README.md

Run your Machine Learning and Deep Learning workloads at any scale.

bears delivers lightning-fast data-processing, whether you need to run a single row or 100M+ rows.

bears works with your existing data tools while optimizing data-layouts between Pandas, Dask, Torch, Python dicts for maximum performance. Drop-in compatibility with the Pandas API means you can use bears today with zero code changes.

Help build a bears!

See CONTRIBUTING for more information.

License

This project is licensed under the Apache-2.0 License.