Ruled based feature engineering for regression
Project description
symfeat is a rule based feature engineering library to be used as a preprocessor for regression tasks.
It is based on:
Mcconaghy, T. (2011). FFX: Fast, Scalable, Deterministic Symbolic Regression Technology. Genetic Programming Theory and Practice IX, 235-260. DOI: 10.1007/978-1-4614-1770-5_13
Features
Builds a features based on all valid rule specified combinations
Discards non-finite transformations
Optional: remove equivalent expressions
Installation
pip install symfeat
Usage
import numpy as np
import symfeat as sf
operators = {"sin": np.sin}
exponents = [1, 2, -1, -2]
x = np.random.normal(size=10).reshape(-1, 1)
sym = sf.SymbolicFeatures(exponents=exponents, operators=operators)
features = sym.fit_transform(x)
names = sym.names
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
symfeat-0.3.0.tar.gz
(3.2 kB
view hashes)