HELSON 1468e4bcfc [zero] add constant placement policy (#1705) ... * fixes memory leak when paramter is in fp16 in ZeroDDP init. * bans chunk releasement in CUDA. Only when a chunk is about to offload, it is allowed to release. * adds a constant placement policy. With it, users can allocate a reserved caching memory space for parameters.		2022-10-14 17:53:16 +08:00
..
components_to_test	[NFC] polish test component gpt code style (#1567)	2022-09-08 16:34:09 +08:00
test_amp	[test] refactored with the new rerun decorator (#763)	2022-04-15 00:33:04 +08:00
test_auto_parallel/test_tensor_shard	[autoparallel] refactored the autoparallel module for organization (#1706)	2022-10-14 13:27:00 +08:00
test_comm	[communication] add p2p_v2.py to support communication with List[Any] (#1407)	2022-08-09 11:40:04 +08:00
test_config	[pipeline] refactor the pipeline module (#1087)	2022-06-10 11:27:38 +08:00
test_context	[test] refactored with the new rerun decorator (#763)	2022-04-15 00:33:04 +08:00
test_data	[unittest] refactored unit tests for change in dependency (#838)	2022-04-22 15:39:07 +08:00
test_data_pipeline_tensor_parallel	[engin/schedule] use p2p_v2 to recontruct pipeline_schedule (#1408)	2022-08-12 11:33:26 +08:00
test_ddp	[feature] A new ZeRO implementation (#1644)	2022-10-09 09:18:51 +08:00
test_device	[tensor] support runtime ShardingSpec apply (#1453)	2022-08-19 13:39:51 +08:00
test_engine	[hotfix] remove potiential circle import (#1307)	2022-07-14 13:44:26 +08:00
test_fx	[autoparallel] fix C version rotor inconsistency (#1691)	2022-10-12 15:21:58 +08:00
test_gemini	[zero] add constant placement policy (#1705)	2022-10-14 17:53:16 +08:00
test_layers	[embedding] rename FreqAwareEmbedding -> CachedEmbedding (#1699)	2022-10-13 22:22:27 +08:00
test_moe	[moe] initialize MoE groups by ProcessGroup (#1640)	2022-09-23 17:20:41 +08:00
test_ops	[FAW] export FAW in _ops (#1438)	2022-08-11 13:43:24 +08:00
test_optimizer	[hotfix] fix CPUAdam kernel nullptr (#1410)	2022-08-05 19:45:45 +08:00
test_pipeline	[pipeline/pytree] add pytree to process args and kwargs | provide `data_process_func` to process args and kwargs after forward (#1642)	2022-09-29 10:58:58 +08:00
test_tensor	[feature] A new ZeRO implementation (#1644)	2022-10-09 09:18:51 +08:00
test_trainer	[pipeline] refactor the pipeline module (#1087)	2022-06-10 11:27:38 +08:00
test_utils	[utils] Add use_reetrant=False in utils.activation_checkpoint (#1460)	2022-08-16 15:39:20 +08:00
test_zero	[feature] A new ZeRO implementation (#1644)	2022-10-09 09:18:51 +08:00
__init__.py	[zero] Update sharded model v2 using sharded param v2 (#323)	2022-03-11 15:50:28 +08:00