Chin Huang

Unfortunately the conda builder has not been worked on or updated in recent months. This was identified as a post-1.7 work item owned by the ArchInfra SIG during a discussion...

@zhaochl Sorry for the long delay. First thank you for providing the fix! Could you describe in more detail how to reproduce the error? Is it during create schema, insert...

Please provide onnx file for debugging.

Is the input some kind of image data in (N x C x H x W)? The problem is coming from the mismatch between input and attributes for maxpool. If...

Can you git clone, run `python util/get_version.py `, and share the environment setup? Can you please provide the onnx file also for debugging?

Environment looks good. Can you possibly share the onnx file so I can investigate the issue?

@devloper13 The model was converted successfully in my Ubuntu environment. Unfortunately I don't have Windows to verify. Hope someone in the community can help investigation further.

You should be able to see the inputs and outputs, name, shape, and data type with the following ``` m = tf.saved_model.load(model_path) print('model serving inputs=>', m.signatures['serving_default'].structured_input_signature) print('model serving outputs=>', m.signatures['serving_default'].structured_outputs)...

Please retry with the latest from master branch. Make sure you have tf 2.x, addons, and onnx can be either master or pip install.

@Ehtasha Have you tried it lately? Please use source build from the tf-1.x branch if you have tensorflow 1.x, and do source build from the master branch for tensorflow 2.x....