turicreate.toolkits.distances.compute_composite_distance

turicreate.toolkits.distances.compute_composite_distance(distance, x, y)

Compute the value of a composite distance function on two dictionaries, typically SFrame rows.

Parameters:

distance : list[list]

A composite distance function. Composite distance functions are a weighted sum of standard distance functions, each of which applies to its own subset of features. Composite distance functions are specified as a list of distance components, each of which is itself a list containing three items:

1. list or tuple of feature names (strings)
2. standard distance name (string)
3. scaling factor (int or float)

x, y : dict

Individual observations, typically rows of an SFrame, in dictionary form. Must include the features specified by distance.

Returns:

out : float

The distance between x and y, as specified by distance.

Examples

>>> sf = turicreate.SFrame({'X1': [0.98, 0.62, 0.11],
...                       'X2': [0.69, 0.58, 0.36],
...                       'species': ['cat', 'dog', 'fossa']})
...
>>> dist_spec = [[('X1', 'X2'), 'euclidean', 2],
...              [('species',), 'levenshtein', 0.4]]
...
>>> d = turicreate.distances.compute_composite_distance(dist_spec, sf[0], sf[1])
>>> print d
1.95286120899