mattpetters/CSc4980Project

Machine Learning Project

Machine Learning Project

Requires Python 3

To install dependencies using pip:

pip3 install pandas sklearn numpy

1. Missing Value Estimator

To run directly:

python3 missingValueEstimate.py path/to/input.txt path/to/output.txt

or run:

sh estimateSampleOne.sh

sh estimateSampleTwo.sh

sh estimateSampleThree.sh

To run the provided sample data.

2. Classification

To run directly (training dataset 1 shown here):

python3 classifiers/bayesNaiveClassifier.py src/TrainData1.txt src/TrainLabel1.txt src/TestData1.txt out/TestDataLabels1.txt

or run

sh trainAndTest1.sh

sh trainAndTest2.sh

sh trainAndTest3.sh

sh trainAndTest4.sh

sh trainAndTest5.sh

NOTE: Due to widely varying input, we chose to implement 5 different classifiers which expect different specific formats for training data and test data. Keep this in mind when trying to run your own arbitrary input. The shell scripts (ie; trainAndTest1.sh) show examples of which classifiers to use with which kinds of data