Fixed n_estimator retriever

app.py

@@ -227,12 +227,12 @@ However, most system only supports the ability to search games by their category
 With that, we wanted to try and make a game recommendation based on description where user can describe the game they are looking for with text and later narrow it down with classification based on their content like genres and price ranges.
 ''')
                     h2("The Model")
-                    p("""The system consists of three model :\nThe first one is the Language Model that will learn users review for a game and use that as a way to describe a game.
-                      The Language Model will be a classifier based on a Gradient Boosting model called XGBClassifier.\n
-                      The second model and third model will be the filter model.\n
+                    p("""The system consists of three model :The first one is the Language Model that will learn users review for a game and use that as a way to describe a game.
+                      The Language Model will be a classifier based on a Gradient Boosting model called XGBClassifier.
+                      The second model and third model will be the filter model.
                       The second model is a collaborative filter model where it will recommend the user a game based on a game that they have liked in the past or a game that they specify similar to the game they are looking for.
-                      This model will learn based on other user who have reviewed a game and a similar game is the game that said user liked other than the input game. This model will use utility matrix and cosine similarity.\n
-                      The third model is a content based model where it will recommend user a game based on their content such as Genres, Categories, Price range, Year Release, etc.\n
+                      This model will learn based on other user who have reviewed a game and a similar game is the game that said user liked other than the input game. This model will use utility matrix and cosine similarity.
+                      The third model is a content based model where it will recommend user a game based on their content such as Genres, Categories, Price range, Year Release, etc.
                       This third model will be a KNeighborsClassifier.""")
             with gr.Column(elem_id="dataset", elem_classes="content-section", visible=False) as dataset_section:
                 """
@@ -601,7 +601,7 @@ classifier.fit(
 )
 """)
                 history = model.text_based_recommender.history
-                n_estimator = len(history['validation_0']['merror'])
+                n_estimator = np.arange(len(history['validation_0']['merror']))
                 
                 h2('Training vs Validation log loss')
                 results = {