Federico Minutoli

> skl2onnx cannot translate such an expression `Identity = lambda: preprocessing.FunctionTransformer()` because it does not know to function a custom function. In your case, you should replace it by `'passthrough'`...

any pointers on that, @xadupre?

After digging deeper, it seems like the problem arises only when exporting the stacking classifier with the lightGBM model as backbone, while the lightGBM model alone and the preprocessing work...

For some reason, the nested pipeline definition was the problem, @xadupre. The following definition works as expected: ```python def abc_Embedder() -> list[tuple[str, Any]]: return [ ("cast64", skl2onnx.sklapi.CastTransformer(dtype=numpy.float64)), ("scaler", preprocessing.StandardScaler()), ("cast32",...

@danielhanchen u have any official guidelines or proven experiments running multi-stage-training? I have had similar issues to the OP multiple times, and have had to revert back not to be...

> I don’t see why it wouldn’t, you may have to configure it differently @thad0ctor, mind sharing hints at what bits of your repository should be configured differently? If any....

@jasonshenj could you provide an example? would help with reproducibility and help us better track down the root cause of your issues