Hi @wxchan, I'm your friendly neighborhood Microsoft Pull Request Bot (You can call me MSBOT). Thanks for your contribution!
You've already signed the contribution license agreement. Thanks!

The agreement was validated by Microsoft and real humans are currently evaluating your PR.

TTYL, MSBOT;

@wxchan ,
for 1, it seems will have many parameters...
for 2. Have you tested to not pop nthread in get_params ? I remember it will cause error..

It passes the test even if I delete pop nthread. What kind of error did you meet?

I forget the details. it seems is something like parameter xxx not found.
If it can pass the test. I think it is ok.

Perhaps some error in c_api already been fixed?

I think we should follow sklearn.base.BaseEstimator though there will be many parameters because it's inherited, and it can let user know where to set drop_rate etc.

Finished.

**kwargs won't work for RandomizedSearchCV & GridSearchCV. other_params need to specify dict of dict. I go along with sklearn.base.BaseEstimator.

I remove pop nthread for now, if there is a bug, add it under fit().