You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you.
As you know, to support sklearn.pipeline , we need fit and transform functions.
Also, in order to support sklearn's GridSearchCV and cross_val_predict, I think that the get_params and set_params, get_support functions are also necessary.
As you know,
from sklearn.model_selectiom import cross_val_predict, GridSearchCV
y_pred_cv = cross_val_predict(pipe, X, y)
gcv = GridSearchCV(pipe, param_grid={"population_size" : [20,30,50,]})
gcv.fit(X,y)
# It is just an example,
# whether population_size optimization is necessary is not discussed here.
I think that it will be very convenient if pipeline can be used as follows
The text was updated successfully, but these errors were encountered: