Skip to content

R
robo-dividend
Commit 9cca8d78 authored by 吕先亚's avatar 吕先亚
Browse files
Options

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

parent 106e8d82
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
ai/training_data_builder.py
View file @ 9cca8d78
...@@ -193,6 +193,6 @@ class TrainingDataBuilder(ABC): ...@@ -193,6 +193,6 @@ class TrainingDataBuilder(ABC):
X_train, X_test, y_train, y_test = train_test_split(scaledX, y, test_size=0.04, shuffle=False) 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() date_index = DataAll['date'][-len(X_test):-self._numForecastDays].to_numpy()
# To avoid data leak, test set should start from numForecastDays later # To avoid data leak, test set should start from numForecastDays later
X_test = X_test[:-self._numForecastDays] X_test = X_test[-self._numForecastDays:]
y_test = y_test[:-self._numForecastDays] y_test = y_test[-self._numForecastDays:]
return X_train, X_test, y_train, y_test, scaledX_forecast, forecastDay, date_index return X_train, X_test, y_train, y_test, scaledX_forecast, forecastDay, date_index
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment