CART (Classification And Regression Trees) is an algorithm that can be used to train Decision trees on various classification and regression tasks. In this case, we had to implement a CART algorithm for assignment in Big Data Analytics Programming (BDAP) course at KU Leuven. We had to implement a CART algorithm with support for both numerical and categorical features. The implementation was graded based on correctness, speed, and style. My implementation is trained (on average) on the whole Covertype dataset in 100 seconds (Weka implementation needs 15 seconds). The algorithm is entirely written in C++.
This project is licensed under the MIT License - see the LICENSE.md file for details