n",
       "      <td>-0.406332</td>\n",
       "      <td>-0.565719</td>\n",
       "      <td>-0.546114</td>\n",
       "      <td>-0.481413</td>\n",
       "      <td>-0.566283</td>\n",
       "      <td>-0.565469</td>\n",
       "      <td>-0.525286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>208 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         fear     anger  surprise   sadness   disgust       joy      hope  \\\n",
       "0   -0.494536 -0.407202 -0.583641 -0.562369 -0.480437 -0.506987 -0.547754   \n",
       "1   -0.489842 -0.412878 -0.613541 -0.456550 -0.442765 -0.532474 -0.541169   \n",
       "2   -0.495759 -0.405055 -0.564010 -0.501943 -0.476336 -0.525681 -0.554068   \n",
       "3   -0.501696 -0.399005 -0.682127 -0.478255 -0.461079 -0.528920 -0.580923   \n",
       "4   -0.511346 -0.405602 -0.560721 -0.558665 -0.478959 -0.550218 -0.541957   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "203 -0.431560 -0.410137 -0.548627 -0.522740 -0.450733 -0.558613 -0.519598   \n",
       "204 -0.508904 -0.404790 -0.564122 -0.602938 -0.482949 -0.540833 -0.555277   \n",
       "205 -0.534702 -0.405044 -0.603539 -0.476945 -0.501007 -0.537009 -0.550371   \n",
       "206 -0.500915 -0.407830 -0.563245 -0.529514 -0.498139 -0.516272 -0.520820   \n",
       "207 -0.490236 -0.406332 -0.565719 -0.546114 -0.481413 -0.566283 -0.565469   \n",
       "\n",
       "      neutral  \n",
       "0   -0.536931  \n",
       "1   -0.559709  \n",
       "2   -0.523637  \n",
       "3   -0.517579  \n",
       "4   -0.528266  \n",
       "..        ...  \n",
       "203 -0.577776  \n",
       "204 -0.527882  \n",
       "205 -0.402117  \n",
       "206 -0.546265  \n",
       "207 -0.525286  \n",
       "\n",
       "[208 rows x 8 columns]"
      ]
     },
     "execution_count": 239,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logits_df = pd.DataFrame()\n",
    "for emotion in emotions:\n",
    "    logit = pd.read_csv(f'./dataset_binary/logits/logits_{emotion}.tsv')\n",
    "    logits_df[emotion] = (logit[f'no-{emotion}'] + logit[emotion] - 0.5).tolist()\n",
    "logits_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i, pred in enumerate(preds_str):\n",
    "    if pred == \"\":\n",
    "        #preds_str[i] = f\"{preds_dev[0].loc[i]}/neutral\"\n",
    "        preds_str[i] = emotions[np.argmax(logits_df.loc[i].tolist())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "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>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>sadness</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sadness</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>joy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>sadness</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>sadness</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>sadness</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>205</th>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>206</th>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>207</th>\n",
       "      <td>sadness</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>208 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           0\n",
       "0    sadness\n",
       "1    sadness\n",
       "2        joy\n",
       "3    neutral\n",
       "4    sadness\n",
       "..       ...\n",
       "203  sadness\n",
       "204  sadness\n",
       "205  neutral\n",
       "206     fear\n",
       "207  sadness\n",
       "\n",
       "[208 rows x 1 columns]"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preds_dev = pd.read_csv('./dataset_binary/predictions_EMO_dev.tsv', header=None)\n",
    "preds_dev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'macro_f1': 0.3917,\n",
       " 'micro_f1': 0.6329,\n",
       " 'micro_jaccard': 0.463,\n",
       " 'macro_precision': 0.4039,\n",
       " 'macro_recall': 0.4093,\n",
       " 'micro_precision': 0.6579,\n",
       " 'micro_recall': 0.6098}"
      ]
     },
     "execution_count": 242,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_scores = compute_metrics(dev_lbl, preds_str, \"EMO\")\n",
    "new_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [],
   "source": [
    "from utils import plot_confusion_matrix, plot_confusion_matrix_per_emotions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x700 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(dev_lbl, preds_str, title=None, path=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix_per_emotions(dev_lbl, preds_str, title = None, path = None)"
   ]
  }
 ],
 "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": "3.10.9"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}