onnx / onnx-coreml

ONNX to Core ML Converter
MIT License
392 stars 79 forks source link

Error in neural network compiler computing minimum sequence length for the model #533

Open Penelope1003 opened 4 years ago

Penelope1003 commented 4 years ago

❓Question

add network input to output, then get this warning, why and how to avoid it?

1/25: Converting Node Type Conv 2/25: Converting Node Type Relu 3/25: Converting Node Type Conv 4/25: Converting Node Type Conv 5/25: Converting Node Type Relu 6/25: Converting Node Type Conv 7/25: Converting Node Type Conv 8/25: Converting Node Type Add 9/25: Converting Node Type Conv 10/25: Converting Node Type Conv 11/25: Converting Node Type Relu 12/25: Converting Node Type Conv 13/25: Converting Node Type Conv 14/25: Converting Node Type Add 15/25: Converting Node Type Conv 16/25: Converting Node Type Reshape 17/25: Converting Node Type Transpose 18/25: Converting Node Type Reshape 19/25: Converting Node Type Transpose 20/25: Converting Node Type Reshape 21/25: Converting Node Type PRelu 22/25: Converting Node Type Conv 23/25: Converting Node Type Upsample 24/25: Converting Node Type Add 25/25: Converting Node Type Relu Translation to CoreML spec completed. Now compiling the CoreML model. /python2.7/site-packages/coremltools/models/model.py:111: RuntimeWarning: You will not be able to run predict() on this Core ML model. Underlying exception message was: Error compiling model: "compiler error: Error in neural network compiler computing minimum sequence length for the model.". RuntimeWarning) Model Compilation done.

oavrahami commented 4 years ago

Hi @Penelope1003 , Did you manage to solve this issue?