IMPACT_lemmatisation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <center> Gender Biais detection</center>\n",
    "## <center> What are the impact of the lematisation on biais detection </center>\n",
    "### <center> Quillivic Robin </center>\n",
    "Sources :Grammatical gender associations outweigh topical gender bias incross linguistic wordembeddings\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Robin\\Anaconda3\\lib\\site-packages\\gensim\\utils.py:1197: UserWarning: detected Windows; aliasing chunkize to chunkize_serial\n",
      "  warnings.warn(\"detected Windows; aliasing chunkize to chunkize_serial\")\n"
     ]
    }
   ],
   "source": [
    "from gensim import *\n",
    "from gensim.test.utils import common_texts, get_tmpfile\n",
    "from gensim.models import Word2Vec\n",
    "import logging\n",
    "import gensim\n",
    "import os\n",
    "\n",
    "import itertools\n",
    "\n",
    "from gensim.test.utils import get_tmpfile\n",
    "from gensim.models import KeyedVectors\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from random import *\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import math\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "import lib\n",
    "import fasttext\n",
    "from gensim.models.wrappers import FastText\n",
    "import codecs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the French Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2019-03-23 15:15:14,297 : INFO : loading projection weights from DATA\\FR\\frWac_no_postag_no_phrase_500_skip_cut100.bin\n",
      "2019-03-23 15:15:16,121 : INFO : loaded (119227, 500) matrix from DATA\\FR\\frWac_no_postag_no_phrase_500_skip_cut100.bin\n",
      "2019-03-23 15:15:16,152 : INFO : loading projection weights from DATA\\FR\\wiki_fr.vec\n",
      "2019-03-23 15:15:28,687 : INFO : loaded (50000, 300) matrix from DATA\\FR\\wiki_fr.vec\n",
      "2019-03-23 15:15:28,703 : INFO : loading projection weights from C:/Users/Robin/Documents/IMPACT/Notebooks/DATA/FR/ppmi_svd_fr/matrix.txt\n",
      "2019-03-23 15:16:01,219 : INFO : loaded (53376, 500) matrix from C:/Users/Robin/Documents/IMPACT/Notebooks/DATA/FR/ppmi_svd_fr/matrix.txt\n",
      "2019-03-23 15:16:01,235 : INFO : precomputing L2-norms of word weight vectors\n",
      "2019-03-23 15:16:02,281 : INFO : precomputing L2-norms of word weight vectors\n",
      "2019-03-23 15:16:02,687 : INFO : precomputing L2-norms of word weight vectors\n"
     ]
    }
   ],
   "source": [
    "# Logging code taken from http://rare-technologies.com/word2vec-tutorial/\n",
    "logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)\n",
    "\n",
    "# Load pre-trained Word2Vec model.\n",
    "#model  = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\FR\\cc.fr.300.vec', binary=False,limit=50000)\n",
    "model_lem  = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\FR\\frWac_no_postag_no_phrase_500_skip_cut100.bin', binary=True)#,limit=50000)\n",
    "#model  = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\FR\\fr_Kyubyong_wordvectors.vec',binary=False,limit=50000)\n",
    "model_wiki  = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\FR\\wiki_fr.vec',binary=False,limit=50000)\n",
    "model_wiki_lem  = gensim.models.KeyedVectors.load_word2vec_format('C:/Users/Robin/Documents/IMPACT/Notebooks/DATA/FR/ppmi_svd_fr/matrix.txt',binary=False,encoding='latin1')\n",
    "# normalizeation :\n",
    "model_lem.init_sims(True)\n",
    "model_wiki.init_sims(True)\n",
    "model_wiki_lem.init_sims(True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the English Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 413,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2019-03-26 09:11:05,131 : INFO : loading projection weights from DATA\\ENG\\GoogleNews-vectors-negative300.bin\n",
      "2019-03-26 09:11:06,410 : INFO : loaded (50000, 300) matrix from DATA\\ENG\\GoogleNews-vectors-negative300.bin\n",
      "2019-03-26 09:11:06,524 : INFO : loading projection weights from DATA\\ENG\\crawl-300d-2M.vec\n",
      "2019-03-26 09:11:27,920 : INFO : loaded (50000, 300) matrix from DATA\\ENG\\crawl-300d-2M.vec\n",
      "2019-03-26 09:11:28,030 : INFO : loading projection weights from DATA\\ENG\\wiki.en.vec\n",
      "2019-03-26 09:11:48,449 : INFO : loaded (50000, 300) matrix from DATA\\ENG\\wiki.en.vec\n",
      "2019-03-26 09:11:48,449 : INFO : precomputing L2-norms of word weight vectors\n",
      "2019-03-26 09:11:49,149 : INFO : precomputing L2-norms of word weight vectors\n"
     ]
    }
   ],
   "source": [
    "model_en = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\ENG\\GoogleNews-vectors-negative300.bin', binary=True,limit=50000)\n",
    "model_en_ff = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\ENG\\crawl-300d-2M.vec', binary=False,limit=50000)\n",
    "model_en_wiki = gensim.models.KeyedVectors.load_word2vec_format(r'DATA\\ENG\\wiki.en.vec', binary=False,limit=50000)\n",
    "# normalizeation :\n",
    "model_en.init_sims(True)\n",
    "model_en_ff.init_sims(True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build the list of test word\n",
    "\n",
    "In the previous notebook we found that in the french model, grammatical gender chnage the represention andlake Biais decttection difficul. We suppose thta lemmatisation remove this difficulty. \n",
    "Now, We want to know if with we find the same kind of biais in lemmatised model than in english one. \n",
    "For this reason we create a set of test word where we know there is a Biais and evaluate the biais for each of this word in french and in english."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 417,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_fr = [#profession\n",
    "           'ingénieur','médecin', 'agriculteur','scientifique','avocat','politique','ministre',\n",
    "           'banquier','sportif','dentiste', 'maçon',\n",
    "           'secrétaire','infirmier', 'boulanger', 'journaliste', 'professeur','comptable','sage-femme',\n",
    "           'vendeur','employé','ménage',\n",
    "           # matière sholaire\n",
    "           'mathématique', 'science','physique','informatique','finance','chimie','histoire','poésie','peinture',\n",
    "           'art','danse','chant','littérature','gestion','production','économie','biologie','astronomie',\n",
    "           'architecture','environnement', 'géographie','géométrie','musique','philosophie','religion','théâtre',\n",
    "           #sport\n",
    "           'football','basket','tennis','pétanque','ping-pong','escalade','course','saut',\n",
    "           'natation','échec','jeu','poker', 'moto','vélo','cyclisme','kayak','surf','ski','athlétisme',\n",
    "           'aviron','aviation','badminton','cirque','curling','catch','équitation','escrime','golf','judo',\n",
    "           'musculation','polo','tango','voile','yoga',#'hand-ball'\n",
    "            #vie quotidienne\n",
    "            'cuisine','maison','jardin','magasin','shopping','sac','chaussure','sécurité','rock',\n",
    "            'jupe','pantalon','argent','table','couteau','serviette','vin','bière']#'bricolage'\n",
    "\n",
    "test_en =[#profession\n",
    "           'engineer','doctor','farmer','scientist','lawyer','politician','minister',\n",
    "           \"banker\", \"sportsman\", \"dentist\", \"builder\",\n",
    "           'secretary','nurse','baker','journalist','teacher','accountant','midwife',\n",
    "           \"salesman\", \"employee\", \"cleaning\",\n",
    "           # sholar material\n",
    "           'mathematics','science','physics','computer','finance','chemistry','history','poetry','painting',\n",
    "           'art','dance','song','literature','management','production','economics','biology','astronomy',\n",
