feat: feature reduction for sklearn pipeline
Closed this issue · 1 comments
MartinBernstorff commented
Why?
We would like to examine the trade-off between n_features
and performance. To do that, we want to train with the 'best' [1, 2, 3, 4, 5, 6] features and so forth, and get performance for each of them.
Looks like sklearn
doesn't output this information. However, mlextend
does!
To get the right type of crossvalidation, we want to specify that as well. It can take a generator, like we use in the crossvalidation trainer.
- Implement sequential feature selection unit test
github-actions commented
This issue is stale because it has been open for 14 days with no activity. Feel free to either 1) remove the stale label or 2) comment. If nothing happens, this will be closed in 7 days.