search

AxeldeRomblay / MLBox

MLBox is a powerful Automated Machine Learning python library.
https://mlbox.readthedocs.io/en/latest/
Other
1.49k stars 274 forks source link

Why does mlbox recognize my multi-class training set as a regression problem #94

Closed gaoexingyun closed 4 years ago

gaoexingyun commented 4 years ago

I originally wanted to train my multi-class training set with mlbox classification, but after importing the data, it automatically recognized it as regression, and changed my target value data type from int to float.

AxeldeRomblay commented 4 years ago

Hello @gaoexingyun, Thanks for the issue ! You can refer to this one : https://github.com/AxeldeRomblay/MLBox/issues/87

To sum it up, MLBox always tries to cast your target into float so if you have something like : "1", "2", ... it is considered as a regression problem, better if your target looks like "C1", "C2", ...

Please tell me if it works now.

gaoexingyun commented 4 years ago

Hello @gaoexingyun, Thanks for the issue ! You can refer to this one : #87

To sum it up, MLBox always tries to cast your target into float so if you have something like : "1", "2", ... it is considered as a regression problem, better if your target looks like "C1", "C2", ...

Please tell me if it works now.

Thank you. It works.