github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-09-09 04:50:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

[setup] support pre-build and jit-build of cuda kernels (#2374)

Browse Source 
* [setup] support pre-build and jit-build of cuda kernels

* polish code

* polish code

* polish code

* polish code

* polish code

* polish code
This commit is contained in:
Frank Lee
2023-01-06 20:50:26 +08:00
committed by GitHub
parent 12c8bf38d7
commit 40d376c566
36 changed files with 414 additions and 390 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
colossalai/nn/optimizer/fused_lamb.py
View File

@@ -76,7 +76,8 @@ class FusedLAMB(torch.optim.Optimizer):
max_grad_norm=max_grad_norm)
super(FusedLAMB, self).__init__(params, defaults)
if multi_tensor_applier.available:
from colossalai.kernel import fused_optim
from colossalai.kernel.op_builder import FusedOptimBuilder
fused_optim = FusedOptimBuilder().load()
self.multi_tensor_l2norm = fused_optim.multi_tensor_l2norm
# Skip buffer