ChebyshevDistance<T>
Models & Types · AiDotNet.Clustering.DistanceMetrics
Computes Chebyshev (L∞) distance between vectors.
Inheritance: DistanceMetricBase<T> → ChebyshevDistance<T>
For Beginners
Chebyshev distance is the largest difference between any pair of corresponding elements. It’s like a “worst case” distance.
Example: The distance between points (0, 0) and (3, 4) is 4 (max(|3-0|, |4-0|) = max(3, 4) = 4)
Named after Pafnuty Chebyshev, it’s useful when:
- You care about the maximum deviation in any single feature
- Movement is possible in all directions at once (like a king in chess)
- You need a robust metric that isn’t sensitive to many small differences
How It Works
Chebyshev distance, also called maximum metric or L-infinity distance, is the maximum absolute difference along any dimension. It’s the limit of Minkowski distance as p approaches infinity.
Formula: d(a, b) = max(|a[i] - b[i]|)
Constructors
| Constructor | Summary |
|---|---|
ChebyshevDistance() |
Properties
| Property | Summary |
|---|---|
string Name { get; } |
Methods
| Method | Summary |
|---|---|
T Compute(Vector<T> a, Vector<T> b) |