Crate neat

A simple crate that implements the Neuroevolution Augmenting Topologies algorithm using genetic-rs

Feature Roadmap:

• base (single-core) crate
• rayon
• crossover

You can get started by

Re-exports

• pub use runnable::*;
• pub use topology::*;

Modules

• builtin
  Built-in nextgen functions and traits to go with them.
• runnable
  A module containing the main NeuralNetwork struct. This has state/cache and will run the predictions. Make sure to run NeuralNetwork::flush_state between uses of NeuralNetwork::predict.
• topology
  A module containing the NeuralNetworkTopology struct. This is what you want to use in the DNA of your agent, as it is the thing that goes through nextgens and suppors mutation.

Structs

• GeneticSim
  The simulation controller.

Traits

• DivisionReproduction
  Used in dividually-reproducing next_gens
• GenerateRandom
  Helper trait used in the generation of random starting populations
• GenerateRandomCollection
  Blanket trait used on collections that contain objects implementing GenerateRandom
• Prunable
  Used in pruning next_gens
• RandomlyMutable
  Used in all of the builtin next_gens to randomly mutate entities a given amount

Functions

• division_pruning_nextgen
  When making a new generation, it despawns half of the entities and then spawns children from the remaining to reproduce. WIP: const generic for mutation rate, will allow for DivisionReproduction::spawn_child to accept a custom mutation rate. Delayed due to current Rust limitations
• scrambling_nextgen
  When making a new generation, it mutates each entity a certain amount depending on their reward. This nextgen is very situational and should not be your first choice.

Type Aliases

• FitnessFn
  Represents a fitness function. Inputs a reference to the entity and outputs an f32.
• NextgenFn
  Represents a nextgen function. Inputs entities and rewards and produces the next generation