timeseriesAI/tsai
Time series Timeseries Deep Learning Machine Learning Python Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai
Provides sklearn-compatible preprocessing pipelines, walk-forward cross-validation for time series, and specialized architectures like PatchTST and attention-based RNNs optimized for sequential data. Leverages fastai's training loop abstractions and PyTorch 2.0 support while offering 128+ pre-downloaded benchmark datasets across classification, forecasting, and regression tasks. Supports univariate/multivariate sequences with reduced memory footprint through modular soft dependencies.
6,010 stars and 14,647 monthly downloads. Used by 1 other package. Available on PyPI.
Stars
6,010
Forks
717
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Mar 08, 2026
Monthly downloads
14,647
Commits (30d)
0
Dependencies
8
Reverse dependents
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/timeseriesAI/tsai"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Related frameworks
aeon-toolkit/aeon
A toolkit for time series machine learning and deep learning
sktime/sktime
A unified framework for machine learning with time series
skforecast/skforecast
Time series forecasting with machine learning models
Nixtla/neuralforecast
Scalable and user friendly neural :brain: forecasting algorithms.
Nixtla/statsforecast
Lightning ⚡️ fast forecasting with statistical and econometric models.