fix flake8 linting errors and add lint check in CI by nilchia · Pull Request #143 · BIMSBbioinfo/flexynesis

nilchia · 2026-04-03T18:30:23Z

This PR adds a new GitHub Actions workflow to automatically lint Python scripts using isort and flake8 on pushes and pull requests.

It also fixes current lint errors from flake8.

I ran black for reformating, flake8, and isort file lint checkin.

Each error is fixed in a commit.

The lint error of long lines is fixed with claude.

Copilot

Pull request overview

This PR introduces a GitHub Actions workflow to lint the repository’s Python code with isort and flake8, and applies broad formatting/import cleanups across the codebase to address existing flake8 issues.

Changes:

Added a CI workflow that runs isort and flake8 on pushes/PRs affecting *.py files.
Reformatted multiple modules (Black-style wrapping) and reorganized imports to reduce lint violations.
Updated a few tests and utility scripts to comply with linting rules (e.g., line length, unused imports).

Reviewed changes

Copilot reviewed 19 out of 19 changed files in this pull request and generated 8 comments.

Show a summary per file

File	Description
`tests/unit/test_smoke.py`	Marks unused import with `noqa` to satisfy flake8.
`tests/test_mps_device.py`	Formatting changes; introduces a regression in memory allocation test and unused locals.
`flexynesis/modules.py`	Import cleanup and formatting adjustments.
`flexynesis/models/triplet_encoder.py`	Import/format refactor; introduces critical bugs in Captum attribution path.
`flexynesis/models/supervised_vae.py`	Import and formatting cleanups for lint compliance.
`flexynesis/models/gnn_early.py`	Import/format updates; minor naming mismatch in helper.
`flexynesis/models/direct_pred.py`	Import and formatting cleanups for lint compliance.
`flexynesis/models/crossmodal_pred.py`	Import and formatting cleanups for lint compliance.
`flexynesis/models/__init__.py`	Reorders imports and formats `__all__`.
`flexynesis/main.py`	Large formatting refactor; introduces a hard-to-read/incorrect early-stop condition expression.
`flexynesis/inference.py`	Formatting and minor refactors to imports / long lines.
`flexynesis/generate_coexpression_network.py`	Formatting refactor; exposes an option that isn’t actually implemented.
`flexynesis/feature_selection.py`	Import ordering and formatting updates.
`flexynesis/config.py`	Formatting and consistent quoting in search-space definitions.
`flexynesis/__init__.py`	Formatting and import ordering updates for lazy import wrapper.
`.github/workflows/lint.yml`	New workflow to run `isort` + `flake8` in CI.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-03T18:34:35Z

+                        ),
+                        n_steps=steps_or_samples,
+                    )
+                    attributions = attributions.split(layer_sizes, dim=3)
+                elif method == "GradientShape":
+                    attributions = explainer.attribute(
+                        input_data,
+                        baseline,
+                        additional_forward_args=(
+                            layer_sizes,
+                            target_var,
+                            steps_or_samples,
+                        ),
+                        n_samples=steps_or_samples,
+                    )
+                    attributions = attributions.split(layer_sizes, dim=3)


Typo in method dispatch: branch checks elif method == "GradientShape" but the supported method name elsewhere is "GradientShap". With method="GradientShap", neither branch runs in the num_class == 1 case, leaving attributions undefined. Rename the branch to "GradientShap" (and ensure both numerical + categorical paths handle it).

It is also not related to the linting fixes of this PR, but it looks valid to me.
Should I change it to "Gradientshap"?

yes, that is a typo/bug, should be fixed.

Copilot · 2026-04-03T18:34:36Z

 def generate_coexpression_network(
    input_file,
    output_file,
-    method='spearman',
+    method="spearman",
    min_correlation=0.3,
    top_k=10,
    remove_self_loops=True,
-    remove_duplicates=True
+    remove_duplicates=True,
 ):


remove_self_loops is exposed in generate_coexpression_network(...) but is never used; self-correlations are always removed in build_network via gene_corr[global_idx] = 0. Either implement the flag (conditionally keep self-loops) or remove the parameter to avoid a non-functional API option.

It is also not related to the linting fixes of this PR, but it looks valid to me.
Should I remove the arg?

