Neural Magic Community Resources

Neural Magic is on a mission to help developers interested in accelerating machine learning performance using automated model optimization technologies. We also are building an engine for those of you who care about GPU-class performance but like the flexibility of CPUs. We’ve gathered some community resources below to get you started.

---

Install and use our model optimization tools

Our open-sourced model optimization tools, Sparsify and SparseML, can be found on PyPi. More information can be found on our Docs website. To see the code, visit our GitHub repo. Check them out using your models, as well as contribute code or simply look around.

Run on CPUs?

If you are interested in running your model inference on CPUs, you can install our DeepSparse Engine from PyPi. More information can be found on our Docs website. 

Sparse Model Zoo

Want to get started quicker using one of our optimized models? Our model repo, the SparseZoo, is publicly available.

Getting Support

For detailed documentation, visit our docs page. 

For general help or questions, use our GitHub discussions. Everyone is welcome!

Discussions for Sparsify
Discussions for SparseML
Discussions for SparseZoo
Discussions for DeepSparse Engine 

For bugs or feature requests, use its repo issue tracker:

Issue tracker for Sparsify
Issue tracker for SparseML
Issue tracker for SparseZoo
Issue tracker for DeepSparse Engine

You can get the latest news, webinar and event invites, research papers, and other ML performance tidbits by subscribing to the Neural Magic community.

For more general questions about Neural Magic, please email us at [email protected]