davidlkl

Hi I am also looking to implement custom loss function. However after reading the source, I think the loss function is implemented in the C library. https://github.com/interpretml/interpret/blob/f298c734f3fedce522795e12da4e7a9788596f06/python/interpret-core/interpret/glassbox/ebm/internal.py#L197

Maybe out of topic but I think you may find this useful: https://github.com/ZebinYang/gaminet They claimed it beats EBM in most of their cases, with similar setting (GA2M).

Hi, Thanks @tczhao for the hard work! Just would like to understand that if it would be sufficient to supply a custom loss by predictor_kwargs, (in other words, is there...

Hi Davmre, Thanks for your reply! I understand that semi-continuous distribution is a tricky point to handle. I am modeling insurance claims which has large amount of zeros (>90%) and...

Hi @brianwa84 , I understand that by sampling I can estimate the variance. What I would like to achieve is to estimate the model parameters (probability of no claim, (expected...

As I originally did it for recognition of English word only, the fitted model cannot recognize digits. However, you can reuse the whole model architecture for this task by modifying...