torch-mlir/python
Sean Silva ef9e4c95f2 torch_mlir.compile: add support for dynamic sizes.
We do this by inroducing a TensorPlaceholder class, which can be used to
specify dynamic sizes. Internally, we canonicalize all example inputs
to TensorPlaceholder's.

This commit also adds some basic testing, which was missing before.
2022-05-17 07:02:32 -07:00
..
test torch_mlir.compile: add support for dynamic sizes. 2022-05-17 07:02:32 -07:00
torch_mlir torch_mlir.compile: add support for dynamic sizes. 2022-05-17 07:02:32 -07:00
torch_mlir_e2e_test Add shape info to `rand_like` + support for `dtype` flag (#851) 2022-05-12 16:00:59 -07:00
CMakeLists.txt Add a new `torch_mlir.compile` method. 2022-04-20 10:06:01 -07:00
TorchMLIRModule.cpp Dual license the torch-mlir project. 2021-10-01 10:46:08 -07:00