    "           'architecture','environment','geography','geometry','music','philosophy','religion','theater',\n",
    "           #sport\n",
    "           'football','basketball','tennis','bowling','softball','climbing','running','jumping',\n",
    "           'swimming','chess','play','poker','motorcycle','bicycle','cycling','kayaking','surfing','skiing','athletics',\n",
    "           'rowing','aviation','badminton','circus','curling','wrestling','riding','fencing','golf','judo',\n",
    "           \"workout\", \"polo\",\"tango\", \"veil\", \"yoga\",#'hand-ball'\n",
    "            #Daily life\n",
    "            'kitchen','house','garden','shop','shopping','bag','shoe','security','rock',#'DIY'\n",
    "            \"skirt\", \"pants\", \"silver\", \"table\", \"knife\", \"towel\", \"wine\", \"beer\"]\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build the direction for each model language"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 341,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Pair_fr = [['elle','il'],['femme','homme'],['fille','garçon'],['mère','père'],['nièce','neveu'],\n",
    "           ['fille','fil|fils'],['reine','roi'],['s\\x9cur','frère'],['tante','oncle'],['maman','papa']]#\n",
    "Pair_fr = [['petite-fille','petit-fils'],['femme','homme'],['fille','garçon'],['mère','père'],['nièce','neveu'],\n",
    "           ['fille','fil|fils'],['reine','roi'],['s\\x9cur','frère'],['tante','oncle'],['Marie','Louis']]#\n",
    "\n",
    "direction_fr, expl_var_fr,eig_values_fr = lib.from_pair_to_direction2(Pair_fr, model_wiki_lem)\n",
    "\n",
    "#############\n",
    "Pair_en = [['she','he'],['woman','man'],['her','his'],['herself','himself'],['daughter','son'],\n",
    "           ['mother','father'],['sister','brother'],['female','male'],['girl','boy'],['gal','guy']]\n",
    "\n",
    "direction_en, expl_var_en,eig_values_en = lib.from_pair_to_direction2(Pair_en, model_en)\n",
    "lib.plot_var_eig(eig_values_fr,expl_var_fr)\n",
    "lib.plot_var_eig(eig_values_en,expl_var_en)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [],
   "source": [
    "def from_pair_to_contribution(Pair, model):\n",
    "    \"\"\"\n",
    "    Extraction of the contribution of each component for a pair of word in PCA\n",
    "     Entry:\n",
    "    - Pair: type:List  size:nx2 ; list of pair word\n",
    "    - model: type:GensimModel\n",
    "    Output:\n",
    "    - direction : principal component of the PCA\n",
    "    - expl_var : explained_variance_ratio_ of the pca\n",
    "    - eig_values: eig_values of the pca\n",
    "    \"\"\"\n",
    "    num_components = len(Pair)\n",
    "    matrix = []\n",
    "    label = []\n",
    "    for a, b in Pair:\n",
    "        center = (model[a] + model[b])/2\n",
    "        matrix.append(model[a] - center)\n",
    "        matrix.append(model[b] - center)\n",
    "        label.append(str(a+'-'+b))\n",
    "        label.append(str(b+'-'+a))\n",
    "    \n",
    "    matrix = np.array(matrix)\n",
    "    pca = PCA(n_components = num_components)\n",
    "    coord = pca.fit_transform(matrix)\n",
    "    eigval = pca.singular_values_\n",
    "    ctr = coord**2 \n",
    "    contrib= pd.DataFrame(index=label)\n",
    "    for j in range(num_components): \n",
    "        ctr[:,j] = ctr[:,j]/(eigval[j]) \n",
    "        contrib['CTR_'+str(j+1)] = ctr[:,j]\n",
    "    return(contrib)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-298-dac02329abe5>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mL\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlib\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0massociation_Genrator\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'tante'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m'oncle'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mmodel_wiki_lem\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mtop_n\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1000\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mL\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Documents\\IMPACT\\Notebooks\\lib.py\u001b[0m in \u001b[0;36massociation_Genrator\u001b[1;34m(w1, w2, model, top_n)\u001b[0m\n\u001b[0;32m     48\u001b[0m     \u001b[1;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mitertools\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcombinations\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mliste\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     49\u001b[0m         \u001b[0mu\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwv\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m-\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwv\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 50\u001b[1;33m         \u001b[0mS\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcosine_similarity\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0ma\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mu\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     51\u001b[0m         \u001b[0mP\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     52\u001b[0m         \u001b[0mi\u001b[0m\u001b[1;33m+=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\Documents\\IMPACT\\Notebooks\\lib.py\u001b[0m in \u001b[0;36mcosine_similarity\u001b[1;34m(w1, w2)\u001b[0m\n\u001b[0;32m    247\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    248\u001b[0m \u001b[1;32mdef\u001b[0m \u001b[0mcosine_similarity\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mw1\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mw2\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 249\u001b[1;33m     \u001b[1;32mreturn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdot\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mw1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mw2\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m/\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlinalg\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnorm\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mw1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m*\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlinalg\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnorm\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mw2\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    250\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    251\u001b[0m \u001b[1;32mdef\u001b[0m \u001b[0mplot_mean\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mrp\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m20\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "L = lib.association_Genrator('tante','oncle',model_wiki_lem,top_n=1000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2c22b097048>"
      ]
     },
     "execution_count": 312,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_contrib = from_pair_to_contribution(Pair_fr, model_wiki_lem)\n",
    "df = df_contrib.iloc[[2*i for i in range(10)],:].sort_values(by='CTR_1')\n",
    "df['CTR_1'].plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2c2264e9cc0>"
      ]
     },
     "execution_count": 296,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_contrib = from_pair_to_contribution(Pair_en, model_en_ff)\n",
    "df = df_contrib.iloc[[2*i for i in range(10)],:].sort_values(by='CTR_1')\n",
    "df['CTR_1'].plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {},
   "outputs": [],
   "source": [
    "Biais_fr = []\n",
    "Biais_en = []\n",
    "for wf,we in zip(test_fr,test_en):\n",
    "    Biais_fr.append(lib.cosine_similarity(model_wiki_lem[wf],direction_fr))\n",
    "    Biais_en.append(lib.cosine_similarity(model_en[we],direction_en))\n",
    "result = pd.DataFrame(data={'fr':Biais_fr,'en':Biais_en},index=test_fr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2c225dbea20>"
      ]
     },
     "execution_count": 303,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result.plot(kind='bar',figsize = (20,10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 323,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.distplot(result['en'], bins=50, label = 'en',kde=True)\n",
    "sns.distplot(result['fr'], bins=50, label = 'fr',kde=True)\n",
    "plt.legend()\n",
    "plt.show()\n",
    "#result.plot(kind='hist',bins = 50,ax=plt.gca(),grid=True,alpha=0.5,kde=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "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>fr</th>\n",
       "      <th>en</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-0.009970</td>\n",
       "      <td>0.011072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.034069</td>\n",
       "      <td>0.095346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-0.102856</td>\n",
       "      <td>-0.215074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.025179</td>\n",
       "      <td>-0.050896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.012330</td>\n",
       "      <td>0.002454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.012073</td>\n",
       "      <td>0.065940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.075136</td>\n",
       "      <td>0.304010</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               fr          en\n",
       "count  100.000000  100.000000\n",
       "mean    -0.009970    0.011072\n",
       "std      0.034069    0.095346\n",
       "min     -0.102856   -0.215074\n",
       "25%     -0.025179   -0.050896\n",
       "50%     -0.012330    0.002454\n",
       "75%      0.012073    0.065940\n",
       "max      0.075136    0.304010"
      ]
     },
     "execution_count": 225,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "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>fr</th>\n",
       "      <th>en</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>dentiste</th>\n",
       "      <td>-0.004192</td>\n",
       "      <td>0.001286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>secrétaire</th>\n",
       "      <td>-0.077823</td>\n",
       "      <td>0.017405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>boulanger</th>\n",
       "      <td>-0.104932</td>\n",
       "      <td>0.047509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>journaliste</th>\n",
       "      <td>-0.036608</td>\n",
       "      <td>0.016265</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>professeur</th>\n",
       "      <td>-0.092069</td>\n",
       "      <td>0.120145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comptable</th>\n",
       "      <td>-0.026212</td>\n",
       "      <td>0.000642</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vendeur</th>\n",
       "      <td>0.000581</td>\n",
       "      <td>-0.166202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employé</th>\n",
       "      <td>-0.016032</td>\n",
       "      <td>0.100549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physique</th>\n",
       "      <td>0.004378</td>\n",
       "      <td>-0.059291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informatique</th>\n",
       "      <td>-0.003480</td>\n",
       "      <td>0.064195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finance</th>\n",
       "      <td>-0.057892</td>\n",
       "      <td>0.011388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peinture</th>\n",
       "      <td>-0.012469</td>\n",
       "      <td>0.005044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>art</th>\n",
       "      <td>-0.016894</td>\n",
       "      <td>0.065928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chant</th>\n",
       "      <td>-0.035663</td>\n",
       "      <td>0.067158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>biologie</th>\n",
       "      <td>-0.001884</td>\n",
       "      <td>0.082304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>environnement</th>\n",
       "      <td>0.004321</td>\n",
       "      <td>-0.012350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>géographie</th>\n",
       "      <td>-0.007596</td>\n",
       "      <td>0.045944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>géométrie</th>\n",
       "      <td>0.002649</td>\n",
       "      <td>-0.015310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>musique</th>\n",
       "      <td>-0.017423</td>\n",
       "      <td>0.042451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>théâtre</th>\n",
       "      <td>-0.015496</td>\n",
       "      <td>0.106619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>basket</th>\n",
       "      <td>-0.005170</td>\n",
       "      <td>0.015120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tennis</th>\n",
       "      <td>-0.003689</td>\n",
       "      <td>0.089158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ping-pong</th>\n",
       "      <td>-0.025767</td>\n",
       "      <td>0.215449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>escalade</th>\n",
       "      <td>0.003782</td>\n",
       "      <td>-0.043977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saut</th>\n",
       "      <td>0.001313</td>\n",
       "      <td>-0.030818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jeu</th>\n",
       "      <td>0.001035</td>\n",
       "      <td>-0.137521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cyclisme</th>\n",
       "      <td>-0.038549</td>\n",
       "      <td>0.031198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>surf</th>\n",
       "      <td>-0.018909</td>\n",
       "      <td>0.026294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ski</th>\n",
       "      <td>-0.007330</td>\n",
       "      <td>0.021120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>aviron</th>\n",
       "      <td>-0.008098</td>\n",
       "      <td>0.090891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>badminton</th>\n",
       "      <td>-0.004263</td>\n",
       "      <td>0.062523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cirque</th>\n",
       "      <td>0.014300</td>\n",
       "      <td>-0.058848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>curling</th>\n",
       "      <td>-0.015859</td>\n",
       "      <td>0.022892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>équitation</th>\n",
       "      <td>0.007110</td>\n",
       "      <td>-0.066746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>judo</th>\n",
       "      <td>0.008152</td>\n",
       "      <td>-0.064823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>polo</th>\n",
       "      <td>-0.071942</td>\n",
       "      <td>0.003622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yoga</th>\n",
       "      <td>-0.024698</td>\n",
       "      <td>0.144753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuisine</th>\n",
       "      <td>-0.013349</td>\n",
       "      <td>0.113183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bricolage</th>\n",
       "      <td>-0.001740</td>\n",
       "      <td>0.081480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>magasin</th>\n",
       "      <td>-0.008188</td>\n",
       "      <td>0.039429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sac</th>\n",
       "      <td>-0.005589</td>\n",
       "      <td>0.004521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chaussure</th>\n",
       "      <td>0.008089</td>\n",
       "      <td>-0.021179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pantalon</th>\n",
