sequitur
A PyTorch library for creating and training autoencoders on sequential data (time series, videos, etc.) in just two lines of code.
What is sequitur?
Sequitur is a PyTorch library that enables quick creation and training of autoencoders for sequential data. It provides pre-implemented architectures and a simplified training loop to learn compressed representations of sequences like time series, videos, or multivariate data. The library solves the problem of implementing autoencoders from scratch, reducing boilerplate and accelerating experimentation.
Data scientists, machine learning engineers, and researchers working with sequential data who need rapid prototyping of autoencoders for tasks like dimensionality reduction, anomaly detection, or feature extraction.
Developers choose Sequitur for its extreme simplicity—training an autoencoder in two lines of code—while offering flexibility through PyTorch integration. It uniquely combines ease of use with specialized architectures for different sequence types (numeric, vector, image sequences).
Overview
Library of autoencoders for sequential data
Use Cases
Best For
- Learning compressed representations of multivariate time series data
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
