Feature: Support ONNX conversion for models
Do you support ONNX conversion?
Currently, the code base does not have code to support ONNX conversion.
We are looking into supporting this and have tried some ideas. However, due to the new architecture, we still need to finalize this.
@AlexanderPfefferle do you have more information about this?
https://github.com/PriorLabs/TabPFN/pull/165 enabled ONNX export The tests contain an example for how to export: https://github.com/PriorLabs/TabPFN/blob/b0bc461e7a44af8c233cdf6eed03d0881dd247a9/tests/test_classifier_interface.py#L227-L289 Currently, you can only use the exported model with the exact same number of features that the example input for exporting had, but @LeoGrin is looking into that.