Xavier Dupré

It is possible to remove zipmap operator by using options. The example shows how to use sklearn-onnx for lightgbm [Convert a pipeline with a LightGBM model](http://onnx.ai/sklearn-onnx/auto_tutorial/plot_gexternal_lightgbm.html). The second one shows...

It could be but I could probably choose a more simple way, just adding a parameter `converrt_lightgbm(..., zipmap=True)`.

Ok I have two bugs to fix then. I'll start by adding a parameter zipmap to convert_lightgbm.

I started to work on this on PR #452.

A new version was released two days ago. Did you with this one?

The converter in onnxmltools only converts one model with one input. Here is the modified code. I changed two lines, both followed by a comment in upper case. ``` import...

A new version of onnxmltools we just released but this impor should still fail. But this is more an issue for winmltools, not onnxmltools.

onnx has just been release on python 3.10, onnxruntime will be soon. onnxmltools should be available after that.

onnx, onnxruntime are now available on python 3.10.

The error message is not clear enough but sparse is not fully supported yet in onnxruntime. It is safer to remove every use of sparse when converting a pipeline to...