aljaz/DALLE2-pytorch
1
1
Fork 0
mirror of https://github.com/lucidrains/DALLE2-pytorch.git synced 2025-12-19 09:44:19 +01:00
Code Issues Packages Projects Releases Wiki Activity

Fixed non deterministic optimizer creation (#130)

Browse Source
This commit is contained in:
Aidan Dempster
2022-05-31 12:03:20 -04:00
committed by GitHub
parent 6f8b90d4d7
commit 09534119a1
1 changed files with 6 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
dalle2_pytorch/optimizer.py
View File

@@ -1,8 +1,10 @@
from torch.optim import AdamW, Adam
def separate_weight_decayable_params(params):
no_wd_params = set([param for param in params if param.ndim < 2])
wd_params = set(params) - no_wd_params
wd_params, no_wd_params = [], []
for param in params:
param_list = no_wd_params if param.ndim < 2 else wd_params
param_list.append(param)
return wd_params, no_wd_params
def get_optimizer(
@@ -25,8 +27,8 @@ def get_optimizer(
wd_params, no_wd_params = separate_weight_decayable_params(params)
params = [
{'params': list(wd_params)},
{'params': list(no_wd_params), 'weight_decay': 0},
{'params': wd_params},
{'params': no_wd_params, 'weight_decay': 0},
]
return AdamW(params, lr = lr, weight_decay = wd, betas = betas, eps = eps)