dbnn 4.7.2

Difference Boosting Neural Network Implementation

DBNN (Difference Boosting Neural Network)

A GPU-optimized implementation of Difference Boosting Neural Networks for classification tasks.

Usage

1. Create a configuration file <dataset_name>.conf in your working directory:

{ "file_path": "your_data.csv", "target_column": "your_data_target_name", "separator": ",", "has_header": true, "likelihood_config": { "feature_group_size": 2, "max_combinations": 1000 } }

Installation