diff --git a/content/course/modelisation/4_featureselection/index.qmd b/content/course/modelisation/4_featureselection/index.qmd
index ea67a8b1e..04b43702a 100644
--- a/content/course/modelisation/4_featureselection/index.qmd
+++ b/content/course/modelisation/4_featureselection/index.qmd
@@ -231,7 +231,7 @@ les variables sélectionnées sont :
 #| echo: false
 
 #3. Estimer un modèle LASSO et afficher les valeurs des coefficients
-lasso1 = Lasso(fit_intercept=True,normalize=False, alpha = 0.1).fit(X_train,y_train)
+lasso1 = Lasso(fit_intercept=True, alpha = 0.1).fit(X_train, y_train)
 
 #np.abs(lasso1.coef_)
 features_selec = df2.select_dtypes(include=np.number).drop("per_gop", axis = 1).columns[np.abs(lasso1.coef_)>0].tolist()
@@ -254,7 +254,7 @@ Par ailleurs, on sélectionne des variables redondantes. Une phase plus approfon
 corr = df2[features_selec].corr()
 
 plt.figure()
-p = corr.style.background_gradient(cmap='coolwarm', axis=None).set_precision(2)
+p = corr.style.background_gradient(cmap='coolwarm', axis=None).format('{:.2f}')
 p
 ```
 
@@ -358,7 +358,7 @@ yindex = df3.columns.get_loc("per_gop")
 df3_scale = scaler.fit(df3).transform(df3)
 # X_train, X_test , y_train, y_test = train_test_split(np.delete(data, yindex, axis = 1),data[:,yindex], test_size=0.2, random_state=0)
 
-lcv = LassoCV(alphas=my_alphas ,normalize=False,fit_intercept=False,random_state=0,cv=5).fit(np.delete(df3_scale, yindex, axis = 1), df3_scale[:,yindex])
+lcv = LassoCV(alphas=my_alphas, fit_intercept=False,random_state=0,cv=5).fit(np.delete(df3_scale, yindex, axis = 1), df3_scale[:,yindex])
 ```
 
 ```{python}