From 074aed88c39b0bd7c56bb135a2058f3eb916acc5 Mon Sep 17 00:00:00 2001
From: Ludovic Moncla <moncla.ludovic@gmail.com>
Date: Fri, 9 Dec 2022 20:59:37 +0100
Subject: [PATCH] Update Predict_XAI.ipynb
---
notebooks/Predict_XAI.ipynb | 104 ++++++++++++++++++++++++++++++++++--
1 file changed, 99 insertions(+), 5 deletions(-)
diff --git a/notebooks/Predict_XAI.ipynb b/notebooks/Predict_XAI.ipynb
index 968a65b..259796a 100644
--- a/notebooks/Predict_XAI.ipynb
+++ b/notebooks/Predict_XAI.ipynb
@@ -738,19 +738,113 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 119,
"metadata": {},
"outputs": [],
"source": [
- "edda_par_path = \"/Users/lmoncla/Nextcloud-LIRIS/GEODE/GEODE - Partage consortium/Corpus/EDdA/EDdA_dataset_articles.tsv\"\n",
+ "edda_par_path = \"/Users/lmoncla/Nextcloud-LIRIS/GEODE/GEODE - Partage consortium/Corpus/EDdA/EDdA_parallel_dataset_articles.tsv\"\n",
"df_EDdA_par = pd.read_csv(edda_par_path, sep=\"\\t\")"
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 120,
"metadata": {},
- "outputs": [],
+ "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>id</th>\n",
+ " <th>tome</th>\n",
+ " <th>filename</th>\n",
+ " <th>content</th>\n",
+ " <th>nb_words</th>\n",
+ " </tr>\n",
+ " </thead>\n",
+ " <tbody>\n",
+ " <tr>\n",
+ " <th>0</th>\n",
+ " <td>T1article100</td>\n",
+ " <td>T1</td>\n",
+ " <td>article100</td>\n",
+ " <td>ABATAGE, s. m. On dit dans un chantier & sur\\n...</td>\n",
+ " <td>64</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>1</th>\n",
+ " <td>T1article1016</td>\n",
+ " <td>T1</td>\n",
+ " <td>article1016</td>\n",
+ " <td>AFFILIATION, s. f. (Jurispr.) s'est dit par le...</td>\n",
+ " <td>69</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>2</th>\n",
+ " <td>T1article1036</td>\n",
+ " <td>T1</td>\n",
+ " <td>article1036</td>\n",
+ " <td>AFFLUENT, adj. terme de rivieres, se dit d'une...</td>\n",
+ " <td>50</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>3</th>\n",
+ " <td>T1article1038</td>\n",
+ " <td>T1</td>\n",
+ " <td>article1038</td>\n",
+ " <td>AFFORAGE, s. terme de Droit, qui se prend dans...</td>\n",
+ " <td>83</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>4</th>\n",
+ " <td>T1article1039</td>\n",
+ " <td>T1</td>\n",
+ " <td>article1039</td>\n",
+ " <td>AFFOUAGE, s. terme de Coûtumes, qui signifie l...</td>\n",
+ " <td>27</td>\n",
+ " </tr>\n",
+ " </tbody>\n",
+ "</table>\n",
+ "</div>"
+ ],
+ "text/plain": [
+ " id tome filename \\\n",
+ "0 T1article100 T1 article100 \n",
+ "1 T1article1016 T1 article1016 \n",
+ "2 T1article1036 T1 article1036 \n",
+ "3 T1article1038 T1 article1038 \n",
+ "4 T1article1039 T1 article1039 \n",
+ "\n",
+ " content nb_words \n",
+ "0 ABATAGE, s. m. On dit dans un chantier & sur\\n... 64 \n",
+ "1 AFFILIATION, s. f. (Jurispr.) s'est dit par le... 69 \n",
+ "2 AFFLUENT, adj. terme de rivieres, se dit d'une... 50 \n",
+ "3 AFFORAGE, s. terme de Droit, qui se prend dans... 83 \n",
+ "4 AFFOUAGE, s. terme de Coûtumes, qui signifie l... 27 "
+ ]
+ },
+ "execution_count": 120,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"df_EDdA_par.head()"
]
@@ -1263,7 +1357,7 @@
}
],
"source": [
- "word_attributions = cls_explainer(content if len(content) < 512 else content[:512])\n",
+ "word_attributions = cls_explainer(content[:512])\n",
"word_attributions"
]
},
--
GitLab