openvino2tensorflow v1.31.1

• Security update
• README update

openvino2tensorflow v1.31.0

• OpenVINO 2021.4.582 -> 2022.1.0
• Remove MXNet

openvino2tensorflow v1.30.3

• CUDA 11.6
• TensorRT 8.4.0
• cuDNN 8.4
• TensorFlow v2.9.0
• TensorFlow Lite v2.9.0
• PyTorch v1.12.0 (with grid_sample)
• onnxruntime-gpu v1.12.0

openvino2tensorflow v1.30.2

• #106 transpose channel for prelu replacement when prelu channel number is 3 @zye1996
• https://github.com/PINTO0309/simple-onnx-processing-tools v1.0.22

openvino2tensorflow v1.30.1

Added workaround for avoiding conversion errors.

• GroupConvolution
• MatMul
• NonZero

openvino2tensorflow v1.30.0

• CUDA 11.6
• TensorRT 8.4.0
• cuDNN 8.4
• TensorFlow v2.9.0-rc0
• TensorFlow Lite v2.9.0-rc0
• PyTorch v1.12.0 (with grid_sample)
• onnxruntime v1.12.0

openvino2tensorflow v1.29.6

• Applying NCHW<->NHWC transposition to Subtract 4D constant
• OpenVINO 2022.1 Added logic to ignore the second output of MaxPool with opset=8
  • #103 openvino int8/fp32 to tflite convertion
  • #100 KeyError
  • #99 KeyError when running openvino2tensorflow

openvino2tensorflow v1.29.5

https://github.com/PINTO0309/simple-onnx-processing-tools

1. sne4onnx
2. snd4onnx
3. snc4onnx
4. scs4onnx
5. sog4onnx
6. sam4onnx
7. soc4onnx
8. scc4onnx

openvino2tensorflow v1.29.4

• Added automatic NCHW conversion function for ONNX
• Add --disable_onnx_nchw_conversion option
  • Disables conversion to NCHW when generating ONNX and generates ONNX in NHWC format

openvino2tensorflow v1.29.3

• Support for OpenVINOExecutionProvider for onnxruntime_gpu
  • https://zenn.dev/pinto0309/scraps/1e9c0a00112cf9
  • https://zenn.dev/pinto0309/scraps/1ecbf4ef73d3ac