Skip to content

Sparse Semi Sturcutred Tutorial audit #3599

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Sep 29, 2025
Merged

Sparse Semi Sturcutred Tutorial audit #3599

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
9e1eb9e
Sparse Semi Sturcutred Tutorial audit
jcaip Sep 29, 2025
52f8823
PR review feedback
jcaip Sep 29, 2025
42f4887
Merge branch 'main' into jcaip/sparse-audit
jcaip Sep 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions advanced_source/semi_structured_sparse.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,8 @@
# - A NVIDIA GPU with semi-structured sparsity support (Compute
# Capability 8.0+).
#
# .. note:: This tutorial is tested on an NVIDIA A100 80GB GPU. You may not see similar speedups on newer GPU architectures, For the latest information on semi-structured sparsity support, please refer to the README `here <https://github.com/pytorch/ao/tree/main/torchao/sparsity#torchao-sparsity>
#
# This tutorial is designed for beginners to semi-structured sparsity and
# sparsity in general. For users with existing 2:4 sparse models,
# accelerating ``nn.Linear`` layers for inference with
Expand All @@ -52,7 +54,6 @@
import torch
from torch.sparse import to_sparse_semi_structured, SparseSemiStructuredTensor
from torch.utils.benchmark import Timer
SparseSemiStructuredTensor._FORCE_CUTLASS = True

# mask Linear weight to be 2:4 sparse
mask = torch.Tensor([0, 0, 1, 1]).tile((3072, 2560)).cuda().bool()
Expand Down Expand Up @@ -207,7 +208,6 @@
import transformers

# force CUTLASS use if ``cuSPARSELt`` is not available
SparseSemiStructuredTensor._FORCE_CUTLASS = True
torch.manual_seed(100)

# Set default device to "cuda:0"
Expand Down
Loading
Loading