There are some version condition in toolchain. Take pytorch as an example, please check the version are "pytorch = 1.4" and "onnx = 1.4" . When export model to onnx from pytorch, make sure the opset version equals 9. You can refer to following command to export:
torch.onnx.export(model, img, onnx_export_file, verbose=False, opset_version=9, keep_initializers_as_inputs=True, input_names=['images'], output_names=['classes', 'boxes'] if y isNoneelse ['output'])
And here is the version of yolov5, it only supports to v2.0 release now.
Comments
Hi Jerome,
There are some version condition in toolchain. Take pytorch as an example, please check the version are "pytorch = 1.4" and "onnx = 1.4" . When export model to onnx from pytorch, make sure the opset version equals 9. You can refer to following command to export:
torch.onnx.export(model, img, onnx_export_file, verbose=False, opset_version=9, keep_initializers_as_inputs=True, input_names=['images'], output_names=['classes', 'boxes'] if y isNoneelse ['output'])
And here is the version of yolov5, it only supports to v2.0 release now.
Hi Ethon,
Thanks!
Did you means I can convert model by using toolchain 2.0?
However, do you support optset=11? optset=9 is too old to do migrate in our environment.
Thanks
Jerome
Hi Jerome,
The latest version of toolchain is v0.12.1, you can convert and compile the model yolov3 on it. Please refer to the instruction manual. (http://doc.kneron.com/docs/#toolchain/manual_520/ )
The toolchain only supports opset = 9 now, optset=11 will be supported in few months later, but the release time has not yet been determined.