ColossalAI/colossalai/fx/passes/algorithms
Super Daniel 3dd6994427
[fx/profiler] assigned UUID to each unrecorded tensor/ improved performance on GPT-2 (#1679)
* [fx/profiler] modify data_ptr into uuid for all tensors.

* [fx] modify uuid.

* [fx/profiler] tune performance on GPT-2.

* [fx] updates.

* [fx] debug.

* [fx] debug.

* [fx] cuda.
2022-10-11 11:03:35 +08:00
..
__init__.py [fx] Add pofo solver (#1608) 2022-09-20 11:20:48 +08:00
build_c_ext.py [autoparallel] add rotor C version (#1658) 2022-10-03 17:13:30 +08:00
ckpt_solver_chen.py [fx/profiler] assigned UUID to each unrecorded tensor/ improved performance on GPT-2 (#1679) 2022-10-11 11:03:35 +08:00
ckpt_solver_pofo.py [fx] refactor code for profiler / enable fake tensor movement. (#1646) 2022-09-27 10:26:52 +08:00
ckpt_solver_rotor.py [fx/profiler] assigned UUID to each unrecorded tensor/ improved performance on GPT-2 (#1679) 2022-10-11 11:03:35 +08:00
dynamic_programs.c [autoparallel] add rotor C version (#1658) 2022-10-03 17:13:30 +08:00
linearize.py [fx/tuning] tune performance on rotor with meta info. (#1599) 2022-09-15 14:46:36 +08:00
operation.py [autoparallel] Add pofo sequence annotation (#1637) 2022-09-24 01:52:57 +08:00