Result Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import pickle\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TrainerState(epoch=5.0, global_step=37500, max_steps=37500, num_train_epochs=5, total_flos=4.531428232585574e+16, log_history=[{'loss': 0.0187, 'learning_rate': 1.866826666666667e-05, 'epoch': 0.33, 'step': 2500}, {'eval_loss': 0.005186041351407766, 'eval_precision': 0.9785880689998168, 'eval_recall': 0.9786136851473745, 'eval_f1': 0.9786008769059615, 'eval_accuracy': 0.9986532346100907, 'eval_runtime': 57.7748, 'eval_samples_per_second': 86.543, 'eval_steps_per_second': 10.818, 'epoch': 0.33, 'step': 2500}, {'loss': 0.0055, 'learning_rate': 1.7334933333333336e-05, 'epoch': 0.67, 'step': 5000}, {'eval_loss': 0.0036667557433247566, 'eval_precision': 0.9795007050712905, 'eval_recall': 0.9818595885032197, 'eval_f1': 0.980678728299519, 'eval_accuracy': 0.9988668185485785, 'eval_runtime': 57.4715, 'eval_samples_per_second': 87.0, 'eval_steps_per_second': 10.875, 'epoch': 0.67, 'step': 5000}, {'loss': 0.0046, 'learning_rate': 1.6003733333333334e-05, 'epoch': 1.0, 'step': 7500}, {'eval_loss': 0.005106177646666765, 'eval_precision': 0.977784730913642, 'eval_recall': 0.9816239987435212, 'eval_f1': 0.9797006034955716, 'eval_accuracy': 0.9989380131947411, 'eval_runtime': 59.3344, 'eval_samples_per_second': 84.268, 'eval_steps_per_second': 10.534, 'epoch': 1.0, 'step': 7500}, {'loss': 0.0033, 'learning_rate': 1.4670400000000001e-05, 'epoch': 1.33, 'step': 10000}, {'eval_loss': 0.003167875576764345, 'eval_precision': 0.9843623336209827, 'eval_recall': 0.98536725825873, 'eval_f1': 0.9848645395915911, 'eval_accuracy': 0.9991385447814325, 'eval_runtime': 57.5338, 'eval_samples_per_second': 86.905, 'eval_steps_per_second': 10.863, 'epoch': 1.33, 'step': 10000}, {'loss': 0.0029, 'learning_rate': 1.3337066666666666e-05, 'epoch': 1.67, 'step': 12500}, {'eval_loss': 0.0035043801181018353, 'eval_precision': 0.9823443376514835, 'eval_recall': 0.9845557824197686, 'eval_f1': 0.9834488168388025, 'eval_accuracy': 0.9989403863496131, 'eval_runtime': 59.8562, 'eval_samples_per_second': 83.534, 'eval_steps_per_second': 10.442, 'epoch': 1.67, 'step': 12500}, {'loss': 0.0033, 'learning_rate': 1.2004266666666668e-05, 'epoch': 2.0, 'step': 15000}, {'eval_loss': 0.0029357888270169497, 'eval_precision': 0.9870962676019474, 'eval_recall': 0.9871996230563845, 'eval_f1': 0.9871479426238091, 'eval_accuracy': 0.9992560159476007, 'eval_runtime': 58.1942, 'eval_samples_per_second': 85.919, 'eval_steps_per_second': 10.74, 'epoch': 2.0, 'step': 15000}, {'loss': 0.0021, 'learning_rate': 1.0671466666666668e-05, 'epoch': 2.33, 'step': 17500}, {'eval_loss': 0.0028338793199509382, 'eval_precision': 0.9860094142259415, 'eval_recall': 0.9869902099366525, 'eval_f1': 0.9864995683001492, 'eval_accuracy': 0.9992061796952869, 'eval_runtime': 57.7068, 'eval_samples_per_second': 86.645, 'eval_steps_per_second': 10.831, 'epoch': 2.33, 'step': 17500}, {'loss': 0.0035, 'learning_rate': 9.339733333333335e-06, 'epoch': 2.67, 'step': 20000}, {'eval_loss': 0.002911886665970087, 'eval_precision': 0.9870010984987184, 'eval_recall': 0.9878278624155803, 'eval_f1': 0.9874143073944214, 'eval_accuracy': 0.9992655085670891, 'eval_runtime': 59.2725, 'eval_samples_per_second': 84.356, 'eval_steps_per_second': 10.545, 'epoch': 2.67, 'step': 20000}, {'loss': 0.0022, 'learning_rate': 8.006933333333335e-06, 'epoch': 3.0, 'step': 22500}, {'eval_loss': 0.0028483825735747814, 'eval_precision': 0.9882171192165693, 'eval_recall': 0.9879325689754463, 'eval_f1': 0.988074823609493, 'eval_accuracy': 0.9992797474963216, 'eval_runtime': 58.0931, 'eval_samples_per_second': 86.069, 'eval_steps_per_second': 10.759, 'epoch': 3.0, 'step': 22500}, {'loss': 0.0014, 'learning_rate': 6.6736e-06, 'epoch': 3.33, 'step': 25000}, {'eval_loss': 0.002810489619150758, 