Add test and delete lint errors by theodchrn · Pull Request #61 · MobsLab/neuroEncoders

theodchrn · 2026-02-04T15:16:18Z

clean: remove old notebooks
fix: typos and minor errors in dealing w bayes matrices
add: tests and possibility to save and load datasets to tfrec and parquet format
improve: plotting modules
improve: test units
change global classes behaviour: no need for duplicate in Mouse_Results, remove dependency on window size and fix copy/paste issues for datahelper (different phases kepts the same Ref)
add: possibility to randomize the spikes inside the dataset for control
add: helper function for comparing bayes and ann
add: data helper behaviour without windowsize ms
fix: lots of typos and ruff linting issues

parquet format

remove dependency on window size and fix copy/paste issues for datahelper (different phases kepts the same Ref) This allows for lighter mouse results, more focuses ref on datahelper

Copilot

Pull request overview

This PR performs extensive code cleanup and adds new functionality for dataset management. The main objectives are to fix linting errors, remove unused code, and add capabilities for saving/loading datasets in TFRecord and Parquet formats.

Changes:

Fixed numerous linting issues: replaced bare except clauses with specific exceptions, converted lambda functions to named functions, fixed type comparisons to use isinstance()
Removed old notebooks and deprecated code (old_code.py with ~1800 lines)
Added dataset saving/loading functionality for TFRecord and Parquet formats
Refactored global classes to remove window size dependencies and fix copy/paste issues
Added spike randomization capability for control experiments
Improved plotting modules and test coverage

Reviewed changes

Copilot reviewed 31 out of 38 changed files in this pull request and generated 26 comments.

Show a summary per file

File	Description
uv.lock, pyproject.toml	Added pyarrow, pdf2image, python-pptx dependencies
tests/test_*.py	Improved test coverage, removed unused variables, added dataset saving tests
runAllMice.py	Logic changes for force flag, mouse filtering, experiment selection; added dry-run flag
neuroencoders/utils/*	Extensive linting fixes, improved exception handling, refactored global classes
neuroencoders/fullEncoder/an_network.py	Added dataset save/load functionality, model rebuilding methods
neuroencoders/simpleBayes/decode_bayes.py	Added spike loading methods, fixed lambda to named function
neuroencoders/importData/*	Linting fixes, improved GUI time formatting, added linear spike sorting
neuroencoders/resultAnalysis/*	Removed old_code.py, updated function calls for API changes
neuroEncoder	Updated function calls, removed try-except wrappers, added debug print
.pre-commit-config.yaml	Fixed YAML formatting, removed --exit-zero from ruff

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

Copilot · 2026-02-04T15:21:58Z

neuroEncoder

        helper.get_true_target(
-            l_function, in_place=True, show=show_figures, speedMask=True
+            windowSizeMS, l_function, in_place=True, show=show_figures, speedMask=True
        )


Incorrect function call: The method signature for get_true_target was changed to accept windowSizeMS as the first parameter (line 436-438), but this call doesn't provide it. This will cause a TypeError since windowSizeMS is now a required positional argument. The call should be: helper.get_true_target(windowSizeMS, l_function, in_place=True, show=show_figures, speedMask=True)

Copilot · 2026-02-04T15:21:58Z

neuroencoders/resultAnalysis/paper_figures.py

+                        if not hasattr(
+                            self.trainerBayes, "spikeMatTimes"
+                        ) or not hasattr(self.trainerBayes, "spikeMat"):
+                            raise ValueError(
+                                """
+                                trainerBayes does not have spikeMatTimes or spikeMat attributes needed to extract spike counts.
+                                Make sure to run decoding with extract_spike_counts=True first. You can run trainerBayes.train_order_by_pos.
+                                """
+                            )
+                        total_count, _ = extract_spike_counts_keops(
                            timesBayes[-1], self.trainerBayes.spikeMatTimes, ws / 1000
                        )
                        total_spikes_count.append(total_count)
-                        matrix_count, _ = extract_spike_counts_from_matrix(
+                        matrix_count, _ = extract_spike_counts_matrix_keops(
                            timesBayes[-1],
-                            self.trainerBayes.spikeMat,
+                            self.trainerBayes.spikeMatLabels,
                            self.trainerBayes.spikeMatTimes,
                            ws / 1000,
                        )


Missing null handling: The try-except at lines 529-537 attempts to extract spike counts, but if trainerBayes.spikeMatTimes or trainerBayes.spikeMat don't exist (as checked in lines 529-531), the code will raise a ValueError but then continues to try accessing these attributes at lines 538-547. The control flow should either return early after raising the ValueError or ensure these attributes exist before proceeding.

Copilot · 2026-02-04T15:21:58Z

runAllMice.py

            "--target",
            target_bayes,
-            "--flat_prior",
+            # "--flat_prior",


Commented out critical parameter: The '--flat_prior' flag has been commented out at line 251. This changes the behavior of the Bayesian decoder. If this is intentional, consider removing the line entirely or adding a comment explaining why. If it's temporary for testing, it should be uncommented before merging.

