Shuffle during CV #490
Answered
by
sonichi
Sukantabasu
asked this question in
Q&A
-
I am using eval_method = cv with n_splits = 5. By default the "uniform split" option is selected. I am not sure if FLAML is shuffling data prior to split. I am expecting it not to. |
Beta Was this translation helpful? Give feedback.
Answered by
sonichi
Mar 21, 2022
Replies: 1 comment 2 replies
-
It shuffles the data prior to split. You can use a custom splitter as "split_type" to prevent the shuffling if you prefer the data not to be shuffled: https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML#data-split-method |
Beta Was this translation helpful? Give feedback.
2 replies
Answer selected by
Sukantabasu
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
It shuffles the data prior to split. You can use a custom splitter as "split_type" to prevent the shuffling if you prefer the data not to be shuffled: https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML#data-split-method