Skip to content
Permalink

Comparing changes

This is a direct comparison between two commits made in this repository or its related repositories. View the default comparison for this range or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: Xilinx/brevitas
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: dfe1f145d738f8adad09a2e897c29b153c289818
Choose a base ref
..
head repository: Xilinx/brevitas
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: 373e6ec66d39784fa406ae283abd60184e7c8abc
Choose a head ref
2 changes: 1 addition & 1 deletion requirements/requirements-llm.txt
Original file line number Diff line number Diff line change
@@ -1,3 +1,3 @@
# optimum-amd[brevitas] @ git+https://github.com/huggingface/optimum-amd.git@main
tqdm
transformers
transformers[sentencepiece]==4.45.2
5 changes: 5 additions & 0 deletions src/brevitas/core/scaling/runtime.py
Original file line number Diff line number Diff line change
@@ -199,6 +199,11 @@ def __init__(
restrict_scaling_impl: Module = FloatRestrictValue(),
restrict_threshold_impl: Optional[Module] = None) -> None:
super(RuntimeDynamicGroupStatsScaling, self).__init__()

# Ensure retro-compatibility with shared threshold/scaling restrict
if restrict_threshold_impl is None:
restrict_threshold_impl = restrict_scaling_impl

self.group_size = group_size
self.group_dim = group_dim
self.scaling_stats_impl = scaling_stats_impl
Loading