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

nilchia · 2026-04-09T21:14:52Z

FileNotFoundError: [Errno 2] No such file or directory: './pancan_pcawg_2020/data_cna.txt'

Looks like this file can not be accessed?

nilchia · 2026-04-09T22:19:54Z

Hi @borauyar,

Can we replace pancan_pcawg_2020 dataset with something else?

borauyar · 2026-04-10T07:45:30Z

Hi @borauyar,

Can we replace pancan_pcawg_2020 dataset with something else?

Hey @nilchia, I don't mind replacing it. I had to change it because there was a problem with the previous MSK dataset. When you download it, the files are just empty. That was breaking the tutorial. If you'd like to change it to something else that demonstrates the cbioportal feature (with at least 2 data modalities), I don't mind changing to something else.

…loaded and unzipped

nilchia · 2026-04-13T12:48:21Z


    def get_cbioportal_data(self, study_id, files=None):
-        archive_path = self.download_study_archive()
-        study_dir = self.extract_archive(archive_path)


This was the culprit.

The commit "b3f9556" deleted those two lines because study_dir was assigned but never referenced again.
The linting flag was correct, but removing the entire call to extract_archive() was the wrong fix.
87631a9 should fix it

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

borauyar · 2026-04-19T10:49:55Z

@nilchia Thank you! Looks like a lot of work :) It took me a while to just go through.

nilchia added 22 commits April 2, 2026 16:27

use black to reformat the scripts

36bafaf

use isort to fix the order

43605c9

remove unused modules

0e8feb8

run black again

7ca24cd

use black to reformat lines to have length < 79

aa8bb8c

remove variables which were never used

b3f9556

fix too many blank lines

c277a1b

fix undefined modules

685a6ed

run black again

d579746

run isort

eaaf05e

fix module level import not at top of the file

d4d8ce2

fix too many blank lines

ec76468

fix f-string is missing placeholders

f89f6f3

fix redefinition of unused

2c0a87a

fix isort

31197d1

fix do not assign a lambda expression, use a def

06155d7

remove unused

a70a745

fix may be undefined or defined from ...

420aa87

fix remaining small fixes except length

5677fbf

add flake8 lint check in CI

5040f06

fix max line length using claude

e185a99

set max len of line to 120 in CI

f686a8d

nilchia requested a review from Copilot April 3, 2026 18:30

Copilot started reviewing on behalf of nilchia April 3, 2026 18:31 View session

Copilot AI reviewed Apr 3, 2026

View reviewed changes

nilchia and others added 4 commits April 3, 2026 20:47

Update .github/workflows/lint.yml

db1cf9d

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

add removed libraries

99b272a

string is imported from data.py

1498d22

isort fix

afa6d77

fix mps memory test

17edca9

nilchia and others added 2 commits April 10, 2026 16:05

Merge branch 'BIMSBbioinfo:main' into lint_py

51b862d

use coad_cptac_2019

8c12ab0

This comment was marked as resolved.

Sign in to view

increase cbioportal timeout and check if the files are correctly down…

154bfbe

…loaded and unzipped

This comment was marked as resolved.

Sign in to view

bring back the removed extract.archive

87631a9

nilchia commented Apr 13, 2026

View reviewed changes

nilchia marked this pull request as ready for review April 13, 2026 13:16

borauyar and others added 2 commits April 19, 2026 12:10

Apply suggestion from @Copilot

f33b98b

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Apply suggestions from code review

4007e35

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

borauyar approved these changes Apr 19, 2026

View reviewed changes

Comment thread flexynesis/main.py

borauyar merged commit 541ba48 into BIMSBbioinfo:main Apr 19, 2026
12 of 14 checks passed

nilchia deleted the lint_py branch April 29, 2026 09:17

Conversation

nilchia commented Apr 3, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI Apr 3, 2026

Choose a reason for hiding this comment

Uh oh!

nilchia Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

borauyar Apr 19, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Copilot AI Apr 3, 2026

Choose a reason for hiding this comment

Uh oh!

nilchia Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

borauyar Apr 19, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

nilchia commented Apr 9, 2026

Uh oh!

nilchia commented Apr 9, 2026

Uh oh!

borauyar commented Apr 10, 2026

Uh oh!

This comment was marked as resolved.

This comment was marked as resolved.

nilchia Apr 13, 2026

Choose a reason for hiding this comment

Uh oh!

nilchia Apr 13, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

borauyar commented Apr 19, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants