Bhushan Sonawane comments

Results 60 comments of


                                            Bhushan Sonawane

output_feature has no shape when output has len(shape)==2

@yuvdar we do handle output of rank 2 convert adds output feature with shape none is that case. Have you encountered any case we don't handle this? please share an...

output_feature has no shape when output has len(shape)==2

@yuvdar when you set None, output shape will be inferred

IndexError: list assignment index out of range

@dzyjjpy apologies for delayed response. What is your ONNX opset version for given model? according to https://github.com/onnx/onnx/blob/master/docs/Changelog.md#Slice-1 Slice should have one input and in that case slice start and end...

Key error pads

@BorisKourt @Arxtage Looks like you are using opset version 11 could you please export to opset version 10? pytorch export has an option to export to opset version 10 (`opset=10`)

Log accurary with CoreML3

@aseemw should we ignore the log precision?

convert to CoreML and then get wrong predictions

@MarcSteven can you share onnx model?

dilated convolution does not support weight as input tensor after convert to CoreML

@edis219 can you pass weight as a constant?

convert to onnx and then convert to CoreML , but the prediction is wrong

@MarcSteven can you provide more details? Can you share ONNX model and input for which you observe incorrect prediction?

Error when 'Converting Node Type Slice'

@howiezhang @fellowProgrammer could you please build from source and try? this should have been fixed with https://github.com/onnx/onnx-coreml/pull/545

Issue converting Slice op when comes after Expand op

@kossnick this is due to shape unknown when expand op is present in onnx shape inference. have a #528 PR to print out error message correctly.