✨ feat: Feature extraction with an identifier

[NielsPraet]

Closes #63

Adds 2 arguments to the FeatureCollection.calculate method:

• group_by_all: creates groups that contains all rows corresponding to the group value
  • Note that this is +/- identical as passing df.groupby(group_by_all) as data to the .calculate method -> (which is now also a valid input for the data argument 🎉)
• group_by_consecutive: creates groups that contain consecutive rows for the group value

Both grouped feature extraction approaches ignore NaNs in the group_by column.

---

Limitations: currently restricted to grouping on only a single column.

---

TODOs

• fix CI-CD benchmarking

[codecov-commenter]

Codecov Report

Attention: 3 lines in your changes are missing coverage. Please review.

Comparison is base (31959d1) 97.91% compared to head (a538af4) 98.02%.

Files	Patch %	Lines
tsflex/features/feature_collection.py	98.18%	3 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #109      +/-   ##
==========================================
+ Coverage   97.91%   98.02%   +0.11%     
==========================================
  Files          23       23              
  Lines        1249     1370     +121     
==========================================
+ Hits         1223     1343     +120     
- Misses         26       27       +1     

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

[codspeed-hq]

CodSpeed Performance Report

Merging #109 will degrade performances by 55.47%

_{Comparing NielsPraet:feat/identifier-feature-extraction (45aa8bd) with main (31959d1)}

Summary

❌ 113 regressions

🆕 268 new benchmarks
⁉️ 226 dropped benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	main	NielsPraet:feat/identifier-feature-extraction	Change
🆕	test_single_series_feature_collection[5s-10s-1-sum]	N/A	247.3 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-1-mean]	N/A	352.7 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-1-std]	N/A	725.1 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-1-amin]	N/A	223.9 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-1-amax]	N/A	223.9 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-1-var]	N/A	661.9 ms	N/A
❌	test_single_series_feature_collection[5s-10s-2-sum]	137.4 ms	227.5 ms	-39.61%
❌	test_single_series_feature_collection[5s-10s-2-mean]	233.2 ms	351.1 ms	-33.57%
❌	test_single_series_feature_collection[5s-10s-2-amax]	136.3 ms	223.6 ms	-39.01%
❌	test_single_series_feature_collection[5s-10s-2-median]	547.5 ms	896.7 ms	-38.94%
❌	test_single_series_feature_collection[5s-10s-2-std]	506.9 ms	719.3 ms	-29.53%
❌	test_single_series_feature_collection[5s-10s-2-var]	464.3 ms	661.1 ms	-29.77%
❌	test_single_series_feature_collection[5s-10s-2-amin]	136.2 ms	223.6 ms	-39.06%
🆕	test_single_series_feature_collection[5s-10s-4-amax]	N/A	223.5 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-4-sum]	N/A	227.5 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-4-amin]	N/A	223.5 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-4-std]	N/A	719.2 ms	N/A
🆕	test_single_series_feature_collection[5s-10s-1-median]	N/A	902.1 ms	N/A
🆕	test_single_series_feature_collection[5s-30s-1-sum]	N/A	227.4 ms	N/A
🆕	test_single_series_feature_collection[5s-30s-1-amin]	N/A	223.3 ms	N/A
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

[jvdd]

LGTM! Waiting for your review @jonasvdd

[jvdd]

@jonasvdd ready to be merged i.m.o.