       "      <td>0.003747</td>\n",
       "      <td>-0.032667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>argent</th>\n",
       "      <td>-0.001021</td>\n",
       "      <td>0.083673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vin</th>\n",
       "      <td>-0.008167</td>\n",
       "      <td>0.057344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bière</th>\n",
       "      <td>0.015244</td>\n",
       "      <td>-0.080741</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     fr        en\n",
       "dentiste      -0.004192  0.001286\n",
       "secrétaire    -0.077823  0.017405\n",
       "boulanger     -0.104932  0.047509\n",
       "journaliste   -0.036608  0.016265\n",
       "professeur    -0.092069  0.120145\n",
       "comptable     -0.026212  0.000642\n",
       "vendeur        0.000581 -0.166202\n",
       "employé       -0.016032  0.100549\n",
       "physique       0.004378 -0.059291\n",
       "informatique  -0.003480  0.064195\n",
       "finance       -0.057892  0.011388\n",
       "peinture      -0.012469  0.005044\n",
       "art           -0.016894  0.065928\n",
       "chant         -0.035663  0.067158\n",
       "biologie      -0.001884  0.082304\n",
       "environnement  0.004321 -0.012350\n",
       "géographie    -0.007596  0.045944\n",
       "géométrie      0.002649 -0.015310\n",
       "musique       -0.017423  0.042451\n",
       "théâtre       -0.015496  0.106619\n",
       "basket        -0.005170  0.015120\n",
       "tennis        -0.003689  0.089158\n",
       "ping-pong     -0.025767  0.215449\n",
       "escalade       0.003782 -0.043977\n",
       "saut           0.001313 -0.030818\n",
       "jeu            0.001035 -0.137521\n",
       "cyclisme      -0.038549  0.031198\n",
       "surf          -0.018909  0.026294\n",
       "ski           -0.007330  0.021120\n",
       "aviron        -0.008098  0.090891\n",
       "badminton     -0.004263  0.062523\n",
       "cirque         0.014300 -0.058848\n",
       "curling       -0.015859  0.022892\n",
       "équitation     0.007110 -0.066746\n",
       "judo           0.008152 -0.064823\n",
       "polo          -0.071942  0.003622\n",
       "yoga          -0.024698  0.144753\n",
       "cuisine       -0.013349  0.113183\n",
       "bricolage     -0.001740  0.081480\n",
       "magasin       -0.008188  0.039429\n",
       "sac           -0.005589  0.004521\n",
       "chaussure      0.008089 -0.021179\n",
       "pantalon       0.003747 -0.032667\n",
       "argent        -0.001021  0.083673\n",
       "vin           -0.008167  0.057344\n",
       "bière          0.015244 -0.080741"
      ]
     },
     "execution_count": 257,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result[(np.sign(result['en'])!=np.sign(result['fr']))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 389,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6161616161616161\n"
     ]
    }
   ],
   "source": [
    "def compute_error(Pair_fr,Pair_en,model_fr,model_en):\n",
    "    direction_en, expl_var_en,eig_values_en = lib.from_pair_to_direction2(Pair_en, model_en)\n",
    "    direction_fr, expl_var_fr,eig_values_fr = lib.from_pair_to_direction2(Pair_fr, model_fr)\n",
    "    Biais_fr = []\n",
    "    Biais_en = []\n",
    "    for wf,we in zip(test_fr,test_en):\n",
    "        Biais_fr.append(lib.cosine_similarity(model_fr[wf],direction_fr)*100)\n",
    "        Biais_en.append(lib.cosine_similarity(model_en[we],direction_en)*100)\n",
    "    \n",
    "    result = pd.DataFrame(data={'fr':Biais_fr,'en':Biais_en},index=test_fr)\n",
    "    print(len(result[np.sign(result['en'])==np.sign(result['fr'])])/len(result))\n",
    "    return(result)\n",
    "\n",
    "#Pair_fr = [['elle','il'],['femme','homme'],['fille','garçon'],['mère','père'],['fille','fils'],\n",
    "             #['la','le'],['reine','roi'],['jument', 'cheval'],['soeur','frère'],['tante','oncle']]\n",
    "result_bd = compute_error(Pair_fr,Pair_en,model_wiki_lem,model_en_ff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 418,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5612244897959183\n"
     ]
    }
   ],
   "source": [
    "result_bd_wiki = compute_error(Pair_fr,Pair_en,model_wiki_lem,model_en_wiki)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 398,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result=-result\n",
    "result.plot(kind='barh',figsize = (10,20))\n",
    "plt.savefig('all_dist_100_wefat.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 419,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2c2113cf748>"
      ]
     },
     "execution_count": 419,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result_bd_wiki.plot(kind='hist',bins = 50,ax=plt.gca(),grid=True,alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "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>fr</th>\n",
       "      <th>en</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ingénieur</th>\n",
       "      <td>-15.101917</td>\n",
       "      <td>-13.204992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>médecin</th>\n",
       "      <td>-9.833717</td>\n",
       "      <td>-2.048113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>agriculteur</th>\n",
       "      <td>-11.125644</td>\n",
       "      <td>-11.482672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scientifique</th>\n",
       "      <td>-2.128866</td>\n",
       "      <td>-5.757140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avocat</th>\n",
       "      <td>-12.176149</td>\n",
       "      <td>-4.954153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>politique</th>\n",
       "      <td>-2.446423</td>\n",
       "      <td>-8.848243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ministre</th>\n",
       "      <td>-12.590605</td>\n",
       "      <td>-5.130417</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>banquier</th>\n",
       "      <td>-15.582640</td>\n",
       "      <td>-8.591282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sportif</th>\n",
       "      <td>-1.446035</td>\n",
       "      <td>-22.185427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dentiste</th>\n",
       "      <td>-0.419151</td>\n",
       "      <td>-1.459419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maçon</th>\n",
       "      <td>-13.598932</td>\n",
       "      <td>-15.142560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>secrétaire</th>\n",
       "      <td>-7.782287</td>\n",
       "      <td>6.511085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>infirmier</th>\n",
       "      <td>15.716924</td>\n",
       "      <td>19.471382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>boulanger</th>\n",
       "      <td>-10.493177</td>\n",
       "      <td>3.017117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>journaliste</th>\n",
       "      <td>-3.660803</td>\n",
       "      <td>0.997578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>professeur</th>\n",
       "      <td>-9.206939</td>\n",
       "      <td>3.585324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comptable</th>\n",
       "      <td>-2.621184</td>\n",
       "      <td>-2.444469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sage-femme</th>\n",
       "      <td>14.206563</td>\n",
       "      <td>20.423308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vendeur</th>\n",
       "      <td>0.058117</td>\n",
       "      <td>-20.815334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employé</th>\n",
       "      <td>-1.603209</td>\n",
       "      <td>1.589333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ménage</th>\n",
       "      <td>6.587706</td>\n",
