Wei-Sheng Chin b0279b14d8 [DORT] Enable Dynamic Shape in DORT and Use Different InferenceSession's when Inputs Are Not Compatible (#16753) ... Sometimes, ONNX exporter generates rank- or shape-dependent sub-graphs. Thus, error could occur when running the ONNX model with different inputs. This PR ([78e736d](78e736d857)) addresses this problem by - if needed, exporting multiple ONNX models with different inputs for the same GraphModule. - implementing a naive mechanism to determine of existing ONNX models (and the associated InferenceSession) can be reused. On the other hand, in the second commit [b5a9b5f](b5a9b5f849), this PR also enables dynamic shapes in DORT by - passing dynamic_shapes = True to exporter (see how DEFAULT_DYNAMIC_BACKEND is created) - calling torch._dynamo.optimize(dynamic_ort_aot, dynamic=True) (see how dynamic_ort_aot is created).		2023-07-24 16:54:01 -07:00
..
__init__.py	[DORT] Use new FX-to-ONNX exporter (#16450)	2023-07-04 13:13:04 -07:00
ort_backend.py	[DORT] Enable Dynamic Shape in DORT and Use Different InferenceSession's when Inputs Are Not Compatible (#16753)	2023-07-24 16:54:01 -07:00
register_backend.py	[DORT] Enable Dynamic Shape in DORT and Use Different InferenceSession's when Inputs Are Not Compatible (#16753)	2023-07-24 16:54:01 -07:00