Welcome to gplearn-CUDA’s documentation!
gplearn-CUDA implements GPU-accelerated Genetic Programming in Python, with
a scikit-learn inspired and compatible API.
This project is a high-performance extension of the original gplearn library developed by Trevor Stephens. It maintains all the original functionality while introducing massive parallelization via NVIDIA CUDA.
gplearn-CUDA supports regression through the SymbolicRegressor,
binary classification with the SymbolicClassifier, and automated
feature engineering with the SymbolicTransformer. All three share the
same GP evolution engine while keeping the familiar scikit-learn estimator
workflow.
gplearn-CUDA retains the familiar scikit-learn fit/predict API and
works with the existing scikit-learn pipeline
and grid search
modules. You can get started with gplearn-CUDA as simply as:
est = SymbolicRegressor()
est.fit(X_train, y_train)
y_pred = est.predict(X_test)
However, don’t let that stop you from exploring all the ways that the evolution can be tailored to your problem. The package attempts to squeeze a lot of functionality into a scikit-learn-style API. While there are a lot of parameters to tweak, reading the documentation here should make the more relevant ones clear for your problem.
gplearn-CUDA is built on scikit-learn and a fairly recent copy is required
for installation. If you come across any issues in running or installing the
package, please submit a bug report.
Next up, read some more details about what Genetic Programming is, and how it works…
Contents: