evan/ai-toolkit
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Set gradient checkpointing on unet enabled by default. Help out immensly with sdxl backprop spikes

Browse Source
This commit is contained in:
Jaret Burkett committed 2023-08-01 15:43:27 -06:00
1 parent f53fd08690
commit 2bf3e529ce
2 files changed
+4 -1
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

toolkit/config_modules.py
+1 -1
View File
@@ -60,7 +60,7 @@ class TrainConfig:
self.noise_offset = kwargs.get('noise_offset', 0.0)
self.optimizer_params = kwargs.get('optimizer_params', {})
self.skip_first_sample = kwargs.get('skip_first_sample', False)
self.gradient_checkpointing = kwargs.get('gradient_checkpointing', False)
self.gradient_checkpointing = kwargs.get('gradient_checkpointing', True)
class ModelConfig: