onnxconverter-common version 1.5.0 is now available! This package was created alongside the onnxmltools package to enable common code to be shared among various ONNX converter offerings.
[onnxmltools](https://github.com/onnx/onnxmltools), [sklearn-onnx](https://github.com/onnx/sklearn-onnx), and [keras-onnx](https://github.com/onnx/keras-onnx) depend on the onnxconverter-common package for shared code and utilities, including:
1. Converting common ONNX operations ([onnx_ops.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/onnx_ops.py))
2. ONNX graph topology and data types ([topology.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/topology.py), [container.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/container.py), [data_types.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/data_types.py))
3. Other shared tools, including the float32->float16 converter ([float16.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/float16.py)), ONNX graph optimizer ([optimizer.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/optimizer.py)) and common shape calculators ([shape_calculator.py](https://github.com/microsoft/onnxconverter-common/blob/master/onnxconverter_common/shape_calculator.py))
How do I use the latest onnxconverter-common package?
pip install onnxconverter-common
python -c "import onnxconverter_common"
This package is automatically installed as part of the dependencies for onnxmltools, keras2onnx, and skl2onnx.
Highlights since the last release
* v1.5.0 for the onnxconverter-common package (https://github.com/onnx/onnxmltools/pull/314)
* [Opset 10] Updates for thresholded relu (https://github.com/onnx/onnxmltools/pull/308)
* [Opset 10] Deprecate Upsample, create Resize op (https://github.com/onnx/onnxmltools/pull/303)
* [Opset 10] Pooling operator updates: AveragePool, MaxPool (https://github.com/onnx/onnxmltools/pull/296)
* Add support for Py3.7, onnx 1.5, onnxruntime 0.4 (https://github.com/onnx/onnxmltools/pull/293)
* Added apply_slice function to enable multiple versions of Slice (https://github.com/onnx/onnxmltools/pull/291)