15 releases (9 breaking)

0.16.1 Apr 3, 2025
0.16.0 Jan 14, 2025
0.15.0 Oct 28, 2024
0.13.2 May 3, 2024
0.8.0 Jul 25, 2023

#1091 in Machine learning

Download history 35/week @ 2024-12-25 80/week @ 2025-01-01 177/week @ 2025-01-08 355/week @ 2025-01-15 192/week @ 2025-01-22 160/week @ 2025-01-29 229/week @ 2025-02-05 196/week @ 2025-02-12 220/week @ 2025-02-19 236/week @ 2025-02-26 88/week @ 2025-03-05 190/week @ 2025-03-12 191/week @ 2025-03-19 100/week @ 2025-03-26 205/week @ 2025-04-02 87/week @ 2025-04-09

606 downloads per month
Used in 3 crates

MIT/Apache

1MB
19K SLoC

burn-import is a crate designed to simplify the process of importing models trained in other machine learning frameworks into the Burn framework. This tool generates a Rust source file that aligns the imported model with Burn's model and converts tensor data into a format compatible with Burn.


Importing Models

The Burn project supports the import of models from various frameworks, emphasizing efficiency and compatibility. Currently, it handles two primary model formats:

  1. ONNX: Facilitates direct import, ensuring the model's performance and structure are maintained.

  2. PyTorch: Enables the loading of PyTorch model weights into Burn’s native model architecture, ensuring seamless integration.

Contribution

Interested in contributing to burn-import? Check out our development guide for more information.

Dependencies

~45–76MB
~1.5M SLoC