       "      <td>5.359682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mathématique</th>\n",
       "      <td>-1.142150</td>\n",
       "      <td>-3.540002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>science</th>\n",
       "      <td>0.133018</td>\n",
       "      <td>-1.758588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physique</th>\n",
       "      <td>0.437775</td>\n",
       "      <td>-6.704398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informatique</th>\n",
       "      <td>-0.348038</td>\n",
       "      <td>-0.631159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finance</th>\n",
       "      <td>-5.789169</td>\n",
       "      <td>-1.079416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chimie</th>\n",
       "      <td>-1.805138</td>\n",
       "      <td>-2.584284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>histoire</th>\n",
       "      <td>-1.591700</td>\n",
       "      <td>-4.955116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poésie</th>\n",
       "      <td>0.432328</td>\n",
       "      <td>2.410852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peinture</th>\n",
       "      <td>-1.246938</td>\n",
       "      <td>0.055065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cirque</th>\n",
       "      <td>1.430002</td>\n",
       "      <td>-3.950743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>curling</th>\n",
       "      <td>-1.585900</td>\n",
       "      <td>1.879825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>catch</th>\n",
       "      <td>-1.808056</td>\n",
       "      <td>-8.900343</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>équitation</th>\n",
       "      <td>0.711040</td>\n",
       "      <td>-3.627175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>escrime</th>\n",
       "      <td>-0.605165</td>\n",
       "      <td>-4.073707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>golf</th>\n",
       "      <td>-4.493179</td>\n",
       "      <td>-10.976909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>judo</th>\n",
       "      <td>0.815246</td>\n",
       "      <td>-2.653974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>musculation</th>\n",
       "      <td>-0.628242</td>\n",
       "      <td>-0.750730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>polo</th>\n",
       "      <td>-7.194246</td>\n",
       "      <td>-8.060221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tango</th>\n",
       "      <td>3.229821</td>\n",
       "      <td>2.154695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>voile</th>\n",
       "      <td>2.421285</td>\n",
       "      <td>11.786567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yoga</th>\n",
       "      <td>-2.469764</td>\n",
       "      <td>14.362338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuisine</th>\n",
       "      <td>-1.334862</td>\n",
       "      <td>6.234980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maison</th>\n",
       "      <td>4.634189</td>\n",
       "      <td>-0.841578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jardin</th>\n",
       "      <td>1.122685</td>\n",
       "      <td>5.666764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bricolage</th>\n",
       "      <td>-0.173990</td>\n",
       "      <td>5.583880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>magasin</th>\n",
       "      <td>-0.818781</td>\n",
       "      <td>1.845101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shopping</th>\n",
       "      <td>6.214960</td>\n",
       "      <td>12.337984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sac</th>\n",
       "      <td>-0.558901</td>\n",
       "      <td>4.647608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chaussure</th>\n",
       "      <td>0.808932</td>\n",
       "      <td>0.296655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sécurité</th>\n",
       "      <td>-1.662870</td>\n",
       "      <td>-3.739334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rock</th>\n",
       "      <td>-0.300745</td>\n",
       "      <td>-1.372232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jupe</th>\n",
       "      <td>4.037632</td>\n",
       "      <td>18.245031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pantalon</th>\n",
       "      <td>0.374723</td>\n",
       "      <td>-2.317839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>argent</th>\n",
       "      <td>-0.102095</td>\n",
       "      <td>0.482660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>table</th>\n",
       "      <td>0.159110</td>\n",
       "      <td>1.073107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>couteau</th>\n",
       "      <td>-0.879321</td>\n",
       "      <td>-6.012172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>serviette</th>\n",
       "      <td>0.137700</td>\n",
       "      <td>0.845516</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vin</th>\n",
       "      <td>-0.816729</td>\n",
       "      <td>1.573648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bière</th>\n",
       "      <td>1.524435</td>\n",
       "      <td>-10.152493</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>99 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     fr         en\n",
       "ingénieur    -15.101917 -13.204992\n",
       "médecin       -9.833717  -2.048113\n",
       "agriculteur  -11.125644 -11.482672\n",
       "scientifique  -2.128866  -5.757140\n",
       "avocat       -12.176149  -4.954153\n",
       "politique     -2.446423  -8.848243\n",
       "ministre     -12.590605  -5.130417\n",
       "banquier     -15.582640  -8.591282\n",
       "sportif       -1.446035 -22.185427\n",
       "dentiste      -0.419151  -1.459419\n",
       "maçon        -13.598932 -15.142560\n",
       "secrétaire    -7.782287   6.511085\n",
       "infirmier     15.716924  19.471382\n",
       "boulanger    -10.493177   3.017117\n",
       "journaliste   -3.660803   0.997578\n",
       "professeur    -9.206939   3.585324\n",
       "comptable     -2.621184  -2.444469\n",
       "sage-femme    14.206563  20.423308\n",
       "vendeur        0.058117 -20.815334\n",
       "employé       -1.603209   1.589333\n",
       "ménage         6.587706   5.359682\n",
       "mathématique  -1.142150  -3.540002\n",
       "science        0.133018  -1.758588\n",
       "physique       0.437775  -6.704398\n",
       "informatique  -0.348038  -0.631159\n",
       "finance       -5.789169  -1.079416\n",
       "chimie        -1.805138  -2.584284\n",
       "histoire      -1.591700  -4.955116\n",
       "poésie         0.432328   2.410852\n",
       "peinture      -1.246938   0.055065\n",
       "...                 ...        ...\n",
       "cirque         1.430002  -3.950743\n",
       "curling       -1.585900   1.879825\n",
       "catch         -1.808056  -8.900343\n",
       "équitation     0.711040  -3.627175\n",
       "escrime       -0.605165  -4.073707\n",
       "golf          -4.493179 -10.976909\n",
       "judo           0.815246  -2.653974\n",
       "musculation   -0.628242  -0.750730\n",
       "polo          -7.194246  -8.060221\n",
       "tango          3.229821   2.154695\n",
       "voile          2.421285  11.786567\n",
       "yoga          -2.469764  14.362338\n",
       "cuisine       -1.334862   6.234980\n",
       "maison         4.634189  -0.841578\n",
       "jardin         1.122685   5.666764\n",
       "bricolage     -0.173990   5.583880\n",
       "magasin       -0.818781   1.845101\n",
       "shopping       6.214960  12.337984\n",
       "sac           -0.558901   4.647608\n",
       "chaussure      0.808932   0.296655\n",
