Sebastian Raschka
Sebastian Raschka
Hi Junting. I just did a rebase because there were some issues in the continuous integration service. Based on the recent tests, it looks like your code still has errors.
Hi @NimaSarajpoor thanks for finding this old PR. Yes, `fixed_features` are already in the `SequentialFeatureSelector` but not in the `ExhaustiveFeatureSelector`, and it would be awesome to have it. I am...
Yup, thanks :)
Looks really nice, thanks a lot! I see that you also addressed the issue with the unit tests raised in #192 . To be honest, I don't know why exactly...
Sorry, but on a second thought, I think we can simplify this greatly. E.g., there's a `check_scoring` function in scikit-learn that we could use to get the "right" `scorer` for...
Thanks, and there's definitely no rush :)
Thanks for the update, and again, no worries regarding the delay. Really appreciate the PR!
Thanks for the update! > maybe guards / try-catch in complement_scorer for scorer methods evaluating within [0,1] I think that would be a good idea. Another thing, and it was...
Thanks for the PR, I see that it is passing now :). Could you maybe include a regressor in the unit tests, just to make sure that it works as...
Thanks for the PR! I agree that overfitting can become an issue. Currently, there is the option - `k_features="parsimonious"` which will select the smallest feature set within 1 standard error...