The rogue-net crate provides a pure Rust implementation of the RogueNet neural network. It can be used to load agents created with the Entity Neural Network Trainer and use them inside Rust applications.
```rust use std::collections::HashMap; use ndarray::prelude::*; use rogue_net::RogueNet;
let roguenet = RogueNet::load("checkpoint-dir"); let mut entities = HashMap::new(); entities.insert("Head".tostring(), array![[3.0, 4.0]]); entities.insert("SnakeSegment".tostring(), array![[3.0, 4.0], [4.0, 4.0]]); entities.insert("Food".tostring(), array![[3.0, 5.0], [8.0, 4.0]]); let (actionprobs, actions) = roguenet.forward(&entities); ```