timeseries.Signal.self_similarity

Signal.self_similarity(window, step=1, start=None, stop=None, dist=<function euclidean>, normalizer=None)

Compare sliding windows of this Signal using a distance function.

Parameters:

window (length of segment)

step (steps to move between windows)

start (index to start at)

stop (index to stop at)

dist (distance function to use. Default:`scipy.spatial.distance.euclidean`

normalizer (function to use to renormalize each window. default:None)

Returns:

an iterator of the window comparisons.

(0,0), (0,1), (0,2), … (0, n-1), (1,1), (1,2), … (n-2, n-1)

The return elements are pairs ((index_lo, index_hi), norm), which

can be used to populate a dictionary or array.

Examples

>>> S = Signal([0.0, 0.0, 3.0, 4.0, 6.0, 8.0])
>>> Sss = list(S.self_similarity(window=2, step=2))
>>> for (ij, d) in Sss:
...     print("{} -> {}".format(ij,d))
(0, 0) -> 0.0
(0, 2) -> 5.0
(0, 4) -> 10.0
(2, 2) -> 0.0
(2, 4) -> 5.0
(4, 4) -> 0.0
>>> D = np.array([d for ij,d in Sss])
>>> print(D)
[  0.   5.  10.   0.   5.   0.]
>>> print(ssd.squareform(D)) # scipy.spatial.distance
[[  0.   0.   5.  10.]
 [  0.   0.   0.   5.]
 [  5.   0.   0.   0.]
 [ 10.   5.   0.   0.]]