light-curve 0.12.5

Feature extractor from noisy time series

light-curve processing toolbox for Python

A Python wrapper for the light-curve-feature and light-curve-dmdt Rust crates, providing high-performance time-series feature extraction for astrophysics.

Full documentation: light-curve.snad.space

Quick start

pip install 'light-curve[full]'

import light_curve as lc
import numpy as np

rng = np.random.default_rng(0)
t = np.sort(rng.uniform(0, 100, 100))
m = 15.0 + 0.01 * t + rng.normal(0, 0.1, 100)
err = np.full(100, 0.1)

# Statistical and variability feature extraction
extractor = lc.Extractor(lc.Amplitude(), lc.BeyondNStd(nstd=1), lc.LinearFit())
result = extractor(t, m, err)
print('\n'.join(f"{name} = {value:.4f}" for name, value in zip(extractor.names, result)))

# dm-dt map — 2D histogram of Δmag vs log-Δt for CNN classifiers
dmdt = lc.DmDt.from_borders(min_lgdt=0, max_lgdt=2, max_abs_dm=1.0, lgdt_size=16, dm_size=16, norm=[])
matrix = dmdt.points(t, m)
print(f"dm-dt map shape: {matrix.shape}")  # (16, 16)

Embed the light curve with a pretrained Astromer2 transformer (model downloads on first use):

from light_curve.embed import Astromer2
import numpy as np

rng = np.random.default_rng(0)
t = np.sort(rng.uniform(0, 100, 100))
m = 15.0 + 0.01 * t + rng.normal(0, 0.1, 100)

model = Astromer2.from_hf(output="mean")   # cached after first download
embedding = model(t, m).squeeze()           # shape (256,)
print(f"Embedding shape: {embedding.shape}")

The package provides:

• 40+ feature extractors — amplitude, period, variability indices, parametric fits, and more. See the Features docs for a full list.
• ML embeddings — embed light curves with pretrained ONNX models (Astromer2, ATCAT). See the Embeddings docs.
• dm-dt maps — 2D histograms of Δm vs log-Δt for CNN input. See the dm-dt docs.

Installation

pip install 'light-curve[full]'
# or
conda install -c conda-forge light-curve-python

The full extra installs optional Python dependencies required by some features (iminuit, scipy).

Binary wheels are available for Linux (x86-64, aarch64), macOS, and Windows. For the full platform support table and build-from-source instructions, see the Installation docs.

Developer guide

Prepare environment

Install a recent Rust toolchain and Python 3.10+. Use rustup to install Rust and keep it updated with rustup update.

Clone the repository, then create and activate a virtual environment:

git clone https://github.com/light-curve/light-curve-python.git
cd light-curve-python/light-curve
python3 -m venv venv
source venv/bin/activate

Install the package in editable mode:

python -mpip install maturin
maturin develop --group dev

Run this command on initial setup. On subsequent runs, activate the environment with source venv/bin/activate and rebuild Rust code with maturin develop. Python-only changes require no rebuild.

Install pre-commit hooks:

pre-commit install

Run tests

python -mpytest

Benchmarks are disabled by default; enable them with --benchmark-enable.

Citation

If you found this project useful for your research, please cite Malanchev et al., 2021:

@ARTICLE{2021MNRAS.502.5147M,
       author = {{Malanchev}, K.~L. and {Pruzhinskaya}, M.~V. and {Korolev}, V.~S. and {Aleo}, P.~D. and {Kornilov}, M.~V. and {Ishida}, E.~E.~O. and {Krushinsky}, V.~V. and {Mondon}, F. and {Sreejith}, S. and {Volnova}, A.~A. and {Belinski}, A.~A. and {Dodin}, A.~V. and {Tatarnikov}, A.~M. and {Zheltoukhov}, S.~G. and {(The SNAD Team)}},
        title = "{Anomaly detection in the Zwicky Transient Facility DR3}",
      journal = {\mnras},
     keywords = {methods: data analysis, astronomical data bases: miscellaneous, stars: variables: general, Astrophysics - Instrumentation and Methods for Astrophysics, Astrophysics - Solar and Stellar Astrophysics},
         year = 2021,
        month = apr,
       volume = {502},
       number = {4},
        pages = {5147-5175},
          doi = {10.1093/mnras/stab316},
archivePrefix = {arXiv},
       eprint = {2012.01419},
 primaryClass = {astro-ph.IM},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2021MNRAS.502.5147M},
      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}