This document lists all the instructions one needs to run this tool.
First of all, you will need Python 3.6+ in order to interpret this program. Download Python from here. In addition, a few libraries are also needed. The following is the list of all libraries needed:
- csv
- datetime
- sys
- warnings
- matplotlib
- pandas
- multiprocessing
- sklearn
- numpy
- time
Most of these libraries are found by default in Python. The rest can be
installed using the pip tool by using the command pip install [library_name]. If pip is not installed, the instructions to install are here.
After having everything installed, you can run the tool by using the following command in terminal/command prompt,
python main.py "path/to/trainingFile.csv" "path/to/testFile.csv" "output.csv". The first two arguments are the paths to training and test files respectively while the last argument is the name (or path) to the output file. All of these files must be in csv format.
After termination, an output file in the csv format will left in the current directory. This file will contain three addtional columns containing predictions from the three models.