Dmitry Khominich

@frontword , thank you for your detailed feedback. ``` leaves does not support the sklearn xgboost model ? ``` Currently, models generated from xgboost sklearn API are not tested. I...

Hello! Thank for your report. ```go e = model.PredictCSR(csr.RowHeaders, csr.ColIndexes, csr.Values, predictions, 50, 5) ``` why do you use only 50 trees to predict? Try use all tress in ensemble,...

Thanks for the issue. I'm going to fix this.

@sbinet , thanks for your comment. For now models implement raw data structure interfaces as below" ```go // PredictCSR calculates predictions from ensembles of trees. `indptr`, `cols`, // `vals` represent...

After #14 , here is full binary compatibility to use `gonum/mat.Dense` without type conversions. As far as I know here is no Sparse matrix implementations in `gonum` (like Compressed Sparse...

Can you show your go code? Note, that `PredictSingle` will return 0.0 in case of error: https://godoc.org/github.com/dmitryikh/leaves#Ensemble.PredictSingle Try to use, `Predict` https://godoc.org/github.com/dmitryikh/leaves#Ensemble.Predict and check the error returned

Hi! Thanks for your feedback. Could be please provide more information: 1. How did you build `model` file? Did you use pythons bindings for xgboost or something? Can you provide...

@randbear , Did you manage with the problem by your self? If not, could you answer to my questions above? Thanks!

Hi! Can anybody provide me with the xgboost model file. Then I will be able to reproduce the bug on my side and finally fix it. Thanks!

Hi all! I assume that this is because of using objective function (`'objective': 'rank:ndcg'`, as example) that isn't supported by leaves. Anyway `leaves` should be not panic and show the...