       "sécurité      -1.662870  -3.739334\n",
       "rock          -0.300745  -1.372232\n",
       "jupe           4.037632  18.245031\n",
       "pantalon       0.374723  -2.317839\n",
       "argent        -0.102095   0.482660\n",
       "table          0.159110   1.073107\n",
       "couteau       -0.879321  -6.012172\n",
       "serviette      0.137700   0.845516\n",
       "vin           -0.816729   1.573648\n",
       "bière          1.524435 -10.152493\n",
       "\n",
       "[99 rows x 2 columns]"
      ]
     },
     "execution_count": 281,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 420,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The direct biais od the 50000 word in the french corpus is  0.015144852274700533\n",
      "The direct biais od the 50000 word in the english corpus is  -0.0044742344877576855\n"
     ]
    }
   ],
   "source": [
    "def Direct_Biais(corpus, Pair, model):\n",
    "    direction_g, expl_var_g,eig_values_g = lib.from_pair_to_direction2(Pair, model)\n",
    "    B=[]\n",
    "    for word in corpus :\n",
    "        B.append(lib.cosine_similarity(direction_g,model[word]))\n",
    "    df_B = pd.DataFrame(data={'Direct Biais':B})\n",
    "    \n",
    "    return(df_B,df_B['Direct Biais'].mean())\n",
    "#\n",
    "#Pair_fr = [['elle','il'],['femme','homme'],['fille','garçon'],['mère','père'],['femelle','mâle'],\n",
    "#           ['la','le'],['reine','roi'],['jument', 'cheval'],['soeur','frère'],['tante','oncle']]\n",
    "\n",
    "df_B_fr,m_fr  = Direct_Biais(list(model_en_wiki.vocab), Pair_en, model_en_wiki)\n",
    "print('The direct biais od the 50000 word in the french corpus is ',m_fr)\n",
    "df_B_en,m_en = Direct_Biais(list(model_en_ff.vocab), Pair_en, model_en_ff)\n",
    "print('The direct biais od the 50000 word in the english corpus is ',m_en)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 421,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,5))\n",
    "plt.subplot(121)\n",
    "df_B_fr.plot(kind='hist',bins = 200,ax=plt.gca(),grid=True)\n",
    "plt.title('Repartition of Direct Biais for french model')\n",
    "plt.subplot(122)\n",
    "df_B_en.plot(kind='hist',bins = 200,ax=plt.gca(),grid=True)\n",
    "plt.title('Repartition of Direct Biais for english model')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 422,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,10))\n",
    "sns.distplot(df_B_fr, bins=250, label = 'fr',kde=True)\n",
    "sns.distplot(df_B_en, bins=250, label = 'en',kde=True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy\n",
    "en =  list(df_B_en.values)\n",
    "fr= list(df_B_fr.values)\n",
    "z,p=scipy.stats.ranksums(en,fr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60.46741417700471, 0.0)"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "z,p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-4.064202474579667, 4.819697446604094e-05)"
      ]
     },
     "execution_count": 331,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "z,p=scipy.stats.ranksums(result['en'],result['fr'])\n",
    "z,p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 332,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.436607632687822\n",
      "0.2982147268891815\n"
     ]
    }
   ],
   "source": [
    "from numpy import std, mean, sqrt\n",
    "def cohen_d(x,y):\n",
    "    nx = len(x)\n",
    "    ny = len(y)\n",
    "    dof = nx + ny - 2\n",
    "    return(mean(x) - mean(y)) / sqrt(((nx-1)*std(x, ddof=1) ** 2 + (ny-1)*std(y, ddof=1) ** 2) / dof)\n",
    "print(cohen_d(result['en'],result['fr']))\n",
    "print(cohen_d(en,fr))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# WEFAT\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "metadata": {},
   "outputs": [],
   "source": [
    "def from_pair_to_attribute(pair):\n",
    "    A = []\n",
    "    B = []\n",
    "    for i in range(len(pair)):\n",
    "        A.append(pair[i][1])\n",
    "        B.append(pair[i][0])\n",
    "    return([A,B])\n",
    "\n",
    "att_fr = from_pair_to_attribute(Pair_fr)\n",
    "att_en = from_pair_to_attribute(Pair_en)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 385,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.696969696969697\n"
     ]
    }
   ],
   "source": [
    "def result(att_fr,att_en,model_fr,model_en):\n",
    "    Biais_fr = []\n",
    "    Biais_en = []\n",
    "    for wf,we in zip(test_fr,test_en):\n",
    "        Biais_fr.append((lib.s(wf,att_fr[0],att_fr[1],model_fr)+0.003)*100)\n",
    "        Biais_en.append(lib.s(we,att_en[0],att_en[1],model_en)*100)\n",
    "    \n",
    "    result = pd.DataFrame(data={'fr':Biais_fr,'en':Biais_en},index=test_fr)\n",
    "    print(len(result[np.sign(result['en'])==np.sign(result['fr'])])/len(result))\n",
    "    return(result)\n",
    "\n",
    "result = result(att_fr,att_en,model_wiki_lem,model_en_ff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 388,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.distplot(result['en'], bins=70, label = 'en',kde=True)\n",
    "sns.distplot(result['fr'], bins=70, label = 'fr',kde=True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2c22d09f898>"
      ]
     },
     "execution_count": 350,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result.plot(kind='bar',figsize = (20,10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 352,
   "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>fr</th>\n",
       "      <th>en</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ingénieur</th>\n",
       "      <td>-10.496599</td>\n",
       "      <td>-7.292019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>médecin</th>\n",
       "      <td>-7.515840</td>\n",
       "      <td>-1.459963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>agriculteur</th>\n",
       "      <td>-8.303619</td>\n",
       "      <td>-6.729674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>scientifique</th>\n",
       "      <td>-1.834142</td>\n",
       "      <td>-2.865345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avocat</th>\n",
       "      <td>-8.603837</td>\n",
       "      <td>-3.265357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>politique</th>\n",
       "      <td>-2.674449</td>\n",
       "      <td>-5.042741</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ministre</th>\n",
       "      <td>-7.980921</td>\n",
       "      <td>-2.565869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>banquier</th>\n",
       "      <td>-9.625424</td>\n",
       "      <td>-5.116218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sportif</th>\n",
       "      <td>-2.092229</td>\n",
       "      <td>-13.108912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dentiste</th>\n",
       "      <td>-0.706403</td>\n",
       "      <td>-0.925478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maçon</th>\n",
       "      <td>-9.519838</td>\n",
       "      <td>-8.775346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>secrétaire</th>\n",
       "      <td>-4.875497</td>\n",
       "      <td>4.213884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>infirmier</th>\n",
       "      <td>9.904173</td>\n",
       "      <td>11.512408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>boulanger</th>\n",
       "      <td>-7.144526</td>\n",
       "      <td>1.972618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>journaliste</th>\n",
       "      <td>-3.134659</td>\n",
       "      <td>1.005381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>professeur</th>\n",
       "      <td>-6.525178</td>\n",
       "      <td>1.985480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comptable</th>\n",