'eval_precision': 0.9877803071931339, 'eval_recall': 0.9881681587351447, 'eval_f1': 0.9879741948991743, 'eval_accuracy': 0.9993319569035075, 'eval_runtime': 57.904, 'eval_samples_per_second': 86.35, 'eval_steps_per_second': 10.794, 'epoch': 3.33, 'step': 25000}, {'loss': 0.0012, 'learning_rate': 5.340266666666667e-06, 'epoch': 3.67, 'step': 27500}, {'eval_loss': 0.0027514391113072634, 'eval_precision': 0.9891082368958475, 'eval_recall': 0.9889011046542066, 'eval_f1': 0.9890046599298391, 'eval_accuracy': 0.9993639944942807, 'eval_runtime': 58.342, 'eval_samples_per_second': 85.702, 'eval_steps_per_second': 10.713, 'epoch': 3.67, 'step': 27500}, {'loss': 0.0016, 'learning_rate': 4.007466666666667e-06, 'epoch': 4.0, 'step': 30000}, {'eval_loss': 0.003099782159551978, 'eval_precision': 0.9858706155814986, 'eval_recall': 0.9881158054552117, 'eval_f1': 0.9869919336915453, 'eval_accuracy': 0.999316531396839, 'eval_runtime': 58.471, 'eval_samples_per_second': 85.512, 'eval_steps_per_second': 10.689, 'epoch': 4.0, 'step': 30000}, {'loss': 0.0008, 'learning_rate': 2.6741333333333338e-06, 'epoch': 4.33, 'step': 32500}, {'eval_loss': 0.0030345479026436806, 'eval_precision': 0.9861800512043472, 'eval_recall': 0.9881419820951782, 'eval_f1': 0.9871600418410043, 'eval_accuracy': 0.9993236508614552, 'eval_runtime': 58.549, 'eval_samples_per_second': 85.399, 'eval_steps_per_second': 10.675, 'epoch': 4.33, 'step': 32500}, {'loss': 0.0011, 'learning_rate': 1.3413333333333334e-06, 'epoch': 4.67, 'step': 35000}, {'eval_loss': 0.002844160422682762, 'eval_precision': 0.9877640660949593, 'eval_recall': 0.9889534579341396, 'eval_f1': 0.9883584041857423, 'eval_accuracy': 0.9993628079168446, 'eval_runtime': 59.8393, 'eval_samples_per_second': 83.557, 'eval_steps_per_second': 10.445, 'epoch': 4.67, 'step': 35000}, {'loss': 0.0009, 'learning_rate': 8e-09, 'epoch': 5.0, 'step': 37500}, {'eval_loss': 0.002769122365862131, 'eval_precision': 0.9897147343627323, 'eval_recall': 0.9899219936128999, 'eval_f1': 0.9898183531382506, 'eval_accuracy': 0.999424509943519, 'eval_runtime': 58.1175, 'eval_samples_per_second': 86.033, 'eval_steps_per_second': 10.754, 'epoch': 5.0, 'step': 37500}, {'train_runtime': 12132.8604, 'train_samples_per_second': 24.726, 'train_steps_per_second': 3.091, 'total_flos': 4.531428232585574e+16, 'train_loss': 0.0035408822377522788, 'epoch': 5.0, 'step': 37500}], best_metric=0.9898183531382506, best_model_checkpoint='./results/checkpoint-37500', is_local_process_zero=True, is_world_process_zero=True, is_hyper_param_search=False, trial_name=None, trial_params=None)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_history = pd.read_pickle('results/checkpoint-best/training_history.pkl')\n",
    "training_history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'loss': 0.0187,\n",
       "  'learning_rate': 1.866826666666667e-05,\n",
       "  'epoch': 0.33,\n",
       "  'step': 2500},\n",
       " {'eval_loss': 0.005186041351407766,\n",
       "  'eval_precision': 0.9785880689998168,\n",
       "  'eval_recall': 0.9786136851473745,\n",
       "  'eval_f1': 0.9786008769059615,\n",
       "  'eval_accuracy': 0.9986532346100907,\n",
       "  'eval_runtime': 57.7748,\n",
       "  'eval_samples_per_second': 86.543,\n",
       "  'eval_steps_per_second': 10.818,\n",
       "  'epoch': 0.33,\n",
       "  'step': 2500},\n",
       " {'loss': 0.0055,\n",
       "  'learning_rate': 1.7334933333333336e-05,\n",
       "  'epoch': 0.67,\n",
       "  'step': 5000},\n",
       " {'eval_loss': 0.0036667557433247566,\n",
       "  'eval_precision': 0.9795007050712905,\n",
       "  'eval_recall': 0.9818595885032197,\n",
       "  'eval_f1': 0.980678728299519,\n",
       "  'eval_accuracy': 0.9988668185485785,\n",
       "  'eval_runtime': 57.4715,\n",
       "  'eval_samples_per_second': 87.0,\n",
