lyg5623 / lightgbm_predict4j Goto Github PK

Noticed here exists a demo suit for v2, and i can run the UseageTest in v2 package successfully, while when i use a model generared by lightgbm v2.0.5, it produced a error: "[Tree] Tree model string format error, should contain default_value field". The default_value is contained in models generated by lightgbm's earlier version. The model file cause the error is attached below.
Thanks for any kind help in advance!

lgbm_test.txt

1 how can i use it in maven ? I can't find this bag in Maven warehouse

2 the lightgbm's new version has reached 3, or4, can the v2 of this code work for the new version of lightgbm,

Hi, I checked the model v2 is right for binary classify , but the result is strange in multi-class classify. For example , output may have -34 or 4.0, etc, not the prob among(0 , 1).
Is the code supported multi-class model?

PS : The results of lightgbm tool looks normal, all classes' prob are among(0 , 1), so I think my model is right