Sample weight support for regression problems

[kmedved]

Hello - I just saw this library written up on Medium and it looks very interesting. I wanted to ask about the possibility of adding Sample Weight support for it? Xgboost already has support for it via the weight parameter in the .fit() call, so I'm not sure what would be needed other than updating the API to allow a user to pass sample weights.

Thanks!

[rsesha]

Hello kmedved:
It should be possible to send the sample_weight parameter. Let me look into it and update this library if possible.
Ram

[kmedved]

Great to hear. Thanks!

[AutoViML]

Hi @kmedved 👍
Sorry it took so long to add sample weight functionality to featurewiz . It now adds sample weights automatically for classification problems. You can see the code here:

if modeltype == 'Multi_Classification':
    wt_array = get_sample_weight_array(y_train)
    dtrain = xgb.DMatrix(x_train, label=y_train, weight=wt_array)

It adds sample weights for multi-class problems and adds scale_pos_weight for binary class problems as you can see here:

scale_pos_weight = get_scale_pos_weight(y_train)

I hope this is good for you. if you want to send in a sample weight separately I need to change the input arguments, hence I thought I will do it automatically. Take a look at the latest version:

pip install featurewiz --upgrade

Thanks
Auto_ViML