emedvedev/attention-ocr

ONNX conversion error

Closed this issue · 2 comments

Guneetkaur03 commented

I am attempting to convert this model to ONNX using the tf2onnx tool. The tool can take a frozen graph or saved model as input and produce an ONNX file. Unfortunately, it fails on aocr, apparently to due the graph being invalid.

I am filing the issue here and not against tf2onnx because the error occurred in Tensorflow while attempting to import the graph that aocr export produced. Here is the error:
tensorflow.python.framework.errors_impl.InvalidArgumentError: Node 'cond/ExpandDims' has an _output_shapes attribute inconsistent with the GraphDef for output #0: Shapes must be equal rank, but are 1 and 0

emedvedev commented

Duplicate of #138.

This is a known issue, but unfortunately I don't have cycles (or ONNX experience) at the moment. Leaving the original one open though, in case someone wants to try and fix it.

Ram-Godavarthi commented

Hi @Guneetkaur03 ,

Did you solve problem of attention ocr to onnx???