diff --git a/notebooks/Classification_Zero-Shot-Learning.ipynb b/notebooks/Classification_Zero-Shot-Learning.ipynb index 1ea10781cb40a4ed28467330fbc236a9e630951b..f6050c8796de4ab7eb85f1555878dc783f30c2c4 100644 --- a/notebooks/Classification_Zero-Shot-Learning.ipynb +++ b/notebooks/Classification_Zero-Shot-Learning.ipynb @@ -583,7 +583,7 @@ }, { "cell_type": "code", - "execution_count": 37, + "execution_count": 48, "metadata": {}, "outputs": [], "source": [ @@ -611,11 +611,7 @@ "\n", "\n", "def get_highest_score(true_probs, hypotheses):\n", - " # print the true probability for each hypothesis\n", - " #for i, hypothesis in enumerate(hypotheses):\n", - " # print(f'Probability that hypothesis \"{hypothesis}\" is true: {true_probs[i]:0.2f}%')\n", - " # print(f'Probability that the label is true: {true_prob:0.2f}%')\n", - "\n", + " \n", " # get index of hypothesis with highest score\n", " highest_index = max(range(len(true_probs)), key=lambda i: true_probs[i])\n", "\n", @@ -626,21 +622,43 @@ " highest_prob = true_probs[highest_index]\n", " \n", " return (highest_hypothesis, highest_prob)\n", + "\n", + "\n", + "def get_sorted_scores(true_probs, hypotheses):\n", + "\n", + " # sort hypotheses based on their scores\n", + " sorted_hypotheses = [hypothesis for _, hypothesis in sorted(zip(true_probs, hypotheses), reverse=True)]\n", + "\n", + " # sort scores\n", + " sorted_scores = sorted(true_probs, reverse=True)\n", + " \n", + " return list(zip(sorted_hypotheses, sorted_scores))\n", " " ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 49, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/plain": [ + "\"ORNIS, s. m. toile des Indes, (Comm.) sortes de\\ntoiles de coton ou de mousseline, qui se font a Brampour ville de l'Indoustan, entre Surate & Agra. Ces\\ntoiles sont par bandes, moitié coton & moitié or &\\nargent. Il y en a depuis quinze jusqu'à vingt aunes.\"" + ] + }, + "execution_count": 49, + "metadata": {}, + "output_type": "execute_result" + } + ], "source": [ "df[column_text].tolist()[0]" ] }, { "cell_type": "code", - "execution_count": 38, + "execution_count": 50, "metadata": {}, "outputs": [ { @@ -649,6 +667,53 @@ "text": [ "The hypothesis with the highest score is: \"Commerce\" with a probability of 70.05%\n" ] + }, + { + "data": { + "text/plain": [ + "[('Commerce', 70.05109786987305),\n", + " ('Anatomie', 68.73835921287537),\n", + " ('Politique', 60.711854696273804),\n", + " ('Géographie', 59.15616154670715),\n", + " ('Architecture', 58.7415874004364),\n", + " ('Histoire', 57.4596107006073),\n", + " ('Agriculture - Economie rustique', 53.530728816986084),\n", + " ('Histoire naturelle', 48.45908284187317),\n", + " ('Antiquité', 46.68468534946442),\n", + " ('Beaux-arts', 42.85620450973511),\n", + " ('Mesure', 41.310253739356995),\n", + " ('Jeu', 41.221246123313904),\n", + " ('Droit - Jurisprudence', 41.13341271877289),\n", + " ('Minéralogie', 38.13730478286743),\n", + " ('Spectacle', 37.80337870121002),\n", + " ('Pêche', 37.214699387550354),\n", + " ('Superstition', 36.72800958156586),\n", + " ('Arts et métiers', 36.512187123298645),\n", + " ('Métiers', 36.505624651908875),\n", + " ('Monnaie', 35.89847385883331),\n", + " ('Musique', 32.74992108345032),\n", + " ('Mathématiques', 32.70103633403778),\n", + " ('Chasse', 29.351988434791565),\n", + " ('Economie domestique', 28.34642231464386),\n", + " ('Philosophie', 27.653351426124573),\n", + " ('Chimie', 25.783848762512207),\n", + " ('Physique - [Sciences physico-mathématiques]', 25.403589010238647),\n", + " ('Médailles', 24.586908519268036),\n", + " ('Grammaire', 22.362521290779114),\n", + " ('Caractères', 20.14842927455902),\n", + " ('Pharmacie', 19.72046047449112),\n", + " ('Militaire (Art) - Guerre - Arme', 19.68255490064621),\n", + " ('Médecine - Chirurgie', 18.616029620170593),\n", + " ('Marine', 18.207980692386627),\n", + " ('Belles-lettres - Poésie', 13.30692172050476),\n", + " ('Blason', 10.476682335138321),\n", + " ('Religion', 9.701979160308838),\n", + " ('Maréchage - Manège', 4.211422428488731)]" + ] + }, + "execution_count": 50, + "metadata": {}, + "output_type": "execute_result" } ], "source": [ @@ -658,7 +723,1068 @@ "highest_score = get_highest_score(true_probs, classes)\n", "\n", "# print the results\n", - "print(f'The hypothesis with the highest score is: \"{highest_score[0]}\" with a probability of {highest_score[1]:0.2f}%')" + "print(f'The hypothesis with the highest score is: \"{highest_score[0]}\" with a probability of {highest_score[1]:0.2f}%')\n", + "\n", + "\n", + "probs = get_sorted_scores(true_probs, classes)\n", + "probs\n" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'Commerce'" + ] + }, + "execution_count": 51, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "probs[0][0]" + ] + }, + { + "cell_type": "code", + "execution_count": 52, + "metadata": {}, + "outputs": [], + "source": [ + "pred_labels = []\n", + "prob_labels = []\n", + "\n", + "for content in df[column_text].tolist():\n", + "\n", + " true_probs = zero_shot_prediction(content[:1024], classes)\n", + " \n", + " pred_labels.append(get_highest_score(true_probs, classes)[0])\n", + " prob_labels.append(get_sorted_scores(true_probs, classes))\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "['Commerce',\n", + " 'Marine',\n", + " 'Géographie',\n", + " 'Histoire',\n", + " 'Histoire',\n", + " 'Marine',\n", + " 'Belles-lettres - Poésie',\n", + " 'Economie domestique',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Beaux-arts',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Histoire naturelle',\n", + " 'Grammaire',\n", + " 'Philosophie',\n", + " 'Histoire',\n", + " 'Médecine - Chirurgie',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Arts et métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Antiquité',\n", + " 'Histoire naturelle',\n", + " 'Histoire naturelle',\n", + " 'Pharmacie',\n", + " 'Grammaire',\n", + " 'Religion',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Médecine - Chirurgie',\n", + " 'Pêche',\n", + " 'Anatomie',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Musique',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Belles-lettres - Poésie',\n", + " 'Commerce',\n", + " 'Jeu',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Caractères',\n", + " 'Métiers',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Histoire naturelle',\n", + " 'Architecture',\n", + " 'Droit - Jurisprudence',\n", + " 'Belles-lettres - Poésie',\n", + " 'Métiers',\n", + " 'Histoire',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Maréchage - Manège',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Grammaire',\n", + " 'Grammaire',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Architecture',\n", + " 'Grammaire',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Chimie',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Blason',\n", + " 'Géographie',\n", + " 'Blason',\n", + " 'Religion',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Grammaire',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Métiers',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Antiquité',\n", + " 'Philosophie',\n", + " 'Grammaire',\n", + " 'Commerce',\n", + " 'Commerce',\n", + " 'Grammaire',\n", + " 'Médecine - Chirurgie',\n", + " 'Métiers',\n", + " 'Chasse',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Mathématiques',\n", + " 'Histoire naturelle',\n", + " 'Marine',\n", + " 'Architecture',\n", + " 'Philosophie',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Architecture',\n", + " 'Métiers',\n", + " 'Histoire',\n", + " 'Antiquité',\n", + " 'Commerce',\n", + " 'Histoire',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Philosophie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Architecture',\n", + " 'Commerce',\n", + " 'Architecture',\n", + " 'Droit - Jurisprudence',\n", + " 'Commerce',\n", + " 'Histoire naturelle',\n", + " 'Histoire naturelle',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Médailles',\n", + " 'Marine',\n", + " 'Superstition',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Marine',\n", + " 'Antiquité',\n", + " 'Histoire naturelle',\n", + " 'Religion',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Histoire naturelle',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Marine',\n", + " 'Grammaire',\n", + " 'Médecine - Chirurgie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Architecture',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Beaux-arts',\n", + " 'Histoire naturelle',\n", + " 'Chasse',\n", + " 'Grammaire',\n", + " 'Métiers',\n", + " 'Agriculture - Economie rustique',\n", + " 'Marine',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Commerce',\n", + " 'Grammaire',\n", + " 'Architecture',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Chasse',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Commerce',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Commerce',\n", + " 'Belles-lettres - Poésie',\n", + " 'Belles-lettres - Poésie',\n", + " 'Histoire naturelle',\n", + " 'Commerce',\n", + " 'Commerce',\n", + " 'Métiers',\n", + " 'Marine',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Grammaire',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Belles-lettres - Poésie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Blason',\n", + " 'Anatomie',\n", + " 