[MXNET-1252] Decouple NNVM to ONNX from NNVM to TensorRT conversion

• Type: Improvement
• Status: In Progress
• Priority: Major
• Resolution: Unresolved
• Assignee: Kellen Sunderland
• Reporter: Kellen Sunderland
• Created: 05/Dec/18 01:57
• Updated: 19/Dec/18 18:39

Description

The current nnvm_to_onnx classes are tightly coupled to TensorRT.  We could extract all of the TensorRT specific functionality and have a proper separation between nnvm_to_onnx and onnx_to_tensorrt.  When structuring nnvm_to_onnx we should make use of object hierarchy to convert to specific opsets of onnx to help us maintain compatibility with different toolsets.  We should create a base class that performs generic onnx conversions.  We should then specialized objects that inherit from the base onnx class and take care of the  differences between opsets.  We should also create unit tests on a per-op basis to make sure we're introducing regressions.

Issue Links

• links to: GitHub Pull Request #13659
• mentioned in: Page Loading...