CTRF 1.0.9

Please add the appropriate description

Overview

CTRF is accepted paper in 27th International Conference on Pattern Recognition (ICPR), Kolkata, India. The mission of CTRF is to help users easily and efficiently transform Random Forests into tractable circuits. CTRF exploits bayesian learning and boolean algebra to achieve high efficiency. Key features of CTRF are as follows.

• Works for any number of variables.
• Reduction in variables and logical operations.
• Supported Operating System(s): Linux and Windows.
• Support classification, tested on UCIML datasets.

Why optimize Random Forests: A survey conducted by Kaggle in 2017 shows that 50%, 46% and 24% of the data mining and machine learning practitioners are users of Decision Trees, Random Forests and GBMs, respectively.

Getting Started

Requirements

1. matplotlib
2. numpy
3. pandas
4. scikit-learn
5. scipy
6. pyeda + related his VS distribution file (https://www.microsoft.com/en-us/download/details.aspx?id=30679)
7. gmm_mml

Quick Install

*  pip install CTRF
*  pip install  github link

• Currently only support python3
• After you have installed CTRF, you can import and use the classifier by:

from CTRF import train ,predictf
clf = train()
clf.predict()

Build from source

git clone GitHub link
cd CTRF
#under the directory of CTRF we can train and predict the info

You will see info about the accuracies of RF, RF-V,BDS, OBDS after successful running predictf()

Related papers

• Choi A, Shih A, Goyanka A, Darwiche A. On symbolically encoding the behavior of random forests. arXiv preprint arXiv:2007.01493. 2020 Jul 3. pdf

• Audemard, Gilles, Frédéric Koriche, and Pierre Marquis. On tractable XAI queries based on compiled representations. Proceedings of the International Conference on Principles of Knowledge Representation and Reasoning. Vol. 17. No. 1. 2020. pdf

##Contact

Principal Investigator

Dr Debdoot Sheet
Department of Electrical Engineering,
Indian Institute of Technology Kharagpur
email: [e-mail]

Contributors

Maddimsetti Srinivas
Department of Electrical Engineering,
Indian Institute of Technology Kharagpur
email: [e-mail]

Sattenapalli Sai hemanth
Scarlet Moose Entertainment
email: [e-mail]

RaviTeja Garlapati
Scarlet Moose Entertainment
email: [e-mail]