'Architecture',\n", + " 'Histoire naturelle',\n", + " 'Musique',\n", + " 'Droit - Jurisprudence',\n", + " 'Commerce',\n", + " 'Histoire',\n", + " 'Pharmacie',\n", + " 'Marine',\n", + " 'Commerce',\n", + " 'Histoire naturelle',\n", + " 'Jeu',\n", + " 'Droit - Jurisprudence',\n", + " 'Grammaire',\n", + " 'Histoire',\n", + " 'Géographie',\n", + " 'Architecture',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Mathématiques',\n", + " 'Médecine - Chirurgie',\n", + " 'Religion',\n", + " 'Chimie',\n", + " 'Blason',\n", + " 'Antiquité',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Belles-lettres - Poésie',\n", + " 'Métiers',\n", + " 'Religion',\n", + " 'Droit - Jurisprudence',\n", + " 'Arts et métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Histoire naturelle',\n", + " 'Beaux-arts',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Economie domestique',\n", + " 'Blason',\n", + " 'Architecture',\n", + " 'Maréchage - Manège',\n", + " 'Pharmacie',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Marine',\n", + " 'Histoire naturelle',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Belles-lettres - Poésie',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Histoire',\n", + " 'Chimie',\n", + " 'Antiquité',\n", + " 'Religion',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Commerce',\n", + " 'Commerce',\n", + " 'Architecture',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Mesure',\n", + " 'Droit - Jurisprudence',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Marine',\n", + " 'Arts et métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Architecture',\n", + " 'Histoire naturelle',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Belles-lettres - Poésie',\n", + " 'Droit - Jurisprudence',\n", + " 'Marine',\n", + " 'Architecture',\n", + " 'Agriculture - Economie rustique',\n", + " 'Histoire naturelle',\n", + " 'Métiers',\n", + " 'Mathématiques',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire',\n", + " 'Musique',\n", + " 'Chasse',\n", + " 'Médecine - Chirurgie',\n", + " 'Antiquité',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Métiers',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Chasse',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Grammaire',\n", + " 'Grammaire',\n", + " 'Médecine - Chirurgie',\n", + " 'Mathématiques',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Mathématiques',\n", + " 'Histoire',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Grammaire',\n", + " 'Marine',\n", + " 'Marine',\n", + " 'Blason',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Histoire',\n", + " 'Marine',\n", + " 'Marine',\n", + " 'Métiers',\n", + " 'Belles-lettres - Poésie',\n", + " 'Religion',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Musique',\n", + " 'Marine',\n", + " 'Marine',\n", + " 'Maréchage - Manège',\n", + " 'Religion',\n", + " 'Géographie',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Belles-lettres - Poésie',\n", + " 'Grammaire',\n", + " 'Histoire',\n", + " 'Géographie',\n", + " 'Chasse',\n", + " 'Métiers',\n", + " 'Histoire',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Antiquité',\n", + " 'Monnaie',\n", + " 'Anatomie',\n", + " 'Métiers',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Anatomie',\n", + " 'Histoire naturelle',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Jeu',\n", + " 'Agriculture - Economie rustique',\n", + " 'Commerce',\n", + " 'Philosophie',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Agriculture - Economie rustique',\n", + " 'Histoire naturelle',\n", + " 'Antiquité',\n", + " 'Histoire naturelle',\n", + " 'Métiers',\n", + " 'Métiers',\n", + " 'Arts et métiers',\n", + " 'Architecture',\n", + " 'Métiers',\n", + " 'Religion',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Histoire',\n", + " 'Mathématiques',\n", + " 'Pêche',\n", + " 'Religion',\n", + " 'Commerce',\n", + " 'Histoire',\n", + " 'Commerce',\n", + " 'Métiers',\n", + " 'Religion',\n", + " 'Religion',\n", + " 'Médecine - Chirurgie',\n", + " 'Métiers',\n", + " 'Musique',\n", + " 'Monnaie',\n", + " 'Antiquité',\n", + " 'Grammaire',\n", + " 'Anatomie',\n", + " 'Marine',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Anatomie',\n", + " 'Métiers',\n", + " 'Métiers',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Architecture',\n", + " 'Belles-lettres - Poésie',\n", + " 'Droit - Jurisprudence',\n", + " 'Arts et métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Mathématiques',\n", + " 'Beaux-arts',\n", + " 'Histoire',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Histoire naturelle',\n", + " 'Minéralogie',\n", + " 'Maréchage - Manège',\n", + " 