Which method is recommended for model online inference?

[hepengfei-ml]

Which method is recommended for model online inference? I want high performance concurrency,
PMML ?

[StrikerRUS]

It depends on your use case.

Initially, LightGBM wasn't optimized for fast predictions. Some time ago *SingleRowFast API was introduced: #2992. You can start with it.

Also, you can try to convert fitted model into if/else cpp code: https://lightgbm.readthedocs.io/en/latest/Parameters.html#convert_model.

In addition, you can find a lot of third-party libraries to inference LightGBM models here: https://github.com/microsoft/LightGBM#external-unofficial-repositories.

I'd personally recommend starting from the treelite library.

[github-actions]

This issue has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this.