Feature extractors: basics¶

This tutorial introduces the light_curve feature extractor interface: creating features, combining them with Extractor, reading names and descriptions, and batch processing.

Setup¶

We create a synthetic sinusoidal light curve with Gaussian noise:

In [ ]:

import light_curve as lc
import numpy as np

rng = np.random.default_rng(42)
t = np.sort(rng.uniform(0, 100, 200))
m = 15.0 + 0.3 * np.sin(2 * np.pi * t / 20) + rng.normal(0, 0.05, 200)
err = np.full(200, 0.05)

print(f'Observations: {len(t)}')
print(f'Time range:   {t[0]:.1f} – {t[-1]:.1f} days')
print(f'Mag range:    {m.min():.2f} – {m.max():.2f}')

import light_curve as lc import numpy as np rng = np.random.default_rng(42) t = np.sort(rng.uniform(0, 100, 200)) m = 15.0 + 0.3 * np.sin(2 * np.pi * t / 20) + rng.normal(0, 0.05, 200) err = np.full(200, 0.05) print(f'Observations: {len(t)}') print(f'Time range: {t[0]:.1f} – {t[-1]:.1f} days') print(f'Mag range: {m.min():.2f} – {m.max():.2f}')

Observations: 200
Time range:   0.7 – 99.2 days
Mag range:    14.60 – 15.41

Single feature¶

Each feature class is callable. It accepts (t, m, sigma) arrays and returns a NumPy array. The .names attribute lists the output column names:

In [ ]:

amp = lc.Amplitude()
result = amp(t, m, err)
print(f'names:  {amp.names}')
print(f'result: {result}')

amp = lc.Amplitude() result = amp(t, m, err) print(f'names: {amp.names}') print(f'result: {result}')

Traceback (most recent call last):
  File "<string>", line 1, in <module>
    import sys; sys.path.insert(0, '/workdir/light-curve-python/light-curve'); amp = lc.Amplitude()
                                                                                     ^^
NameError: name 'lc' is not defined

.descriptions gives a human-readable explanation of each output:

In [ ]:

print(amp.descriptions)

print(amp.descriptions)

Traceback (most recent call last):
  File "<string>", line 1, in <module>
    import sys; sys.path.insert(0, '/workdir/light-curve-python/light-curve'); print(amp.descriptions)
                                                                                     ^^^
NameError: name 'amp' is not defined

Combining features with Extractor¶

Extractor combines multiple features into a single callable evaluated in one pass:

In [ ]:

ext = lc.Extractor(
    lc.Amplitude(),
    lc.BeyondNStd(nstd=1),
    lc.BeyondNStd(nstd=2),
    lc.StandardDeviation(),
    lc.WeightedMean(),
    lc.LinearFit(),
    lc.StetsonK(),
)
result = ext(t, m, err)
for name, value in zip(ext.names, result):
    print(f'  {name:35s} = {value:.5f}')

ext = lc.Extractor( lc.Amplitude(), lc.BeyondNStd(nstd=1), lc.BeyondNStd(nstd=2), lc.StandardDeviation(), lc.WeightedMean(), lc.LinearFit(), lc.StetsonK(), ) result = ext(t, m, err) for name, value in zip(ext.names, result): print(f' {name:35s} = {value:.5f}')

Traceback (most recent call last):
  File "<string>", line 1, in <module>
    import sys; sys.path.insert(0, '/workdir/light-curve-python/light-curve'); ext = lc.Extractor(
                                                                                     ^^
NameError: name 'lc' is not defined

Batch processing with .many()¶

.many() processes a list of (t, m, sigma) tuples in parallel with reduced Python–Rust call overhead — the preferred path for large datasets:

In [ ]:

n_lc = 5000
light_curves = [
    (
        np.sort(rng.uniform(0, 100, rng.integers(50, 200))),
        15.0 + rng.normal(0, 0.2, rng.integers(50, 200)),
        np.full(rng.integers(50, 200), 0.05),
    )
    for _ in range(n_lc)
]
# Batch extraction — one call for all light curves
results = lc.Amplitude().many(light_curves)
print(f'Extracted from {n_lc} light curves: shape = {results.shape}')
print(f'Mean amplitude = {results.mean():.4f} mag')

n_lc = 5000 light_curves = [ ( np.sort(rng.uniform(0, 100, rng.integers(50, 200))), 15.0 + rng.normal(0, 0.2, rng.integers(50, 200)), np.full(rng.integers(50, 200), 0.05), ) for _ in range(n_lc) ] # Batch extraction — one call for all light curves results = lc.Amplitude().many(light_curves) print(f'Extracted from {n_lc} light curves: shape = {results.shape}') print(f'Mean amplitude = {results.mean():.4f} mag')

Traceback (most recent call last):
  File "<string>", line 4, in <module>
    np.sort(rng.uniform(0, 100, rng.integers(50, 200))),
    ^^
NameError: name 'np' is not defined

Next steps¶

• Feature table — all 40+ extractors grouped by category
• API reference — full signatures and equations
• Periodogram tutorial — Lomb–Scargle and period search