'Métiers',\n", + " 'Blason',\n", + " 'Marine',\n", + " 'Droit - Jurisprudence',\n", + " 'Maréchage - Manège',\n", + " 'Caractères',\n", + " 'Métiers',\n", + " 'Histoire',\n", + " 'Droit - Jurisprudence',\n", + " 'Philosophie',\n", + " 'Anatomie',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Religion',\n", + " 'Arts et métiers',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Philosophie',\n", + " 'Agriculture - Economie rustique',\n", + " 'Commerce',\n", + " 'Histoire',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Musique',\n", + " 'Histoire naturelle',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Histoire naturelle',\n", + " 'Arts et métiers',\n", + " 'Maréchage - Manège',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Architecture',\n", + " 'Droit - Jurisprudence',\n", + " 'Architecture',\n", + " 'Commerce',\n", + " 'Musique',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Politique',\n", + " 'Grammaire',\n", + " 'Commerce',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Médecine - Chirurgie',\n", + " 'Métiers',\n", + " 'Marine',\n", + " 'Histoire',\n", + " 'Histoire',\n", + " 'Géographie',\n", + " 'Monnaie',\n", + " 'Agriculture - Economie rustique',\n", + " 'Métiers',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Histoire naturelle',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Anatomie',\n", + " 'Anatomie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Chimie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Grammaire',\n", + " 'Pharmacie',\n", + " 'Grammaire',\n", + " 'Métiers',\n", + " 'Métiers',\n", + " 'Histoire naturelle',\n", + " 'Arts et métiers',\n", + " 'Antiquité',\n", + " 'Pharmacie',\n", + " 'Histoire naturelle',\n", + " 'Marine',\n", + " 'Religion',\n", + " 'Histoire',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Maréchage - Manège',\n", + " 'Jeu',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Agriculture - Economie rustique',\n", + " 'Antiquité',\n", + " 'Médecine - Chirurgie',\n", + " 'Chimie',\n", + " 'Métiers',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Politique',\n", + " 'Histoire',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Grammaire',\n", + " 'Marine',\n", + " 'Métiers',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Marine',\n", + " 'Commerce',\n", + " 'Grammaire',\n", + " 'Histoire naturelle',\n", + " 'Histoire naturelle',\n", + " 'Maréchage - Manège',\n", + " 'Métiers',\n", + " 'Grammaire',\n", + " 'Métiers',\n", + " 'Métiers',\n", + " 'Histoire naturelle',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Musique',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Commerce',\n", + " 'Métiers',\n", + " 'Marine',\n", + " 'Marine',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Marine',\n", + " 'Religion',\n", + " 'Droit - Jurisprudence',\n", + " 'Musique',\n", + " 'Histoire naturelle',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Mesure',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Musique',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Histoire',\n", + " 'Médecine - Chirurgie',\n", + " 'Blason',\n", + " 'Belles-lettres - Poésie',\n", + " 'Métiers',\n", + " 'Economie domestique',\n", + " 'Géographie',\n", + " 'Mathématiques',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Chimie',\n", + " 'Géographie',\n", + " 'Belles-lettres - Poésie',\n", + " 'Marine',\n", + " 'Anatomie',\n", + " 'Commerce',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Métiers',\n", + " 'Marine',\n", + " 'Géographie',\n", + " 'Histoire',\n", + " 'Métiers',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Belles-lettres - Poésie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Histoire',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Agriculture - Economie rustique',\n", + " 'Histoire naturelle',\n", + " 'Médecine - Chirurgie',\n", + " 'Architecture',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Commerce',\n", + " 'Antiquité',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Religion',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Architecture',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Histoire',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Jeu',\n", + " 'Marine',\n", + " 'Métiers',\n", + " 'Marine',\n", + " 'Architecture',\n", + " 'Architecture',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Musique',\n", + " 'Métiers',\n", + " 'Histoire naturelle',\n", + " 'Antiquité',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Marine',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Marine',\n", + " 'Géographie',\n", + " 'Architecture',\n", + " 