       "  'eval_steps_per_second': 10.875,\n",
       "  'epoch': 0.67,\n",
       "  'step': 5000},\n",
       " {'loss': 0.0046,\n",
       "  'learning_rate': 1.6003733333333334e-05,\n",
       "  'epoch': 1.0,\n",
       "  'step': 7500},\n",
       " {'eval_loss': 0.005106177646666765,\n",
       "  'eval_precision': 0.977784730913642,\n",
       "  'eval_recall': 0.9816239987435212,\n",
       "  'eval_f1': 0.9797006034955716,\n",
       "  'eval_accuracy': 0.9989380131947411,\n",
       "  'eval_runtime': 59.3344,\n",
       "  'eval_samples_per_second': 84.268,\n",
       "  'eval_steps_per_second': 10.534,\n",
       "  'epoch': 1.0,\n",
       "  'step': 7500},\n",
       " {'loss': 0.0033,\n",
       "  'learning_rate': 1.4670400000000001e-05,\n",
       "  'epoch': 1.33,\n",
       "  'step': 10000},\n",
       " {'eval_loss': 0.003167875576764345,\n",
       "  'eval_precision': 0.9843623336209827,\n",
       "  'eval_recall': 0.98536725825873,\n",
       "  'eval_f1': 0.9848645395915911,\n",
       "  'eval_accuracy': 0.9991385447814325,\n",
       "  'eval_runtime': 57.5338,\n",
       "  'eval_samples_per_second': 86.905,\n",
       "  'eval_steps_per_second': 10.863,\n",
       "  'epoch': 1.33,\n",
       "  'step': 10000},\n",
       " {'loss': 0.0029,\n",
       "  'learning_rate': 1.3337066666666666e-05,\n",
       "  'epoch': 1.67,\n",
       "  'step': 12500},\n",
       " {'eval_loss': 0.0035043801181018353,\n",
       "  'eval_precision': 0.9823443376514835,\n",
       "  'eval_recall': 0.9845557824197686,\n",
       "  'eval_f1': 0.9834488168388025,\n",
       "  'eval_accuracy': 0.9989403863496131,\n",
       "  'eval_runtime': 59.8562,\n",
       "  'eval_samples_per_second': 83.534,\n",
       "  'eval_steps_per_second': 10.442,\n",
       "  'epoch': 1.67,\n",
       "  'step': 12500},\n",
       " {'loss': 0.0033,\n",
       "  'learning_rate': 1.2004266666666668e-05,\n",
       "  'epoch': 2.0,\n",
       "  'step': 15000},\n",
       " {'eval_loss': 0.0029357888270169497,\n",
       "  'eval_precision': 0.9870962676019474,\n",
       "  'eval_recall': 0.9871996230563845,\n",
       "  'eval_f1': 0.9871479426238091,\n",
       "  'eval_accuracy': 0.9992560159476007,\n",
       "  'eval_runtime': 58.1942,\n",
       "  'eval_samples_per_second': 85.919,\n",
       "  'eval_steps_per_second': 10.74,\n",
       "  'epoch': 2.0,\n",
       "  'step': 15000},\n",
       " {'loss': 0.0021,\n",
       "  'learning_rate': 1.0671466666666668e-05,\n",
       "  'epoch': 2.33,\n",
       "  'step': 17500},\n",
       " {'eval_loss': 0.0028338793199509382,\n",
       "  'eval_precision': 0.9860094142259415,\n",
       "  'eval_recall': 0.9869902099366525,\n",
       "  'eval_f1': 0.9864995683001492,\n",
       "  'eval_accuracy': 0.9992061796952869,\n",
       "  'eval_runtime': 57.7068,\n",
       "  'eval_samples_per_second': 86.645,\n",
       "  'eval_steps_per_second': 10.831,\n",
       "  'epoch': 2.33,\n",
       "  'step': 17500},\n",
       " {'loss': 0.0035,\n",
       "  'learning_rate': 9.339733333333335e-06,\n",
       "  'epoch': 2.67,\n",
       "  'step': 20000},\n",
       " {'eval_loss': 0.002911886665970087,\n",
       "  'eval_precision': 0.9870010984987184,\n",
       "  'eval_recall': 0.9878278624155803,\n",
       "  'eval_f1': 0.9874143073944214,\n",
       "  'eval_accuracy': 0.9992655085670891,\n",
       "  'eval_runtime': 59.2725,\n",
       "  'eval_samples_per_second': 84.356,\n",
       "  'eval_steps_per_second': 10.545,\n",
       "  'epoch': 2.67,\n",
       "  'step': 20000},\n",
       " {'loss': 0.0022,\n",
       "  'learning_rate': 8.006933333333335e-06,\n",
       "  'epoch': 3.0,\n",
       "  'step': 22500},\n",
       " {'eval_loss': 0.0028483825735747814,\n",
       "  'eval_precision': 0.9882171192165693,\n",
       "  'eval_recall': 0.9879325689754463,\n",
       "  'eval_f1': 0.988074823609493,\n",
       "  'eval_accuracy': 0.9992797474963216,\n",
       "  'eval_runtime': 58.0931,\n",
       "  'eval_samples_per_second': 86.069,\n",
       "  'eval_steps_per_second': 10.759,\n",
       "  'epoch': 3.0,\n",
       "  'step': 22500},\n",
       " {'loss': 0.0014, 'learning_rate': 6.6736e-06, 'epoch': 3.33, 'step': 25000},\n",
       " {'eval_loss': 0.002810489619150758,\n",
       "  'eval_precision': 0.9877803071931339,\n",
       "  'eval_recall': 0.9881681587351447,\n",
       "  'eval_f1': 0.9879741948991743,\n",
       "  'eval_accuracy': 0.9993319569035075,\n",
       "  'eval_runtime': 57.904,\n",
       "  'eval_samples_per_second': 86.35,\n",
       "  'eval_steps_per_second': 10.794,\n",
       "  'epoch': 3.33,\n",
       "  'step': 25000},\n",
       " {'loss': 0.0012,\n",
       "  'learning_rate': 5.340266666666667e-06,\n",
       "  'epoch': 3.67,\n",
       "  'step': 27500},\n",
       " {'eval_loss': 0.0027514391113072634,\n",
       "  'eval_precision': 0.9891082368958475,\n",
       "  'eval_recall': 0.9889011046542066,\n",
       "  'eval_f1': 0.9890046599298391,\n",
       "  'eval_accuracy': 0.9993639944942807,\n",
       "  'eval_runtime': 58.342,\n",
       "  'eval_samples_per_second': 85.702,\n",
       "  'eval_steps_per_second': 10.713,\n",
       "  'epoch': 3.67,\n",
       "  'step': 27500},\n",
       " {'loss': 0.0016,\n",
       "  'learning_rate': 4.007466666666667e-06,\n",
       "  'epoch': 4.0,\n",
       "  'step': 30000},\n",
       " {'eval_loss': 0.003099782159551978,\n",
       "  'eval_precision': 0.9858706155814986,\n",
       "  'eval_recall': 0.9881158054552117,\n",
       "  'eval_f1': 0.9869919336915453,\n",
       "  'eval_accuracy': 0.999316531396839,\n",
       "  'eval_runtime': 58.471,\n",
       "  'eval_samples_per_second': 85.512,\n",
       "  'eval_steps_per_second': 10.689,\n",
       "  'epoch': 4.0,\n",
       "  'step': 30000},\n",
       " {'loss': 0.0008,\n",
       "  'learning_rate': 2.6741333333333338e-06,\n",
       "  'epoch': 4.33,\n",
       "  'step': 32500},\n",
       " {'eval_loss': 0.0030345479026436806,\n",
       "  'eval_precision': 0.9861800512043472,\n",
       "  'eval_recall': 0.9881419820951782,\n",
       "  'eval_f1': 0.9871600418410043,\n",
       "  'eval_accuracy': 0.9993236508614552,\n",
       "  'eval_runtime': 58.549,\n",
       "  'eval_samples_per_second': 85.399,\n",
       "  'eval_steps_per_second': 10.675,\n",
       "  'epoch': 4.33,\n",
       "  'step': 32500},\n",
       " {'loss': 0.0011,\n",
       "  'learning_rate': 1.3413333333333334e-06,\n",
       "  'epoch': 4.67,\n",
       "  'step': 35000},\n",
       " {'eval_loss': 0.002844160422682762,\n",
       "  'eval_precision': 0.9877640660949593,\n",
       "  'eval_recall': 0.9889534579341396,\n",
       "  'eval_f1': 0.9883584041857423,\n",
       "  'eval_accuracy': 0.9993628079168446,\n",
       "  'eval_runtime': 59.8393,\n",
       "  'eval_samples_per_second': 83.557,\n",
       "  'eval_steps_per_second': 10.445,\n",
       "  'epoch': 4.67,\n",
       "  'step': 35000},\n",
       " {'loss': 0.0009, 'learning_rate': 8e-09, 'epoch': 5.0, 'step': 37500},\n",
       " {'eval_loss': 0.002769122365862131,\n",
       "  'eval_precision': 0.9897147343627323,\n",
       "  'eval_recall': 0.9899219936128999,\n",
       "  'eval_f1': 0.9898183531382506,\n",
       "  'eval_accuracy': 0.999424509943519,\n",
       "  'eval_runtime': 58.1175,\n",
       "  'eval_samples_per_second': 86.033,\n",
       "  'eval_steps_per_second': 10.754,\n",
       "  'epoch': 5.0,\n",
       "  'step': 37500},\n",
       " {'train_runtime': 12132.8604,\n",
       "  'train_samples_per_second': 24.726,\n",
       "  'train_steps_per_second': 3.091,\n",
       "  'total_flos': 4.531428232585574e+16,\n",
