Skip to content

Add set_local support for static quant with pt2e #1870

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 9 commits into from
Jun 20, 2024
Merged

Add set_local support for static quant with pt2e #1870

Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
c036ee0
support set_local for pt2e
yiliu30 Jun 14, 2024
b9ca5e9
fixed ut
yiliu30 Jun 17, 2024
f82b9c9
use np 1.26
yiliu30 Jun 17, 2024
33ea758
Merge branch 'master' into set_local
yiliu30 Jun 17, 2024
0f3618a
fix coverage
yiliu30 Jun 18, 2024
d8021e2
remove .coverage
yiliu30 Jun 18, 2024
2182fcb
Merge branch 'master' into set_local
yiliu30 Jun 19, 2024
256fc0a
disable coverage check
yiliu30 Jun 19, 2024
4b303f5
disable some check
yiliu30 Jun 20, 2024
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
Prev Previous commit
disable some check 
Signed-off-by: yiliu30 <[email protected]>
  • Loading branch information
@yiliu30
yiliu30 committed Jun 20, 2024
commit 4b303f54d85d4275a546753f627c710b09f7c1a6
4 changes: 2 additions & 2 deletions neural_compressor/torch/algorithms/pt2e_quant/half_precision_rewriter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ def is_target_node_in_candidate_list(match, original_graph, pattern_graph):
if node.target == target_op:
target_node = node
break
if target_node is None:
if target_node is None: # pragma: no cover
return False
matched_node = match.nodes_map[target_node]
return matched_node in node_list
Expand Down Expand Up @@ -138,7 +138,7 @@ def get_unquantized_node_set(gm: torch.fx.GraphModule):
if meta := getattr(node, "meta"):
if quantization_annotation := meta.get(xiq.QUANT_ANNOTATION_KEY):
none_annotation = xiq._X86InductorQuantizationAnnotation(_annotated=True)
if quantization_annotation != none_annotation:
if quantization_annotation != none_annotation: # pragma: no cover
continue
unquantized_node_set.add(node)
return unquantized_node_set
Expand Down