'Métiers',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Histoire naturelle',\n", + " 'Marine',\n", + " 'Métiers',\n", + " 'Pêche',\n", + " 'Droit - Jurisprudence',\n", + " 'Marine',\n", + " 'Histoire naturelle',\n", + " 'Droit - Jurisprudence',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Agriculture - Economie rustique',\n", + " 'Histoire naturelle',\n", + " 'Grammaire',\n", + " 'Architecture',\n", + " 'Anatomie',\n", + " 'Commerce',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Histoire naturelle',\n", + " 'Antiquité',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Droit - Jurisprudence',\n", + " 'Chasse',\n", + " 'Métiers',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire naturelle',\n", + " 'Métiers',\n", + " 'Grammaire',\n", + " 'Architecture',\n", + " 'Philosophie',\n", + " 'Géographie',\n", + " 'Chasse',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Métiers',\n", + " 'Histoire naturelle',\n", + " 'Antiquité',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire naturelle',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Antiquité',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Chasse',\n", + " 'Architecture',\n", + " 'Histoire',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Histoire',\n", + " 'Droit - Jurisprudence',\n", + " 'Marine',\n", + " 'Agriculture - Economie rustique',\n", + " 'Architecture',\n", + " 'Commerce',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Anatomie',\n", + " 'Histoire',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Médecine - Chirurgie',\n", + " 'Histoire naturelle',\n", + " 'Architecture',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Marine',\n", + " 'Antiquité',\n", + " 'Géographie',\n", + " 'Marine',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Géographie',\n", + " 'Architecture',\n", + " 'Géographie',\n", + " 'Chimie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Mathématiques',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Médecine - Chirurgie',\n", + " 'Chasse',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Commerce',\n", + " 'Religion',\n", + " 'Agriculture - Economie rustique',\n", + " 'Marine',\n", + " 'Commerce',\n", + " 'Anatomie',\n", + " 'Anatomie',\n", + " 'Agriculture - Economie rustique',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Musique',\n", + " 'Architecture',\n", + " 'Marine',\n", + " 'Histoire',\n", + " 'Géographie',\n", + " 'Grammaire',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Superstition',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Histoire naturelle',\n", + " 'Agriculture - Economie rustique',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire naturelle',\n", + " 'Métiers',\n", + " 'Pharmacie',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Droit - Jurisprudence',\n", + " 'Grammaire',\n", + " 'Histoire',\n", + " 'Agriculture - Economie rustique',\n", + " 'Droit - Jurisprudence',\n", + " 'Histoire naturelle',\n", + " 'Histoire',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Histoire',\n", + " 'Histoire',\n", + " 'Belles-lettres - Poésie',\n", + " 'Histoire',\n", + " 'Architecture',\n", + " 'Physique - [Sciences physico-mathématiques]',\n", + " 'Philosophie',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Religion',\n", + " 'Droit - Jurisprudence',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Antiquité',\n", + " 'Grammaire',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Marine',\n", + " 'Mathématiques',\n", + " 'Antiquité',\n", + " 'Caractères',\n", + " 'Commerce',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Géographie',\n", + " 'Métiers',\n", + " 'Médecine - Chirurgie',\n", + " 'Belles-lettres - Poésie',\n", + " 'Métiers',\n", + " 'Médecine - Chirurgie',\n", + " 'Droit - Jurisprudence',\n", + " 'Antiquité',\n", + " 'Maréchage - Manège',\n", + " 'Economie domestique',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Agriculture - Economie rustique',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Commerce',\n", + " 'Beaux-arts',\n", + " 'Mathématiques',\n", + " 'Géographie',\n", + " 'Antiquité',\n", + " 'Géographie',\n", + " 'Histoire naturelle',\n", + " 'Médecine - Chirurgie',\n", + " 'Géographie',\n", + " 'Droit - Jurisprudence',\n", + " 'Droit - Jurisprudence',\n", + " 'Métiers',\n", + " 'Militaire (Art) - Guerre - Arme',\n", + " 'Chasse',\n", + " 'Marine',\n", + " 'Géographie',\n", + " 'Agriculture - Economie rustique',\n", + " 'Métiers',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " 'Géographie',\n", + " ...]" + ] + }, + "execution_count": 40, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "true_labels = df[column_class].tolist()" ] }, {