diff --git a/notebooks/Classification_Zero-Shot-Learning.ipynb b/notebooks/Classification_Zero-Shot-Learning.ipynb
index 4a7ab888a7f66428cc6e486d2b1668d3636fa511..ad73e6a87ccb0cc4f6fc2f22426c6ee08e68e004 100644
--- a/notebooks/Classification_Zero-Shot-Learning.ipynb
+++ b/notebooks/Classification_Zero-Shot-Learning.ipynb
@@ -143,7 +143,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
@@ -158,7 +158,7 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 3,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
@@ -325,7 +325,7 @@
"4 INSPECTEUR, s. m. inspector ; () celui \\nà qui... 102 "
]
},
- "execution_count": 18,
+ "execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
@@ -337,7 +337,7 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": 4,
"metadata": {},
"outputs": [
{
@@ -346,7 +346,7 @@
"(15854, 13)"
]
},
- "execution_count": 19,
+ "execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
@@ -357,7 +357,7 @@
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
@@ -368,7 +368,7 @@
},
{
"cell_type": "code",
- "execution_count": 22,
+ "execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
@@ -377,7 +377,7 @@
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": 7,
"metadata": {},
"outputs": [
{
@@ -386,7 +386,7 @@
"(13441, 13)"
]
},
- "execution_count": 23,
+ "execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
@@ -397,7 +397,7 @@
},
{
"cell_type": "code",
- "execution_count": 32,
+ "execution_count": 8,
"metadata": {},
"outputs": [
{
@@ -443,7 +443,7 @@
" 'Spectacle']"
]
},
- "execution_count": 32,
+ "execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
@@ -471,13 +471,13 @@
},
{
"cell_type": "code",
- "execution_count": 30,
+ "execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "e5a45c55993f47019fbdc0aceda84def",
+ "model_id": "cf58eccecc1847e48d520a83040e3ec7",
"version_major": 2,
"version_minor": 0
},
@@ -491,7 +491,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "a7285f3fe7154920a0bb05fdb921d6f9",
+ "model_id": "2e91d1ab181f409f9c6263a255991c9f",
"version_major": 2,
"version_minor": 0
},
@@ -505,7 +505,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "0cb38fe0e7934c49bbce246e88dd6e53",
+ "model_id": "4c23b1cb783d4c25b8332447bf25755a",
"version_major": 2,
"version_minor": 0
},
@@ -519,7 +519,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "2c9dd205446a4b25848f1f06beeac8ae",
+ "model_id": "203385bbd7664b8b87250e4883cc300f",
"version_major": 2,
"version_minor": 0
},
@@ -533,7 +533,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "032334b5310d4588993b7d85329d916c",
+ "model_id": "c068998923a14abf8e38d8e0d89248ad",
"version_major": 2,
"version_minor": 0
},
@@ -583,7 +583,7 @@
},
{
"cell_type": "code",
- "execution_count": 48,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
@@ -638,7 +638,7 @@
},
{
"cell_type": "code",
- "execution_count": 49,
+ "execution_count": 11,
"metadata": {},
"outputs": [
{
@@ -647,7 +647,7 @@
"\"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,
+ "execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
@@ -658,7 +658,7 @@
},
{
"cell_type": "code",
- "execution_count": 50,
+ "execution_count": 12,
"metadata": {},
"outputs": [
{
@@ -671,47 +671,47 @@
{
"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)]"
+ "[('Commerce', 70.05096077919006),\n",
+ " ('Anatomie', 68.73840689659119),\n",
+ " ('Politique', 60.71174740791321),\n",
+ " ('Géographie', 59.156250953674316),\n",
+ " ('Architecture', 58.74174237251282),\n",
+ " ('Histoire', 57.459235191345215),\n",
+ " ('Agriculture - Economie rustique', 53.53081226348877),\n",
+ " ('Histoire naturelle', 48.459288477897644),\n",
+ " ('Antiquité', 46.68458700180054),\n",
+ " ('Beaux-arts', 42.856183648109436),\n",
+ " ('Mesure', 41.31035804748535),\n",
+ " ('Jeu', 41.22118949890137),\n",
+ " ('Droit - Jurisprudence', 41.1332905292511),\n",
+ " ('Minéralogie', 38.137245178222656),\n",
+ " ('Spectacle', 37.80339956283569),\n",
+ " ('Pêche', 37.214648723602295),\n",
+ " ('Superstition', 36.727988719940186),\n",
+ " ('Arts et métiers', 36.511969566345215),\n",
+ " ('Métiers', 36.5054726600647),\n",
