int-brain-lab / ibleatools / 29150248236
60%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 27
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 55.333% (+11.5%) from 43.823%
29150248236

push

github

web-flow 
Add compute_features_from_pid_oop (OOP-backed twin of compute_features_from_pid)

* Add engine feature_params plumbing, CSD scale flag, and rms_lf_no_car

PR1 of the OOP feature-computation refactor. In-place, additive engine changes
with no dependency on the feature_calculators package (avoids a circular import):

- features.py: add `scale: bool = True` to csd() and forward it to both
  current_source_density calls (n=2 and n=1). Add rms_lf_no_car to ModelLfFeatures
  and distance_to_tip_um (channel metadata, ASCII "um") to ChannelDataFrameSchema.
- feature_computation.py: add duck-typed `feature_params` to compute_features_from_raw
  (defaults reproduce today's kwargs exactly). Compute rms_lf_no_car inline in the lf
  branch, gated by feature_params.lf.compute_rms_no_car (default off, so no extra
  destripe unless requested).
- aggregation.py: coerce the channels.pqt merge keys to Int64 and thread a
  presence-filtered distance_to_tip_um through as pass-through channel metadata. Because
  it lives only in ChannelDataFrameSchema it never enters the median groupby or the
  denoise path (no skip-list). No-op until channels.pqt carries the column.
- tests: new tests/test_engine_params.py; update voltage feature-set counts in
  test_features.py (52->53, 35->36) for the new rms_lf_no_car feature.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Add compute_features_from_pid_oop delegating to IBLPIDFeatureCalculator

Introduce the OOP feature-calculator layer (ported from oop_refactor) and a parity twin of compute_features_from_pid that delegates to it, so the procedural entry point and the OOP classes share one implementation. The original function is unchanged; the twin is added alongside for verification and will replace it once parity is confirmed (PR3).

- Port feature_calculators/{types,provenance,base,ibl,spikeglx}.py + a trimmed __init__ (nwb deferred).
- Add IBLPIDFeatureCalculator._resolve_channel_labels cbin path, mirrori... (continued)

1520 of 2747 relevant lines covered (55.33%)

0.55 hits per line

Coverage Regressions

Lines Coverage ∆ File
114
52.25
 20.77% feature_computation.py
55
77.94
 0.11% features.py
4
93.1
 18.33% aggregation.py
Jobs
ID Job ID Ran Files Coverage
1 29150248236.1 27
55.33
 GitHub Action Run
Source Files on build 29150248236