Hey guys. Just wanted to know if there is any difference between neo-ai/tvm and apache/incubator-tvm. I heard somewhere that neo-ai/tvm supports conversion of sklearn random forest as well. Is that true? If not can anyone pls explain what is the difference between two?
Ping @hcho3 who is the author of treelite
neo-ai/tvm is a fork of apache/incubator-tvm, and the fork is hosted by AWS (*). The reason for the fork is to host features that are specifically useful to the org but may be outside the scope of the wider TVM community.
I heard somewhere that neo-ai/tvm supports conversion of sklearn random forest as well. Is that true?
No, neo-ai/tvm does not support random forests. You should use dmlc/treelite for that purpose.
Thank you so much
Do u know any other open source platform that supports random forest other than dmlc/treelite? I happen to come across sklearn-compiledtrees but it only deals with Regression models.