API Reference
Modules
A module for storing and (de)serializing ANNs (artificial neural networks)
A module for storing connections between nodes in an ANN
A module for simulating ANNs
This module provides the means for backpropogation
Code for evolving neural networks to reproduce the behavior of the XOR logic gates (only with 0’s repalced by -1’s)
GradAprox uses iterative gradient approximation and descent/ascent to optimize a set of non-discrete numeric parameters using only a fitness/error function. The function takes the parameter map as its argument and returns a value, which in turn is either minimized or maximized
GradAprox.NeuralTrainer uses iterative gradient approximation and descent/ascent to optimize the weights of a neural network using only a fitness/error function. The function takes the ann as its argument and returns a value, which in turn is either minimized or maximized
Evaluates a nerual network.
Usage: mix eval {input1} [input2] [input3] [...]
for as many inputs as are needed.
Once the command is issued, a prompt appears asking for the JSON for the neural network. Paste it in as one line, then hit enter
Usage: mix fishsim [printEvery] [minutes] [file_to_record]
Runs a fishhy evolution process. Runs for the given number of minutes (default 60), and will print out ascii art, or record to the given file every printEvery
generations, default is 1. Ascii art is used if file is not specified
Spawns a Density.Shark
Spawns slower Fish, and a faster Std.Shark
Spawns a Std.Shark and a Density.Shark
mix xor.multi
runs multiple xor evolutions simultaneously, printing out the average number of generations taken
mix xor.single
will run a single xor evolution, printing out information about the process, and the best network
mix xor.single.profile
will run a single xor evolution, printing out profiling information
Neuro-Evolution of Augmenting Topologies (NEAT) is an algorithm for developing Artificial Neural Networks (ANNs) through the process of evolution
Below are the options that can be configured the Neat evolution process