The Naive Bayes classifier implementation with support of Gaussian NB, multinomial NB and Bernoulli NB.
Task from Machine Learning for Greenhorns – Winter 2020/21. Part of this code was provided by Milan Straka.
The implementation is in the file naive_bayes.py. It's functionality is shown on some aritficial digits data from sklearn.datasets.
The example of the invocation of the program are:
python naive_bayes.py --classes=3 --naive_bayes_type=bernoulli
python naive_bayes.py --classes=3 --naive_bayes_type=multinomial
python naive_bayes.py --classes=3 --naive_bayes_type=gaussian
python naive_bayes.py --classes=10 --naive_bayes_type=gaussian --alpha=10 --seed=41