patrick-kidger/signatory

Differentiable computations of the signature and logsignature transforms, on both CPU and GPU. (ICLR 2021)

43
/ 100
Emerging

This tool helps machine learning practitioners transform sequential data, like time series, into a more informative representation. It takes your raw stream of data as input and outputs a 'signature' or 'logsignature' that captures order and area information, which can then be used as features for your machine learning models. It's designed for data scientists and ML engineers working with sequential data.

298 stars. No commits in the last 6 months.

Use this if you are working with time series or other sequential data and need to extract features that robustly capture complex, non-linear relationships for machine learning.

Not ideal if you are not working with sequential data, or if your primary interest is frequency analysis rather than order and area relationships.

time-series-analysis feature-engineering sequential-data machine-learning data-transformation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

298

Forks

36

Language

C++

License

Apache-2.0

Last pushed

Jan 11, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/patrick-kidger/signatory"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.