       "  'train_loss': 0.0035408822377522788,\n",
       "  'epoch': 5.0,\n",
       "  'step': 37500}]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_history.log_history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_evaluation_overall = []\n",
    "for i in range(0, len(training_history.log_history)):\n",
    "    if \"eval_f1\" in training_history.log_history[i]:\n",
    "        training_evaluation_overall.append(training_history.log_history[i]['eval_f1'])\n",
    "    else:\n",
    "        continue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(training_evaluation_overall)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_evaluation_langs = {'DE': [0.9783721466451465, 0.9761882872115473, 0.9775497113534317, 0.9837758112094396, 0.9830414746543779, \n",
    "                                    0.9848491825926778, 0.9847458408221578, 0.9856873302959178, 0.9858636091541189, 0.9864466162640605, \n",
    "                                    0.9879729044744482, 0.9834859009154054, 0.982427086208483, 0.9850402761795166, 0.9884358442755126],\n",
    "                             'EN': [0.9831772127700248, 0.9840011461865419, 0.9855861015654829, 0.9869683517112988, 0.986272540297508,\n",
    "                                    0.9901984221850347, 0.9893530675578895, 0.9910202521971724, 0.9911597457829597, 0.9909281894576013,\n",
    "                                    0.9918800152846771, 0.9899365669862165, 0.990885230255309, 0.9917884082879785, 0.9923583914413985],\n",
    "                             'ES': [0.9781283614198638, 0.9846467554276119, 0.97672186225102, 0.9879894306990152, 0.9887276651876724,\n",
    "                                    0.9895583293326933, 0.9890228540579449, 0.9894976894916883, 0.990868677159678, 0.9905150678352743,\n",
    "                                    0.9906340057636889, 0.9904613354130423, 0.990160787137029, 0.9909996399855995, 0.9912385981757081],\n",
    "                             'FR': [0.9737428203024265, 0.9784575865491272, 0.9781388823942015, 0.980636883238073, 0.9754141238012206,\n",
    "                                    0.9839897160219704, 0.9827757342208209, 0.983165770399813, 0.9843887037361867, 0.9838248175182481,\n",
    "                                    0.9852072735777349, 0.9844589857443328, 0.9856884163794615, 0.985800268801496, 0.9870798012277112]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "steps = list(np.arange(2500,40000, 2500))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(steps, training_evaluation_langs['DE'], '.-', label='German')\n",
    "plt.plot(steps, training_evaluation_langs['EN'], '.-', label='English')\n",
    "plt.plot(steps, training_evaluation_langs['ES'], '.-', label='Spanish')\n",
    "plt.plot(steps, training_evaluation_langs['FR'], '.-', label='French')\n",
    "plt.plot(steps, training_evaluation_overall, '*--', label='Overall', color = \"black\")\n",
    "plt.xticks(steps, rotation = 45)\n",
    "plt.xlabel(\"Steps\")\n",
    "plt.ylabel(\"F1 Score\")\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.title(\"F1 Scores of Different Languages during Training Evaluation\")\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train_tl = pd.read_hdf('data/marc_multilingual-merged_train_tokenized_labeled.h5', 'df')\n",
    "df_test_tl = pd.read_hdf('data/marc_multilingual-merged_test_tokenized_labeled.h5', 'df')\n",
    "df_validation_tl = pd.read_hdf('data/marc_multilingual-merged_validation_tokenized_labeled.h5', 'df')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>reviews</th>\n",
       "      <th>labels</th>\n",
       "      <th>languages</th>\n",
       "      <th>review_tokens</th>\n",
       "      <th>token_labels</th>\n",
       "      <th>token_languages</th>\n",
       "      <th>unique_language_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[Carga algo lento., I do like the range of col...</td>\n",
