guyang3532 ba9a585fcc Fix the tensor save for backward release problem (#13679) ... Motivation: PythonOp is saving input for backward, it's risky since ONNX Runtime backend is not aware of this, the tensor buffer may be "released" by ORT, then potentially modified by other operators before backward function executes. Fix: This pr just clone all input of PythonOp before forward is invoked. This may be high overhead, it's just a workaround before a better fix.		2022-11-22 17:32:19 +08:00
..
deprecated	Format all python files under onnxruntime with black and isort (#11324)	2022-04-26 09:35:16 -07:00
training	Fix the tensor save for backward release problem (#13679)	2022-11-22 17:32:19 +08:00
checkpointing_utils.py	Format all python files under onnxruntime with black and isort (#11324)	2022-04-26 09:35:16 -07:00
ort_trainer.py	Deprecate ORTTrainer (#13022)	2022-09-23 18:10:09 -07:00
orttraining_pybind_common.h	resolve the provider options before create training session in orttrainer (#9199)	2021-10-12 09:30:45 -07:00
orttraining_pybind_state.cc	Training API to export the eval model to an inference model (#13345)	2022-10-27 09:34:01 -07:00
orttraining_python_module.cc	Allow CUDA EP enable or disable TunableOp via session options and environment variable (#13601)	2022-11-15 14:43:54 +08:00
orttraining_python_module_eager.h	Abjindal/clean eager backend (#10055)	2022-01-19 14:20:09 -08:00
pt_patch.py	Format all python files under onnxruntime with black and isort (#11324)	2022-04-26 09:35:16 -07:00