Copilot · 2026-02-04T15:21:59Z

neuroencoders/utils/global_classes.py


        # regarding data augmentation
-        self.dataAugmentation = kwargs.pop("dataAugmentation", True)
+        self.dataAugmentation = kwargs.pop("dataAugmentation", False)


The dataAugmentation default has been changed from True to False (line 1706). This is a significant behavior change that will affect model training. Ensure this is intentional and documented, as it may impact model performance. Consider whether this should be mentioned in the PR description or release notes.

Suggested change

self.dataAugmentation = kwargs.pop("dataAugmentation", False)

self.dataAugmentation = kwargs.pop("dataAugmentation", True)

Copilot · 2026-02-04T15:21:59Z

neuroencoders/utils/global_classes.py

+        self.xyOutput, positions = self._get_XYOutput_morph_maze(
            positions, self.shock_zone_mask, self.ref, self.ratioIMAonREAL
        )


Missing return statement: The function _get_XYOutput_morph_maze now returns both XYOutput and positions (line 1027), but the assignment at line 865 captures only xyOutput, discarding the updated positions. This should be: self.xyOutput, positions = self._get_XYOutput_morph_maze(...). Without this change, the positions variable will not reflect any transformations applied inside the function.

Copilot · 2026-02-04T15:22:03Z

neuroencoders/importData/rawdata_parser.py

                                try:
                                    ls[dim][2][iaxis].remove()
-                                except:
+                                except (AttributeError, KeyError):


'except' clause does nothing but pass and there is no explanatory comment.

Copilot · 2026-02-04T15:22:04Z

neuroencoders/importData/rawdata_parser.py

                            try:
                                ls[dim][2].remove()
-                            except:
+                            except (AttributeError, KeyError):


'except' clause does nothing but pass and there is no explanatory comment.

Copilot · 2026-02-04T15:22:04Z

neuroencoders/importData/rawdata_parser.py

                            try:
                                l3.remove()
-                            except:
+                            except (AttributeError, KeyError):


'except' clause does nothing but pass and there is no explanatory comment.

Copilot · 2026-02-04T15:22:04Z

neuroencoders/importData/gui_elements.py

                    try:
                        mask = indices == frame
-                    except:
+                    except (TypeError, AttributeError):


This statement is unreachable.

Copilot · 2026-02-04T15:22:04Z

neuroencoders/importData/gui_elements.py

            try:
                mask = indices == frame
-            except:
+            except (TypeError, AttributeError):


This statement is unreachable.

theodchrn · 2026-02-05T11:22:37Z

@copilot open a new pull request to apply changes based on the comments in this thread

Copilot · 2026-02-05T11:22:47Z

@theodchrn I've opened a new pull request, #62, to work on those changes. Once the pull request is ready, I'll request review from you.

…ming Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

…g builtin Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

…nused code Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

Address PR review comments: remove debug code, fix variable shadowing, improve error handling

theodchrn added 10 commits January 15, 2026 16:53

clean: remove old notebooks

7da37a7

fix: typos and minor errors in dealing w bayes matrices

e381a8e

add: tests and possibility to save and load datasets to tfrec and

9d7d7a7

parquet format

improve: plotting modules

840196b

improve: test units

b576664

change global classes behaviour: no need for duplicate in Mouse_Results,

e5302f4

remove dependency on window size and fix copy/paste issues for datahelper (different phases kepts the same Ref) This allows for lighter mouse results, more focuses ref on datahelper

add: possibility to randomize the spikes inside the dataset for control

bd17f14

add: helper function for comparing bayes and ann

8c665ad

add: data helper behaviour without windowsize ms

84bd76e

fix: lots of typos and ruff linting issues

701805e

Copilot AI review requested due to automatic review settings February 4, 2026 15:16

Copilot started reviewing on behalf of theodchrn February 4, 2026 15:16 View session

Copilot AI reviewed Feb 4, 2026

View reviewed changes

Initial plan

475d9de

Copilot AI mentioned this pull request Feb 5, 2026

Address PR review comments: remove debug code, fix variable shadowing, improve error handling #62

Merged

Copilot AI and others added 5 commits February 5, 2026 11:26

Fix review comments: remove debug code, add comments, fix variable na…

8d8a660

…ming Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

Remove unused variables and rename dir to directory to avoid shadowin…

0f0bbc1

…g builtin Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

Fix linting issues: remove unused assignment, fix imports, clean up u…

8c9b092

…nused code Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

Improve exception handling and variable naming clarity

4bc80c7

Co-authored-by: theodchrn <48409582+theodchrn@users.noreply.github.com>

Merge pull request #62 from MobsLab/copilot/sub-pr-61

f754b45

Address PR review comments: remove debug code, fix variable shadowing, improve error handling

	self.dataAugmentation = kwargs.pop("dataAugmentation", False)
	self.dataAugmentation = kwargs.pop("dataAugmentation", True)

Conversation

theodchrn commented Feb 4, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

theodchrn commented Feb 5, 2026

Uh oh!

Copilot AI commented Feb 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants