mnemosine.vec._distance

Kind: function

Signature

_distance(a, b, metric: str) -> float

Documentation

Compute a distance between two same-length vectors.

Supports l2 (Euclidean), dot (negated dot product) and cosine (1 - cosine similarity; returns 1.0 if either vector has zero magnitude).

Args:

• a (sequence of float): First vector.
• b (sequence of float): Second vector.
• metric (str): "l2", "dot" or "cosine".

Returns:

• float: The distance. Lower means closer.

Raises:

• VectorError: If the vectors have different lengths.

Example:

>>> from mnemosine.vec import _distance
>>> round(_distance([1, 0], [1, 0], "cosine"), 3)
0.0