torch-mlir/lib/Conversion/TorchOnnxToTorch
Rob Suderman f97cd4893f
[torch] Improve shape inference for dynamic shapes (#3091)
Shapes can be processed as tensors to represent the set of dimensions.
As reshapes take a list of scalars this can result in a single dynamic
dimension blocking the adjacent static dimensions.

This pass attempts to de-couple tensor computations related to shapes
and propagate values to better support lowering scalar tensor
computations.
2024-04-02 16:19:57 -07:00
..
CMakeLists.txt [MLIR][ONNX] Add OnnxToTorch support for Maxpool Op (#2695) 2024-01-12 14:54:38 -08:00
DefaultDomainAtoF.cpp [ONNX] Preliminary Work Towards Supporting QuantizedMLP_basic onnx e2e test (#3089) 2024-04-01 16:21:05 -07:00
DefaultDomainGtoP.cpp [MLIR][ONNX] Fix onnx.gather_nd implementation (#3070) 2024-04-01 20:17:09 +05:30
DefaultDomainQtoZ.cpp [torch] Improve shape inference for dynamic shapes (#3091) 2024-04-02 16:19:57 -07:00
PassDetail.h Initial TorchOnnxToTorch conversion pipeline. (#2585) 2023-11-21 21:02:55 -08:00
Passes.cpp Initial TorchOnnxToTorch conversion pipeline. (#2585) 2023-11-21 21:02:55 -08:00
Patterns.cpp Initial TorchOnnxToTorch conversion pipeline. (#2585) 2023-11-21 21:02:55 -08:00
TorchOnnxToTorch.cpp Initial TorchOnnxToTorch conversion pipeline. (#2585) 2023-11-21 21:02:55 -08:00
Utils.cpp [onnx] Convert `onnx.QLinearConv` to `torch` (#2851) 2024-02-05 16:09:41 -08:00