Update app.py

app.py

@@ -20,29 +20,28 @@ scaler = joblib.load('scaler.joblib')
 #fonction de predictions simple
 
 def pred(HS,	PS,	EA,	SH,	SQPP):
-  EA = encoder.transform([EA])[0]
-  x_new = np.arange([HS,	PS,	EA,	SH,	SQPP])
-  #normaliser les valeurs 
-  x_new = scaler.transform(x_new)
-  x_new = x_new.reshape(1, -1)
-  x_pred = model.predict(x_new)
+    EA = encoder.transform([EA])[0]
+    x_new = np.array([HS,  PS, EA,  SH,  SQPP])
+    x_new = x_new.reshape(1, -1)
+    x_new = scaler.transform(x_new)
+    y_pred = model.predict(x_new)
+    y_pred = round(y_pred[0], 2)
 
-  y_pred = np.round(y_pred[0], 2)
-
-  return f"la performance de cet etudiant est: {str(y_pred)}"
+    return f"la performance de cet etudiant est: {str(y_pred)}"
 
 
 def pred_csv(file):
-  df = pd.read_csv(file)
+   df = pd.read_csv(file)
 
-  prediction = []
+   prediction = []
 
-  for row in df.iloc[:, :].values:
+   for row in df.iloc[:, :].values:
         prediction.append(pred(row[0], row[1], encoder.transform([row[2][0]]), row[3]))
 
-  df['Performance Index']  = prediction
-  df.to_csv('perfo_etud.csv', index= False)
-  return 'perfo_etud.csv'
+   df['Performance Index']  = prediction
+   df.to_csv('perfo_etud.csv', index= False)
+   return 'perfo_etud.csv'
+
     
 demo = gr.Blocks(theme= gr.themes.Origin())