ml
PCA Variance
Eigenvalue input or raw data matrix — QR algorithm covariance decomposition, scree plot, cumulative explained variance, components to reach threshold.
Input Mode
Data Matrix (rows = observations, cols = features, max 10×10)
Variance Threshold (%)
PCA Summary
Original dimensions2
Components for 95% variance1
Compression ratio2×
Component Breakdown
| PC | Eigenvalue | Explained % | Cumulative % |
|---|---|---|---|
| PC1 | 1.284 | 96.3% | 96.32% |
| PC2 | 0.04908 | 3.68% | 100% |
Scree Plot
▋ explained variance▋ past threshold—— cumulative--- threshold (95%)
C = XᵀX/(n-1) · QR iteration (20 steps) · explained(i) = λᵢ / Σλ