gradboard 2.3.0__tar.gz → 2.3.1__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of gradboard might be problematic. Click here for more details.
- {gradboard-2.3.0 → gradboard-2.3.1}/PKG-INFO +1 -1
- {gradboard-2.3.0 → gradboard-2.3.1}/gradboard/optimiser.py +2 -1
- {gradboard-2.3.0 → gradboard-2.3.1}/pyproject.toml +1 -1
- {gradboard-2.3.0 → gradboard-2.3.1}/LICENSE +0 -0
- {gradboard-2.3.0 → gradboard-2.3.1}/README.md +0 -0
- {gradboard-2.3.0 → gradboard-2.3.1}/gradboard/__init__.py +0 -0
- {gradboard-2.3.0 → gradboard-2.3.1}/gradboard/cycles.py +0 -0
- {gradboard-2.3.0 → gradboard-2.3.1}/gradboard/scheduler.py +0 -0
|
@@ -142,7 +142,8 @@ def get_optimiser(model, optimiser=AdamW, lr=1e-3, weight_decay=1e-2):
|
|
|
142
142
|
"norm",
|
|
143
143
|
"embedding",
|
|
144
144
|
"swiglu_beta",
|
|
145
|
-
"sigma_reparam"
|
|
145
|
+
"sigma_reparam",
|
|
146
|
+
"input_query",
|
|
146
147
|
"reentrant_query",
|
|
147
148
|
]:
|
|
148
149
|
weight_decay_exclude += [
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|