Skip to content

Fix Windows PermissionError in test_patch_merger #160

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
justinchuby merged 1 commit into from
Apr 10, 2026
Merged

Fix Windows PermissionError in test_patch_merger #160

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
28 changes: 15 additions & 13 deletions src/mobius/components/_pixtral_vision_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,8 +115,6 @@ def test_patch_merger_matches_hf_unfold_ordering():
the outermost loop). The ONNX implementation must reproduce this
ordering so the learned ``merging_layer`` projection is correct.
"""
import tempfile

import torch
from onnxscript._internal.builder import GraphBuilder

Expand Down Expand Up @@ -181,17 +179,21 @@ def test_patch_merger_matches_hf_unfold_ordering():

model = ir.Model(graph, ir_version=11)

with tempfile.NamedTemporaryFile(suffix=".onnx", delete=True) as f:
ir.save(model, f.name)
sess = ort.InferenceSession(f.name, providers=["CPUExecutionProvider"])
onnx_out = sess.run(
None,
{
"x": x,
"grid_h": np.array(grid_h, dtype=np.int64),
"grid_w": np.array(grid_w, dtype=np.int64),
},
)[0]
# Serialize to protobuf in-memory (avoids Windows PermissionError
# from concurrent file access with tempfile + ir.save).
proto = ir.serde.serialize_model(model)
sess = ort.InferenceSession(
proto.SerializeToString(),
providers=["CPUExecutionProvider"],
Comment thread
justinchuby marked this conversation as resolved.
)
onnx_out = sess.run(
None,
{
"x": x,
"grid_h": np.array(grid_h, dtype=np.int64),
"grid_w": np.array(grid_w, dtype=np.int64),
},
)[0]

np.testing.assert_allclose(
onnx_out.squeeze(0),
Expand Down
Loading