       "      <td>[2, 1, 1, 1, 1, 0, 1, 3, 1, 2, 1, 1, 1, 3, 1, ...</td>\n",
       "      <td>[es, en, en, en, en, de, en, fr, en, es, en, e...</td>\n",
       "      <td>[Carga, algo, lento., I, do, like, the, range,...</td>\n",
       "      <td>[2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...</td>\n",
       "      <td>[es, es, es, en, en, en, en, en, en, en, en, e...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[Los imanes son potentes pero el pegamento que...</td>\n",
       "      <td>[2, 0, 3, 3, 1, 2]</td>\n",
       "      <td>[es, de, fr, fr, en, es]</td>\n",
       "      <td>[Los, imanes, son, potentes, pero, el, pegamen...</td>\n",
       "      <td>[2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, ...</td>\n",
       "      <td>[es, es, es, es, es, es, es, es, es, es, es, e...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[déçue de mon achat, je m'attendais à mieux !!...</td>\n",
       "      <td>[3, 0, 0, 2, 2, 3, 2, 2, 3, 0, 3, 0, 3, 1]</td>\n",
       "      <td>[fr, de, de, es, es, fr, es, es, fr, de, fr, d...</td>\n",
       "      <td>[déçue, de, mon, achat,, je, m'attendais, à, m...</td>\n",
       "      <td>[3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>[fr, fr, fr, fr, fr, fr, fr, fr, fr, de, de, d...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[Je suis très déçue car je n’ai pas reçu la bo...</td>\n",
       "      <td>[3, 0, 1, 1, 2]</td>\n",
       "      <td>[fr, de, en, en, es]</td>\n",
       "      <td>[Je, suis, très, déçue, car, je, n’ai, pas, re...</td>\n",
       "      <td>[3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ...</td>\n",
       "      <td>[fr, fr, fr, fr, fr, fr, fr, fr, fr, fr, fr, f...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[El diseño de La foto es más llamativo que las...</td>\n",
       "      <td>[2, 2, 0, 1, 0, 0, 1, 3, 2, 0, 3, 3, 2]</td>\n",
       "      <td>[es, es, de, en, de, de, en, fr, es, de, fr, f...</td>\n",
       "      <td>[El, diseño, de, La, foto, es, más, llamativo,...</td>\n",
       "      <td>[2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, ...</td>\n",
       "      <td>[es, es, es, es, es, es, es, es, es, es, es, e...</td>\n",
       "      <td>4</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59995</th>\n",
       "      <td>[Le tactile fonctionne bien., Die beschriebene...</td>\n",
       "      <td>[3, 0, 2, 1, 3, 2, 1, 2, 0, 0, 3, 0]</td>\n",
       "      <td>[fr, de, es, en, fr, es, en, es, de, de, fr, de]</td>\n",
       "      <td>[Le, tactile, fonctionne, bien., Die, beschrie...</td>\n",
       "      <td>[3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>[fr, fr, fr, fr, de, de, de, de, de, de, de, d...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59996</th>\n",
       "      <td>[Very disappointed with this purchase., Had to...</td>\n",
       "      <td>[1, 1, 1, 0, 0, 0, 1, 2, 3, 1]</td>\n",
       "      <td>[en, en, en, de, de, de, en, es, fr, en]</td>\n",
       "      <td>[Very, disappointed, with, this, purchase., Ha...</td>\n",
       "      <td>[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...</td>\n",
       "      <td>[en, en, en, en, en, en, en, en, en, en, en, e...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59997</th>\n",
       "      <td>[When the battery arrived put in laptop and it...</td>\n",
       "      <td>[1, 2, 1, 2, 1, 3, 2, 1, 2, 2, 0, 2, 2, 0]</td>\n",
       "      <td>[en, es, en, es, en, fr, es, en, es, es, de, e...</td>\n",
       "      <td>[When, the, battery, arrived, put, in, laptop,...</td>\n",
       "      <td>[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, ...</td>\n",
       "      <td>[en, en, en, en, en, en, en, en, en, en, en, e...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59998</th>\n",