       "      <td>-2.385847</td>\n",
       "      <td>-1.200846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sage-femme</th>\n",
       "      <td>9.221149</td>\n",
       "      <td>11.773248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vendeur</th>\n",
       "      <td>-0.873697</td>\n",
       "      <td>-11.839924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employé</th>\n",
       "      <td>-2.902588</td>\n",
       "      <td>1.576270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ménage</th>\n",
       "      <td>3.259402</td>\n",
       "      <td>3.122715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mathématique</th>\n",
       "      <td>-0.835066</td>\n",
       "      <td>-1.902082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>science</th>\n",
       "      <td>-0.411509</td>\n",
       "      <td>-0.782505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physique</th>\n",
       "      <td>0.000236</td>\n",
       "      <td>-3.726605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informatique</th>\n",
       "      <td>-0.282358</td>\n",
       "      <td>-0.774564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>finance</th>\n",
       "      <td>-4.258355</td>\n",
       "      <td>-0.342521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chimie</th>\n",
       "      <td>-1.098793</td>\n",
       "      <td>-1.317772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>histoire</th>\n",
       "      <td>-1.274924</td>\n",
       "      <td>-2.981467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poésie</th>\n",
       "      <td>-0.078059</td>\n",
       "      <td>1.547720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peinture</th>\n",
       "      <td>-0.643594</td>\n",
       "      <td>0.041083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cirque</th>\n",
       "      <td>0.960815</td>\n",
       "      <td>-2.405485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>curling</th>\n",
       "      <td>-1.469130</td>\n",
       "      <td>0.656473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>catch</th>\n",
       "      <td>-1.599311</td>\n",
       "      <td>-5.296403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>équitation</th>\n",
       "      <td>-0.409235</td>\n",
       "      <td>-2.347498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>escrime</th>\n",
       "      <td>-0.802625</td>\n",
       "      <td>-3.215626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>golf</th>\n",
       "      <td>-3.822528</td>\n",
       "      <td>-6.290050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>judo</th>\n",
       "      <td>-0.127481</td>\n",
       "      <td>-2.042955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>musculation</th>\n",
       "      <td>-0.938529</td>\n",
       "      <td>-0.282723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>polo</th>\n",
       "      <td>-5.515223</td>\n",
       "      <td>-5.670493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tango</th>\n",
       "      <td>1.807810</td>\n",
       "      <td>1.084191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>voile</th>\n",
       "      <td>1.052649</td>\n",
       "      <td>7.073573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yoga</th>\n",
       "      <td>-2.058303</td>\n",
       "      <td>8.355502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuisine</th>\n",
       "      <td>-1.953737</td>\n",
       "      <td>3.642878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maison</th>\n",
       "      <td>3.511948</td>\n",
       "      <td>-0.870962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jardin</th>\n",
       "      <td>0.322140</td>\n",
       "      <td>3.352022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bricolage</th>\n",
       "      <td>-0.985828</td>\n",
       "      <td>3.464979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>magasin</th>\n",
       "      <td>-1.273764</td>\n",
       "      <td>1.530419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>shopping</th>\n",
       "      <td>3.832000</td>\n",
       "      <td>7.130126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sac</th>\n",
       "      <td>-0.263557</td>\n",
       "      <td>2.903395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chaussure</th>\n",
       "      <td>0.191086</td>\n",
       "      <td>0.174761</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sécurité</th>\n",
       "      <td>-1.389807</td>\n",
       "      <td>-2.184001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rock</th>\n",
       "      <td>-0.549941</td>\n",
       "      <td>-0.819107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jupe</th>\n",
       "      <td>2.696635</td>\n",
       "      <td>11.182336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pantalon</th>\n",
       "      <td>0.207962</td>\n",
       "      <td>-1.389910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>argent</th>\n",
       "      <td>-0.237517</td>\n",
       "      <td>0.043919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>table</th>\n",
       "      <td>-0.366538</td>\n",
       "      <td>0.970131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>couteau</th>\n",
       "      <td>-0.894147</td>\n",
       "      <td>-3.646948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>serviette</th>\n",
       "      <td>-0.155255</td>\n",
       "      <td>0.647838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vin</th>\n",
       "      <td>-0.775499</td>\n",
       "      <td>1.272379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bière</th>\n",
       "      <td>0.722999</td>\n",
       "      <td>-5.707219</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>99 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     fr         en\n",
       "ingénieur    -10.496599  -7.292019\n",
       "médecin       -7.515840  -1.459963\n",
       "agriculteur   -8.303619  -6.729674\n",
       "scientifique  -1.834142  -2.865345\n",
       "avocat        -8.603837  -3.265357\n",
       "politique     -2.674449  -5.042741\n",
       "ministre      -7.980921  -2.565869\n",
       "banquier      -9.625424  -5.116218\n",
       "sportif       -2.092229 -13.108912\n",
       "dentiste      -0.706403  -0.925478\n",
       "maçon         -9.519838  -8.775346\n",
       "secrétaire    -4.875497   4.213884\n",
       "infirmier      9.904173  11.512408\n",
       "boulanger     -7.144526   1.972618\n",
       "journaliste   -3.134659   1.005381\n",
       "professeur    -6.525178   1.985480\n",
       "comptable     -2.385847  -1.200846\n",
       "sage-femme     9.221149  11.773248\n",
       "vendeur       -0.873697 -11.839924\n",
       "employé       -2.902588   1.576270\n",
       "ménage         3.259402   3.122715\n",
       "mathématique  -0.835066  -1.902082\n",
       "science       -0.411509  -0.782505\n",
       "physique       0.000236  -3.726605\n",
       "informatique  -0.282358  -0.774564\n",
       "finance       -4.258355  -0.342521\n",
       "chimie        -1.098793  -1.317772\n",
       "histoire      -1.274924  -2.981467\n",
       "poésie        -0.078059   1.547720\n",
       "peinture      -0.643594   0.041083\n",
       "...                 ...        ...\n",
       "cirque         0.960815  -2.405485\n",
       "curling       -1.469130   0.656473\n",
       "catch         -1.599311  -5.296403\n",
       "équitation    -0.409235  -2.347498\n",
       "escrime       -0.802625  -3.215626\n",
       "golf          -3.822528  -6.290050\n",
       "judo          -0.127481  -2.042955\n",
       "musculation   -0.938529  -0.282723\n",
