PROPHET

A neural net implementation written in Rust with a focus on cache-efficiency and sequential performance.

Take a look into the documentation to get to know how to use it!

Link to crates.io.

Usage

• pull this project into your workspace
• build and run with cargo

Future Plans (in no specific order)

• convolutional neural network implementation
• implement neural net generically for num::Float types
• improve error handling: less panics, more results
• enhance interfaces with stronger type safety
• add usage examples and benchmarks
• increase test code coverage
• parallel implementation for large layers