Clojure curve fitting using genetic algorithms
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
public
src/myapp
test/myapp
.gitignore
README.md
example1.png
example2.png
genetic.cpp
project.clj

README.md

Genetic algorithm

Curve fitting is a well researched problem. There are few approaches that use genetic algorithms for curve fitting. We try a new approach to curve fitting where we don't focus on performance but rather try to create "real" equations. This differs from the traditional approach, where parameters of a harmonic function are varied to produce a specific curve.

The visualization is written in D3 and javascript; however, the server and genetic algorithm are written entirely in clojure.

Examples

Generated by code

For goal graph - blue is randomly generated, red is our match. An example graph

Generated by code

For goal graph - blue is randomly generated, red is our match. An example graph

Setup and run

lein run -m myapp.core

Then navigate your browser to localhost:5000 to see some graphs