Backup from original repository https://csil-git1.cs.surrey.sfu.ca/rkm3/mlclass-1777-randomforest/tree/master Contributors: Richard Mar, Jae Lee, and Robin White

FILES

TemplateDataSetPreprocessor.py - contains a template to follow for additional preprocessors to work with our random forest implementation.
HockeyDataSetPreprocessor.py - contains logic to take the hockey dataset and extract it into training and test sets.
RandomForest.py - contains logic for orchestrating learning and predicting with multiple trees.
README.md - this read me file.
runner.py - runs the random forests program.
Tree.py - contains logic for a tree in the random forest.

System Requirements

must be run using Python 3.x
numpy v 1.12.1
pandas v 0.21.0
sklearn v 0.19.1
matplotlib v 2.1.0

How to Run

python runner.py -d <inputfile> -c <targetColumnName> --use_hockey_preprocessor

or if you want to use a custom preprocessor:

python runner.py -d <inputfile> -c <targetColumnName> --use_custom_preprocessor <preprocessorFileName>

Please use the -h/--help option for additional details.

Output

Processing the data using Hockey Dataset Preprocessor...
I am tree number: 0
I am tree number: 1
I am tree number: 2
I am tree number: 3
Predicting on the given dataset!
Predicting on the given dataset!

Train Percent Correct: 100.0
Test Percent Correct: 62.30366492146597


Time for train: 1min 4sec
Time for prediction: 0min 0sec

For Developers

Please ensure that the pre-commit file in this repository is added as a Git hook for this project.

It will ensure that the runner.py script is run before making a commit, which will prevent people from breaking the this project.

To add the pre-commit hook, please do the following:

Ensure that you are in the mlclass-1777-randomforest directory.
Run: cp pre-commit .git/hooks -i

And you're done!

Compiling the report