1, it is said in scikit-learn that 'CNB regularly outperforms MNB (often by a considerable margin) on text classification tasks.'
2, the training part of CNB is highly similar to existing MNB, so it is a easy win to support CNB.
- links to
1, it is said in scikit-learn that 'CNB regularly outperforms MNB (often by a considerable margin) on text classification tasks.'
2, the training part of CNB is highly similar to existing MNB, so it is a easy win to support CNB.