toinsson/pysdtw

Torch implementation of Soft-DTW, supports CUDA.

49
/ 100
Emerging

This project offers a highly optimized tool for calculating the similarity between different time series data. It takes in various time series, even those with different lengths, and outputs a 'distance' score indicating how alike or unalike they are. This is particularly useful for machine learning engineers, data scientists, and researchers working with sequential data who need an efficient way to compare time-series patterns.

Used by 1 other package. Available on PyPI.

Use this if you are a machine learning engineer or data scientist working with time series data and need to calculate differentiable distances between sequences efficiently, especially on GPU hardware.

Not ideal if you are an end-user needing to compare time series without writing Python code or if you don't require gradient computation for machine learning models.

time-series-analysis machine-learning-engineering data-science pattern-recognition sequence-comparison
Maintenance 10 / 25
Adoption 9 / 25
Maturity 25 / 25
Community 5 / 25

How are scores calculated?

Stars

49

Forks

2

Language

Python

License

MIT

Last pushed

Jan 25, 2026

Commits (30d)

0

Dependencies

2

Reverse dependents

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/toinsson/pysdtw"

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