       "      <td>[Al principio no conectaba con los textos., I’...</td>\n",
       "      <td>[2, 1, 1, 3, 3, 0, 2, 2, 2, 2, 1, 1, 2, 3]</td>\n",
       "      <td>[es, en, en, fr, fr, de, es, es, es, es, en, e...</td>\n",
       "      <td>[Al, principio, no, conectaba, con, los, texto...</td>\n",
       "      <td>[2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, ...</td>\n",
       "      <td>[es, es, es, es, es, es, es, en, en, en, en, e...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59999</th>\n",
       "      <td>[Geliefert wurde eine sehr dünne rote Schlauch...</td>\n",
       "      <td>[0, 2, 0, 1, 0, 2, 2, 1, 1, 3, 0]</td>\n",
       "      <td>[de, es, de, en, de, es, es, en, en, fr, de]</td>\n",
       "      <td>[Geliefert, wurde, eine, sehr, dünne, rote, Sc...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 0, 0, 0, 0, ...</td>\n",
       "      <td>[de, de, de, de, de, de, de, es, es, es, es, d...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>60000 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 reviews  \\\n",
       "0      [Carga algo lento., I do like the range of col...   \n",
       "1      [Los imanes son potentes pero el pegamento que...   \n",
       "2      [déçue de mon achat, je m'attendais à mieux !!...   \n",
       "3      [Je suis très déçue car je n’ai pas reçu la bo...   \n",
       "4      [El diseño de La foto es más llamativo que las...   \n",
       "...                                                  ...   \n",
       "59995  [Le tactile fonctionne bien., Die beschriebene...   \n",
       "59996  [Very disappointed with this purchase., Had to...   \n",
       "59997  [When the battery arrived put in laptop and it...   \n",
       "59998  [Al principio no conectaba con los textos., I’...   \n",
       "59999  [Geliefert wurde eine sehr dünne rote Schlauch...   \n",
       "\n",
       "                                                  labels  \\\n",
       "0      [2, 1, 1, 1, 1, 0, 1, 3, 1, 2, 1, 1, 1, 3, 1, ...   \n",
       "1                                     [2, 0, 3, 3, 1, 2]   \n",
       "2             [3, 0, 0, 2, 2, 3, 2, 2, 3, 0, 3, 0, 3, 1]   \n",
       "3                                        [3, 0, 1, 1, 2]   \n",
       "4                [2, 2, 0, 1, 0, 0, 1, 3, 2, 0, 3, 3, 2]   \n",
       "...                                                  ...   \n",
       "59995               [3, 0, 2, 1, 3, 2, 1, 2, 0, 0, 3, 0]   \n",
       "59996                     [1, 1, 1, 0, 0, 0, 1, 2, 3, 1]   \n",
       "59997         [1, 2, 1, 2, 1, 3, 2, 1, 2, 2, 0, 2, 2, 0]   \n",
       "59998         [2, 1, 1, 3, 3, 0, 2, 2, 2, 2, 1, 1, 2, 3]   \n",
       "59999                  [0, 2, 0, 1, 0, 2, 2, 1, 1, 3, 0]   \n",
       "\n",
       "                                               languages  \\\n",
       "0      [es, en, en, en, en, de, en, fr, en, es, en, e...   \n",
       "1                               [es, de, fr, fr, en, es]   \n",
       "2      [fr, de, de, es, es, fr, es, es, fr, de, fr, d...   \n",
       "3                                   [fr, de, en, en, es]   \n",
       "4      [es, es, de, en, de, de, en, fr, es, de, fr, f...   \n",
       "...                                                  ...   \n",
       "59995   [fr, de, es, en, fr, es, en, es, de, de, fr, de]   \n",
       "59996           [en, en, en, de, de, de, en, es, fr, en]   \n",
       "59997  [en, es, en, es, en, fr, es, en, es, es, de, e...   \n",
       "59998  [es, en, en, fr, fr, de, es, es, es, es, en, e...   \n",
       "59999       [de, es, de, en, de, es, es, en, en, fr, de]   \n",
       "\n",
       "                                           review_tokens  \\\n",
       "0      [Carga, algo, lento., I, do, like, the, range,...   \n",
       "1      [Los, imanes, son, potentes, pero, el, pegamen...   \n",
