#machinelearning #inference

cervo-asset

A unified asset type for Cervo using ONNX or NNEF models

4 releases (breaking)

Uses new Rust 2021

0.4.0 Nov 23, 2022
0.3.0 Oct 4, 2022
0.2.0 Jul 11, 2022
0.1.0 Jun 2, 2022

#252 in Science

Download history 64/week @ 2022-08-18 72/week @ 2022-08-25 64/week @ 2022-09-01 100/week @ 2022-09-08 62/week @ 2022-09-15 75/week @ 2022-09-22 111/week @ 2022-09-29 52/week @ 2022-10-06 90/week @ 2022-10-13 68/week @ 2022-10-20 85/week @ 2022-10-27 43/week @ 2022-11-03 21/week @ 2022-11-10 52/week @ 2022-11-17 54/week @ 2022-11-24 95/week @ 2022-12-01

229 downloads per month
Used in 2 crates

MIT/Apache

80KB
1.5K SLoC

🧠 cervo

Wrapper around tract used for ML workloads in our games.

Embark Embark

Build status

Cervo is intended to be a thin wrapper around tract, at a slightly higher abstraction level and with common utilities we need. While not a goal, our current use-cases has led to a design centered around dynamic batching and dictionary inputs for reinforcement-learning based agents.

As of currently, Cervo offers a set of inferers, noise generators (for continuous-action/parametrized policies), and a unified asset format.

Contribution

Contributor Covenant

We welcome community contributions to this project.

Please read our Contributor Guide for more information on how to get started. Please also read our Contributor Terms before you make any contributions.

Any contribution intentionally submitted for inclusion in an Embark Studios project, shall comply with the Rust standard licensing model (MIT OR Apache 2.0) and therefore be dual licensed as described below, without any additional terms or conditions:

License

This contribution is dual licensed under EITHER OF

at your option.

For clarity, "your" refers to Embark or any other licensee/user of the contribution.

Dependencies

~9–15MB
~306K SLoC