ml
Distance Metrics
Euclidean, Manhattan, Chebyshev, Minkowski(p), Cosine, Hamming, Jaccard — per-dimension breakdown, radar/scatter visualization, dominant-dimension highlighting.
Vector A (comma or space separated)
Vector B
p (Minkowski parameter)
Hamming and Jaccard require binary (0/1) vectors — skipped
Distances
Euclidean3
Manhattan5
Chebyshev2
Minkowski (p=3)2.5713
Cosine similarity0.8895
Cosine distance0.1105
Hamming distanceN/A (non-binary)
Jaccard distanceN/A (non-binary)
Per-Dimension Breakdown
| Dim | A | B | |A-B| | (A-B)² |
|---|---|---|---|---|
| 1 | 3 | 1 | 2 | 4 |
| 2 | 4 | 2 | 2 | 4 |
| 3 | 1 | 1 | 0 | 0 |
| 4 | 0 | 1 | 1 | 1 |
Radar Chart (4 dims)
— A— B
Euclidean = √Σ(aᵢ-bᵢ)² · Cosine = A·B / (‖A‖‖B‖) · Minkowski(p) = (Σ|aᵢ-bᵢ|ᵖ)^(1/p)