FFX 1.3.3

FFX is a technique for symbolic regression. It is:

- Fast - runtime 5-60 seconds, depending on problem size (1GHz cpu)
- Scalable - 1000 input variables, no problem!
- Deterministic - no need to "hope and pray".

Installation

To install from PyPI, simply run:

pip install ffx

Usage

A simple example of how to use FFX is:

import ffx
models = ffx.run(train_X, train_y, test_X, test_y, varnames)
for model in models:
 yhat = model.simulate(X)
 print model

Presently, the FFX package only exposes a single API method, ffx.run().


Product's homepage

Requirements:

· Python
· NumPy
· SciPy
· scikits.learn