onnxruntime/orttraining/orttraining/python/training
Adam Louly 4ce7bbf6f1
Add LayerSpec Support to ORTPipelineModule (#20410)
### Description
In Deepspeed's Pipeline Parallel Implementation, there is a class used
to instantiate the object after it's moved to the device and assigned in
a stage.

This approach helps reduce peak memory usage. 

In this PR, we're adding support to ORT for wrapping this LayerSpec.
2024-04-23 17:57:08 -07:00
..
amp [Better Engineering] Bump ruff to 0.0.278 and fix new lint errors (#16789) 2023-07-21 12:53:41 -07:00
api Introduce a Nominal Checkpoint for On-Device Training (#19232) 2024-01-30 22:11:25 -08:00
experimental
onnxblock Introduce a Nominal Checkpoint for On-Device Training (#19232) 2024-01-30 22:11:25 -08:00
optim Bump ruff to 0.3.2 and black to 24 (#19878) 2024-03-13 10:00:32 -07:00
ort_triton Support BFloat16 for Triton Codegen (#20353) 2024-04-18 17:15:11 +08:00
ortmodule Add LayerSpec Support to ORTPipelineModule (#20410) 2024-04-23 17:57:08 -07:00
utils Fix and enable few ORTModule Unit Tests (#19847) 2024-03-12 10:49:19 +08:00
__init__.py Bump ruff to 0.3.2 and black to 24 (#19878) 2024-03-13 10:00:32 -07:00
_utils.py Removed all the deprecated python training code and related tests and utils (#18333) 2023-11-17 18:19:21 -08:00
artifacts.py Add support for SGD optimizer in minimal build (#19901) 2024-03-14 11:31:20 -07:00