From af98a95beabc2411a44d9c74ff4b0eada8bb0b47 Mon Sep 17 00:00:00 2001
From: Arthur DANJOU <arthurdanjou@outlook.fr>
Date: Mon, 13 Oct 2025 18:24:13 +0200
Subject: [PATCH] =?UTF-8?q?Ajout=20de=20l'ex=C3=A9cution=20de=20cellules?=
 =?UTF-8?q?=20pour=20le=20One=20Hot=20Encoding,=20la=20normalisation=20des?=
 =?UTF-8?q?=20variables=20num=C3=A9riques=20et=20la=20s=C3=A9paration=20de?=
 =?UTF-8?q?s=20donn=C3=A9es=20en=20ensembles=20d'apprentissage=20et=20de?=
 =?UTF-8?q?=20test.?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .../TP_3/2025_TP_3_M2_ISF.ipynb               | 26 ++++++++++++++-----
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/M2/Machine Learning/TP_3/2025_TP_3_M2_ISF.ipynb b/M2/Machine Learning/TP_3/2025_TP_3_M2_ISF.ipynb
index e894875..ea17293 100644
--- a/M2/Machine Learning/TP_3/2025_TP_3_M2_ISF.ipynb	
+++ b/M2/Machine Learning/TP_3/2025_TP_3_M2_ISF.ipynb	
@@ -2104,7 +2104,7 @@
     "\n",
     "# Visualiser avec Plotly\n",
     "fig = px.imshow(correlation_cat,\n",
-    "                text_auto='.2f',\n",
+    "                text_auto='.2f', # type: ignore\n",
     "                aspect=\"auto\",\n",
     "                color_continuous_scale='RdBu_r',\n",
     "                title='Matrice de corrélation des variables catégorielles (V de Cramér)')\n",
@@ -3066,11 +3066,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 38,
    "id": "b8530717",
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "encoder = preproc.OneHotEncoder(sparse_output=False, drop='first')\n",
+    "encoder.fit(vars_categorielles)\n",
+    "vars_categorielles_enc = encoder.transform(vars_categorielles)\n",
+    "vars_categorielles_enc = pd.DataFrame(vars_categorielles_enc, columns=encoder.get_feature_names_out()) # type: ignore"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -3082,11 +3087,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 39,
    "id": "4ff3847d",
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "scaler = preproc.StandardScaler()\n",
+    "scaler.fit(vars_numeriques)\n",
+    "vars_numeriques_scaled = scaler.transform(vars_numeriques)\n",
+    "vars_numeriques_scaled = pd.DataFrame(vars_numeriques_scaled, columns=vars_numeriques.columns)"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -3106,11 +3116,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 40,
    "id": "6a1c7907",
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "train, test = train_test_split(data_model, test_size=0.2, random_state=42)"
+   ]
   },
   {
    "cell_type": "markdown",