       "polo          -5.515223  -5.670493\n",
       "tango          1.807810   1.084191\n",
       "voile          1.052649   7.073573\n",
       "yoga          -2.058303   8.355502\n",
       "cuisine       -1.953737   3.642878\n",
       "maison         3.511948  -0.870962\n",
       "jardin         0.322140   3.352022\n",
       "bricolage     -0.985828   3.464979\n",
       "magasin       -1.273764   1.530419\n",
       "shopping       3.832000   7.130126\n",
       "sac           -0.263557   2.903395\n",
       "chaussure      0.191086   0.174761\n",
       "sécurité      -1.389807  -2.184001\n",
       "rock          -0.549941  -0.819107\n",
       "jupe           2.696635  11.182336\n",
       "pantalon       0.207962  -1.389910\n",
       "argent        -0.237517   0.043919\n",
       "table         -0.366538   0.970131\n",
       "couteau       -0.894147  -3.646948\n",
       "serviette     -0.155255   0.647838\n",
       "vin           -0.775499   1.272379\n",
       "bière          0.722999  -5.707219\n",
       "\n",
       "[99 rows x 2 columns]"
      ]
     },
     "execution_count": 352,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 368,
   "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>fr</th>\n",
       "      <th>en</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>secrétaire</th>\n",
       "      <td>-4.875497</td>\n",
       "      <td>4.213884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>boulanger</th>\n",
       "      <td>-7.144526</td>\n",
       "      <td>1.972618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>journaliste</th>\n",
       "      <td>-3.134659</td>\n",
       "      <td>1.005381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>professeur</th>\n",
       "      <td>-6.525178</td>\n",
       "      <td>1.985480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>employé</th>\n",
       "      <td>-2.902588</td>\n",
       "      <td>1.576270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>physique</th>\n",
       "      <td>0.000236</td>\n",
       "      <td>-3.726605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poésie</th>\n",
       "      <td>-0.078059</td>\n",
       "      <td>1.547720</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peinture</th>\n",
       "      <td>-0.643594</td>\n",
       "      <td>0.041083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>art</th>\n",
       "      <td>-1.731530</td>\n",
       "      <td>2.116745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chant</th>\n",
       "      <td>-2.525896</td>\n",
       "      <td>1.137007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>littérature</th>\n",
       "      <td>-0.632948</td>\n",
       "      <td>1.246928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>production</th>\n",
       "      <td>0.012924</td>\n",
       "      <td>-1.500099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>architecture</th>\n",
       "      <td>0.130277</td>\n",
       "      <td>-0.642771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>théâtre</th>\n",
       "      <td>-1.421326</td>\n",
       "      <td>2.200979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tennis</th>\n",
       "      <td>-1.245610</td>\n",
       "      <td>1.114684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ping-pong</th>\n",
       "      <td>-1.815173</td>\n",
       "      <td>9.831171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>kayak</th>\n",
       "      <td>-0.293791</td>\n",
       "      <td>0.894484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>badminton</th>\n",
       "      <td>-0.859103</td>\n",
       "      <td>1.422842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cirque</th>\n",
       "      <td>0.960815</td>\n",
       "      <td>-2.405485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>curling</th>\n",
       "      <td>-1.469130</td>\n",
       "      <td>0.656473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>yoga</th>\n",
       "      <td>-2.058303</td>\n",
       "      <td>8.355502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cuisine</th>\n",
       "      <td>-1.953737</td>\n",
       "      <td>3.642878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>maison</th>\n",
       "      <td>3.511948</td>\n",
       "      <td>-0.870962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bricolage</th>\n",
       "      <td>-0.985828</td>\n",
       "      <td>3.464979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>magasin</th>\n",
       "      <td>-1.273764</td>\n",
       "      <td>1.530419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sac</th>\n",
       "      <td>-0.263557</td>\n",
       "      <td>2.903395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pantalon</th>\n",
       "      <td>0.207962</td>\n",
       "      <td>-1.389910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>argent</th>\n",
       "      <td>-0.237517</td>\n",
       "      <td>0.043919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>table</th>\n",
       "      <td>-0.366538</td>\n",
       "      <td>0.970131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>serviette</th>\n",
       "      <td>-0.155255</td>\n",
       "      <td>0.647838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>vin</th>\n",
       "      <td>-0.775499</td>\n",
       "      <td>1.272379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bière</th>\n",
       "      <td>0.722999</td>\n",
       "      <td>-5.707219</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    fr        en\n",
       "secrétaire   -4.875497  4.213884\n",
       "boulanger    -7.144526  1.972618\n",
       "journaliste  -3.134659  1.005381\n",
       "professeur   -6.525178  1.985480\n",
       "employé      -2.902588  1.576270\n",
       "physique      0.000236 -3.726605\n",
       "poésie       -0.078059  1.547720\n",
       "peinture     -0.643594  0.041083\n",
       "art          -1.731530  2.116745\n",
       "chant        -2.525896  1.137007\n",
       "littérature  -0.632948  1.246928\n",
       "production    0.012924 -1.500099\n",
       "architecture  0.130277 -0.642771\n",
       "théâtre      -1.421326  2.200979\n",
       "tennis       -1.245610  1.114684\n",
       "ping-pong    -1.815173  9.831171\n",
       "kayak        -0.293791  0.894484\n",
       "badminton    -0.859103  1.422842\n",
       "cirque        0.960815 -2.405485\n",
       "curling      -1.469130  0.656473\n",
       "yoga         -2.058303  8.355502\n",
       "cuisine      -1.953737  3.642878\n",
       "maison        3.511948 -0.870962\n",
       "bricolage    -0.985828  3.464979\n",
       "magasin      -1.273764  1.530419\n",
       "sac          -0.263557  2.903395\n",
       "pantalon      0.207962 -1.389910\n",
       "argent       -0.237517  0.043919\n",
       "table        -0.366538  0.970131\n",
       "serviette    -0.155255  0.647838\n",
       "vin          -0.775499  1.272379\n",
       "bière         0.722999 -5.707219"
      ]
     },
     "execution_count": 368,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-result[(np.sign(result['en'])!=np.sign(result['fr']))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 403,
   "metadata": {},
   "outputs": [],
   "source": [
    "final = pd.DataFrame(data = {'fr_db': result_bd['fr'].values,'en_db': result_bd['en'].values,'fr_wefat': result['fr'],'en_wefat': result['en']},index=test_fr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 412,
   "metadata": {},
   "outputs": [],
   "source": [
    "final = final.apply(lambda x:round(x,2))\n",
    "final.to_csv('final_biais.csv',encoding='latin1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}