check_model gives 'Bad node spec' error

[aviaisr]

Hi,

I'm trying to convert onnx to pb. The onnx file loading is done successfully. Then, I call prepare (a function of onnx_tf.beckend). This function calls onnx.checker.check_model -> that calls C.check_model(model.SerializeToString()).

I get these error lines: Traceback (most recent call last): File "C:/CP/DL_REPOSITORIES/pytorch-onnx-tensorflow-pb/convert_pytorch2onnx2tfpb.py", line 103, in tf_rep = prepare(onnx_model, strict=False) File "C:\ProgramData\Anaconda3\lib\site-packages\onnx_tf\backend.py", line 53, in prepare super(TensorflowBackend, cls).prepare(model, device, **kwargs) File "C:\ProgramData\Anaconda3\lib\site-packages\onnx\backend\base.py", line 74, in prepare onnx.checker.check_model(model) File "C:\ProgramData\Anaconda3\lib\site-packages\onnx\checker.py", line 91, in check_model C.check_model(model.SerializeToString()) onnx.onnx_cpp2py_export.checker.ValidationError: Node () has input size 1 not in range [min=2, max=3].

==> Context: Bad node spec: input: "708" output: "715" op_type: "Pad" attribute { name: "mode" s: "constant" type: STRING } attribute { name: "pads" ints: 0 ints: 0 ints: 0 ints: 0 ints: 0 ints: 0 ints: 0 ints: 0 type: INTS } attribute { name: "value" f: 0 type: FLOAT }

Can you please help?

[aviaisr]

I'm working with this onnx version version = '1.6.0' git_version = '553df22c67bee5f0fe6599cff60f1afc6748c635'

[askhade]

Pad operator was recently updated (https://github.com/onnx/onnx/blob/master/docs/Changelog.md#Pad-13) Per the new spec min of 2 inputs are required for this operator. In order to fix this issue you should update your Pad operator. Alternatively you can use an older opset for onnx