+ " ('Monnaie', 35.89862287044525),\n",
+ " ('Musique', 32.74966776371002),\n",
+ " ('Mathématiques', 32.70111680030823),\n",
+ " ('Chasse', 29.35197949409485),\n",
+ " ('Economie domestique', 28.346234560012817),\n",
+ " ('Philosophie', 27.653270959854126),\n",
+ " ('Chimie', 25.783824920654297),\n",
+ " ('Physique - [Sciences physico-mathématiques]', 25.4037082195282),\n",
+ " ('Médailles', 24.58679974079132),\n",
+ " ('Grammaire', 22.36253321170807),\n",
+ " ('Caractères', 20.14845609664917),\n",
+ " ('Pharmacie', 19.720394909381866),\n",
+ " ('Militaire (Art) - Guerre - Arme', 19.682711362838745),\n",
+ " ('Médecine - Chirurgie', 18.615825474262238),\n",
+ " ('Marine', 18.208028376102448),\n",
+ " ('Belles-lettres - Poésie', 13.306896388530731),\n",
+ " ('Blason', 10.476677119731903),\n",
+ " ('Religion', 9.702161699533463),\n",
+ " ('Maréchage - Manège', 4.211411997675896)]"
]
},
- "execution_count": 50,
+ "execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
@@ -732,7 +732,7 @@
},
{
"cell_type": "code",
- "execution_count": 51,
+ "execution_count": 13,
"metadata": {},
"outputs": [
{
@@ -741,7 +741,7 @@
"'Commerce'"
]
},
- "execution_count": 51,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
@@ -752,1041 +752,110 @@
},
{
"cell_type": "code",
- "execution_count": 52,
+ "execution_count": null,
"metadata": {},
"outputs": [],
"source": [
- "pred_labels = []\n",
- "prob_labels = []\n",
- "\n",
- "for content in tqdm(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"
+ "y_true = df[column_class].tolist()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
+ "outputs": [],
+ "source": [
+ "def get_tsv_content(y_true, prob_labels):\n",
+ " c = ''\n",
+ " for i, row in enumerate(prob_labels):\n",
+ " c += y_true[i] + '\\t'\n",
+ " for t in row:\n",
+ " c += t[0] + '\\t' + str(t[1])+'\\t'\n",
+ " c += '\\n'\n",
+ "\n",
+ " return c"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 35,
+ "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",
- " ...]"
+ "672.05"
]
},
- "execution_count": 40,
+ "execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
- "true_labels = df[column_class].tolist()"
+ "len(texts) / 20"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 36,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ " 1%| | 5/673 [10:28<23:19:26, 125.70s/it]\n"
+ ]
+ },
+ {
+ "ename": "IndexError",
+ "evalue": "string index out of range",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[36], line 12\u001b[0m\n\u001b[1;32m 8\u001b[0m prob_labels \u001b[39m=\u001b[39m []\n\u001b[1;32m 10\u001b[0m \u001b[39mfor\u001b[39;00m j, content \u001b[39min\u001b[39;00m \u001b[39menumerate\u001b[39m(batch):\n\u001b[0;32m---> 12\u001b[0m true_probs \u001b[39m=\u001b[39m zero_shot_prediction(content[i][:\u001b[39m1024\u001b[39m], classes)\n\u001b[1;32m 14\u001b[0m \u001b[39m#pred_labels.append(get_highest_score(true_probs, classes)[0])\u001b[39;00m\n\u001b[1;32m 15\u001b[0m prob_labels\u001b[39m.\u001b[39mappend(get_sorted_scores(true_probs, classes))\n",
+ "\u001b[0;31mIndexError\u001b[0m: string index out of range"
+ ]
+ }
+ ],
+ "source": [
+ "texts = df[column_text].tolist()\n",
+ "batch_size = 20\n",
+ "\n",
+ "for i in tqdm(range(0, len(texts), batch_size)):\n",
+ " batch = texts[i:i+batch_size]\n",
+ " batch_y_true = y_true[i:i+batch_size]\n",
+ "\n",
+ " prob_labels = []\n",
+ "\n",
+ " for j, content in enumerate(batch):\n",
+ "\n",
+ " true_probs = zero_shot_prediction(content[j][: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",
+ "\n",
+ " with open('zero-shot-classification.tsv', 'a') as f:\n",
+ " f.write(get_tsv_content(batch_y_true, prob_labels))\n",
+ " #print(prob_labels) \n"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
{
"cell_type": "code",
"execution_count": null,
@@ -1817,11 +886,11 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.13"
+ "version": "3.9.15"
},
"vscode": {
"interpreter": {
- "hash": "16fac9c2d845f8e1f8c6fffffe3d3a0def61c7e42da17a08d00f279ad4dea797"
+ "hash": "c5e52a88b09a5dede77937220424e7ea0969cc6c60aea64b9a5e920a2b19e73c"
}
}
},