Yiqing-Zhou
|
5e6b747baf
|
[fix] add patch to fix DeepSpeedStrategy offload 'zero_force_ds_cpu_optimizer' issue
|
2023-05-09 23:00:28 +08:00 |
|
Yiqing-Zhou
|
8a5e2043bb
|
[optimize] map_location='cpu' for load_from_checkpoint
|
2023-05-09 00:37:52 +08:00 |
|
Yiqing-Zhou
|
3f92bbbaa2
|
[feature] new args learning_rate max_epochs
|
2023-05-09 00:35:28 +08:00 |
周以晴
|
dc2941f790
|
Create LICENSE
|
2023-05-08 00:26:38 +08:00 |
|
Yiqing-Zhou
|
70ff2acaf0
|
[fix] add patch to fix FSDPStrategy checkpoint issue
|
2023-05-07 16:51:57 +08:00 |
|
Yiqing-Zhou
|
5392a845f7
|
[feature] export model checkpoint from pl.LightningModule
|
2023-05-07 13:18:58 +08:00 |
|
Yiqing-Zhou
|
09507449f7
|
[code] refactor
|
2023-05-07 13:01:02 +08:00 |
|
Yiqing-Zhou
|
939be31c10
|
[feature] new arg use_tril_attention_mask
|
2023-05-06 21:06:18 +08:00 |
|
Yiqing-Zhou
|
0324eb4103
|
[code] update requirements
|
2023-05-06 21:05:53 +08:00 |
|
Yiqing-Zhou
|
9b7f9b9d60
|
[feature] persistent_workers; new args accumulate_grad_batches strategy
|
2023-05-06 19:39:24 +08:00 |
|
Yiqing-Zhou
|
45fa065530
|
Initial Commit
|
2023-05-04 21:52:25 +08:00 |