chore(notebook): met à jour le chemin vers le dataset nettoyé FR

notebooks/01_exploration.ipynb

@@ -67,7 +67,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 14,
    "id": "0ff4a780",
    "metadata": {},
    "outputs": [
@@ -92,6 +92,7 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
+       "      <th>id</th>\n",
        "      <th>comment_text</th>\n",
        "      <th>toxic</th>\n",
        "      <th>severe_toxic</th>\n",
@@ -104,6 +105,7 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
+       "      <td>000113f07ec002fd</td>\n",
        "      <td>Hé mec, je n'essaye vraiment pas de modifier l...</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
@@ -114,6 +116,7 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
+       "      <td>00025465d4725e87</td>\n",
        "      <td>\"\\n\\nFélicitations de ma part également, utili...</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
@@ -124,6 +127,7 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
+       "      <td>0009801bd85e5806</td>\n",
        "      <td>Le point Mitsurugi n'avait aucun sens - pourqu...</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
@@ -134,6 +138,7 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
+       "      <td>0001b41b1c6bb37e</td>\n",
        "      <td>\"\\nPlus\\nJe ne peux pas faire de véritables su...</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
@@ -144,6 +149,7 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
+       "      <td>00040093b2687caa</td>\n",
        "      <td>alignement sur ce sujet et qui sont contraires...</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
@@ -157,22 +163,22 @@
        "</div>"
       ],
       "text/plain": [
-       "                                        comment_text  toxic  severe_toxic  \\\n",
-       "0  Hé mec, je n'essaye vraiment pas de modifier l...      0             0   \n",
-       "1  \"\\n\\nFélicitations de ma part également, utili...      0             0   \n",
-       "2  Le point Mitsurugi n'avait aucun sens - pourqu...      0             0   \n",
-       "3  \"\\nPlus\\nJe ne peux pas faire de véritables su...      0             0   \n",
-       "4  alignement sur ce sujet et qui sont contraires...      0             0   \n",
+       "                 id                                       comment_text  toxic  \\\n",
+       "0  000113f07ec002fd  Hé mec, je n'essaye vraiment pas de modifier l...      0   \n",
+       "1  00025465d4725e87  \"\\n\\nFélicitations de ma part également, utili...      0   \n",
+       "2  0009801bd85e5806  Le point Mitsurugi n'avait aucun sens - pourqu...      0   \n",
+       "3  0001b41b1c6bb37e  \"\\nPlus\\nJe ne peux pas faire de véritables su...      0   \n",
+       "4  00040093b2687caa  alignement sur ce sujet et qui sont contraires...      0   \n",
        "\n",
-       "   obscene  threat  insult  identity_hate  \n",
-       "0        0       0       0              0  \n",
-       "1        0       0       0              0  \n",
-       "2        0       0       0              0  \n",
-       "3        0       0       0              0  \n",
-       "4        0       0       0              0  "
+       "   severe_toxic  obscene  threat  insult  identity_hate  \n",
+       "0             0        0       0       0              0  \n",
+       "1             0        0       0       0              0  \n",
+       "2             0        0       0       0              0  \n",
+       "3             0        0       0       0              0  \n",
+       "4             0        0       0       0              0  "
       ]
      },
-     "execution_count": 9,
+     "execution_count": 14,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -189,7 +195,7 @@
     "plt.rcParams[\"figure.figsize\"] = (10, 6)\n",
     "\n",
     "# Chargement des données\n",
-    "df = pd.read_csv(\"../data/jigsaw_toxic_fr_clean.csv\")\n",
+    "df = pd.read_csv(\"../data/jigsaw-toxic-comment-train-google-fr-cleaned.csv\")\n",
     "\n",
     "# Nettoyage de colonnes inutiles si présentes\n",
     "df = df.loc[:, ~df.columns.str.startswith(\"Unnamed\")]\n",