Xavier Dupré

I would recommend to add a function or an operator svd in onnx. Implementing the algorithm requires many accesses to one coefficient of the matrix. I doubt any implementation with...

Thanks for the feedback. The other implementation I tried is #4281. However, depending on how protobuf 4 is integrated into onnx, this option may not be available anymore, see https://developers.google.com/protocol-buffers/docs/news/2022-05-06.

How to choose between int8 or uint8? I don't remember if tensorflow is giving enough information to make the choice. Otherwise, it is possible to add an option to tensorflow...

There is no easy way to do hashes with ONNX. There is no dedicated operator and hashing with current operator is not straightforward. This implementation may take some time.

There was no recent development on coremltools converters. Would you like to contribute?

I tried your script but it works for me. What versions are you using? ``` import numpy from xgboost import XGBRegressor from onnxruntime import InferenceSession from skl2onnx.common.shape_calculator import calculate_linear_regressor_output_shapes from...

I can't replicate the issue. Do you obtain the same error with the unit test? Which version of pyspark, onnxruntime, onnx, onnxmltools are you using?

PR #795 should fix that bug. I'll make a new release soon.

onnx has released a new version and changed the TreeEnsembleClassifier specifications. You need to change the following line: ``onx = to_onnx(clf, np.array(X).astype(np.float32), target_opset={'': 15, 'ai.onnx.ml': 2})``

skl2onnx only converts model from scikit-learn, any other framework require a different library. I suggest using onnxmltools to convert pyspark models.