To avoid data leak, test set should start from numForecastDays later

ai/training_data_builder.py

@@ -197,8 +197,8 @@ class TrainingDataBuilder(ABC):
         else:
             # Step 2: Split data into train set and test set
             X_train, X_test, y_train, y_test = train_test_split(scaledX, y, test_size=0.04, shuffle=False)
-            date_index = DataAll['date'][-len(X_test):-self._numForecastDays].to_numpy()
             # To avoid data leak, test set should start from numForecastDays later
             X_test = X_test[self._numForecastDays:]
             y_test = y_test[self._numForecastDays:]
+            date_index = DataAll['date'][:-len(X_test)].to_numpy()
         return X_train, X_test, y_train, y_test, scaledX_forecast, forecastDay, date_index