       "2      [déçue, de, mon, achat,, je, m'attendais, à, m...   \n",
       "3      [Je, suis, très, déçue, car, je, n’ai, pas, re...   \n",
       "4      [El, diseño, de, La, foto, es, más, llamativo,...   \n",
       "...                                                  ...   \n",
       "59995  [Le, tactile, fonctionne, bien., Die, beschrie...   \n",
       "59996  [Very, disappointed, with, this, purchase., Ha...   \n",
       "59997  [When, the, battery, arrived, put, in, laptop,...   \n",
       "59998  [Al, principio, no, conectaba, con, los, texto...   \n",
       "59999  [Geliefert, wurde, eine, sehr, dünne, rote, Sc...   \n",
       "\n",
       "                                            token_labels  \\\n",
       "0      [2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
       "1      [2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, ...   \n",
       "2      [3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, 0, ...   \n",
       "3      [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ...   \n",
       "4      [2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, ...   \n",
       "...                                                  ...   \n",
       "59995  [3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...   \n",
       "59996  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
       "59997  [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, ...   \n",
       "59998  [2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, ...   \n",
       "59999  [0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 0, 0, 0, 0, ...   \n",
       "\n",
       "                                         token_languages  \\\n",
       "0      [es, es, es, en, en, en, en, en, en, en, en, e...   \n",
       "1      [es, es, es, es, es, es, es, es, es, es, es, e...   \n",
       "2      [fr, fr, fr, fr, fr, fr, fr, fr, fr, de, de, d...   \n",
       "3      [fr, fr, fr, fr, fr, fr, fr, fr, fr, fr, fr, f...   \n",
       "4      [es, es, es, es, es, es, es, es, es, es, es, e...   \n",
       "...                                                  ...   \n",
       "59995  [fr, fr, fr, fr, de, de, de, de, de, de, de, d...   \n",
       "59996  [en, en, en, en, en, en, en, en, en, en, en, e...   \n",
       "59997  [en, en, en, en, en, en, en, en, en, en, en, e...   \n",
       "59998  [es, es, es, es, es, es, es, en, en, en, en, e...   \n",
       "59999  [de, de, de, de, de, de, de, es, es, es, es, d...   \n",
       "\n",
       "       unique_language_count  \n",
       "0                          4  \n",
       "1                          4  \n",
       "2                          4  \n",
       "3                          4  \n",
       "4                          4  \n",
       "...                      ...  \n",
       "59995                      4  \n",
       "59996                      4  \n",
       "59997                      4  \n",
       "59998                      4  \n",
       "59999                      4  \n",
       "\n",
       "[60000 rows x 7 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train_tl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='unique_language_count', ylabel='count'>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=df_train_tl[\"unique_language_count\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='unique_language_count', ylabel='count'>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=df_test_tl[\"unique_language_count\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='unique_language_count', ylabel='count'>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=df_validation_tl[\"unique_language_count\"])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "lds",
   "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": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}