Remove intervals dimension from decoding results

[Copilot]

The intervals dimension caused memory waste from zero-padding and confused users with its 3D structure. Decoding results now concatenate along time with an interval_labels coordinate for grouping.

Changes

• Core logic (sorted_spikes.py, clusterless.py): Changed xr.concat(results, dim="intervals") to dim="time" with interval_labels=("time", [0,0,...,1,1,...]) coordinate
• Tests: Updated mock decoders and added tests validating no intervals dimension and proper interval_labels tracking

Impact

Memory: 12-48% reduction by eliminating padding zeros (varies with interval length distribution)

Structure:

• Before: (n_intervals, max_time, n_position) with padding
• After: (total_time, n_position) without padding

Usage:

results = table.fetch_results()

# Select specific interval
interval_0 = results.sel(time=results.interval_labels == 0)

# Iterate intervals
for idx, data in results.groupby('interval_labels'):
    process(data)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Get rid of intervals dimension in decoding.</issue_title>
<issue_description>The intervals dimension often confuses people and can take up more memory because we have to pad to the longest time series with zeros.

We should store it as one time series. There can be an additional variable that tracks the intervals so that people can do a groupby operation or selection to get the interval.</issue_description>

<agent_instructions>Use TDD</agent_instructions>

Comments on the Issue (you are @copilot in this section)

• Fixes Get rid of intervals dimension in decoding. #1474

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[Copilot]

Pull Request Overview

This PR removes the intervals dimension from decoding results and replaces it with concatenation along the time dimension, adding an interval_labels coordinate to track interval membership. This change eliminates memory waste from zero-padding shorter time series and reduces confusion for users.

Key changes:

• Concatenate decoding results along time instead of intervals dimension
• Add interval_labels coordinate for tracking which interval each time point belongs to
• Update tests to verify the new structure and demonstrate 12.4% memory savings

Reviewed Changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
src/spyglass/decoding/v1/sorted_spikes.py	Modified to concatenate results along time dimension and add interval_labels tracking
src/spyglass/decoding/v1/clusterless.py	Modified to concatenate results along time dimension and add interval_labels tracking
tests/decoding/conftest.py	Updated mock decoder fixtures to simulate the new concatenation behavior
tests/decoding/test_intervals_removal.py	Integration tests verifying the new structure with actual decoder tables
tests/decoding/test_intervals_removal_simple.py	Standalone unit tests demonstrating memory savings and basic functionality

---

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

[Copilot]

The mock decoder handles the single-interval case specially but doesn't handle the case where results_list is empty. Add a check: if not results_list: raise ValueError(\"No intervals to decode\") before the length check.

[Copilot]

The mock decoder handles the single-interval case specially but doesn't handle the case where results_list is empty. Add a check: if not results_list: raise ValueError(\"No intervals to decode\") before the length check.

[codecov]

Codecov Report

❌ Patch coverage is 44.44444% with 10 lines in your changes missing coverage. Please review.
✅ Project coverage is 71.09%. Comparing base (32b6950) to head (281d949).

Files with missing lines	Patch %	Lines
src/spyglass/decoding/v1/sorted_spikes.py	0.00%	9 Missing ⚠️
src/spyglass/decoding/v1/clusterless.py	88.88%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1476      +/-   ##
==========================================
- Coverage   71.12%   71.09%   -0.03%     
==========================================
  Files         114      114              
  Lines       13306    13318      +12     
==========================================
+ Hits         9464     9469       +5     
- Misses       3842     3849       +7     

Flag	Coverage Δ
full-tests	71.09% <44.44%> (-0.03%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.