Uncorrelated Linear Discriminant Analysis (ULDA), modified based on Ye, J., & Yu, B. (2005).
- Provide a more robust LDA module compared to one in
sklearn
, especially when handling perfect separation in high-dimensional data. - Faster performance.
pip install ulda
import numpy as np
from ulda import ULDA
X = np.array([[0, 0], [0,1], [1, 1], [1, 2], [2, 2], [2, 3]])
y = np.array(['A', 'A', 'B', 'B', 'C', 'C'])
lda = ULDA()
lda.fit(X, y)
print(lda.predict([[1, 3]]))