bace
Project description
bace
A deck of Naive Bayes algorithms with sklearn-like API.
Algorithms
Complement Naive Bayes
Negation Naive Bayes
Universal-set Naive Bayes
Selective Naive Bayes
Installation
You can install this module directly from GitHub repo with command:
pip install git+https://github.com/krzjoa/Bayes.git
or as a PyPI package
pip install bayes-variants
Usage
Bayes classifiers API mimics Scikit-Learn API, so usage is very simple.
from bace import ComplementNB
from sklearn.datasets import fetch_20newsgroups
from sklearn.feature_extraction.text import CountVectorizer
vectorizer = CountVectorizer()
categories = ['alt.atheism', 'talk.religion.misc',
'comp.graphics', 'sci.space']
# Train set
newsgroups_train = fetch_20newsgroups(subset='train',
categories=categories, shuffle=True)
X_train = vectorizer.fit_transform(newsgroups_train.data)
y_train = newsgroups_train.target
# Test set
newsgroups_test = fetch_20newsgroups(subset='test',
categories=categories, shuffle=True)
X_test = vectorizer.fit_transform(newsgroups_test.data)
y_test = newsgroups_test.target
# Score
cnb = ComplementNB()
cnb.fit(X_train, y_train).accuracy_score(X_test, y_test)
TODO list
Weighted Complement Naive Bayes
Locally Weighted Naive Bayes
Documentation
The full documentation is at http://bace.rtfd.org.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
bace-1.0.0.tar.gz
(9.4 kB
view hashes)
Built Distribution
bace-1.0.0-py3-none-any.whl
(18.1 kB
view hashes)