stumpy/stumpy/stomp.py at main · stumpy-dev/stumpy · GitHub

History

192 lines (156 loc) · 5.81 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

# STUMPY

import warnings

import numpy as np

from . import config, core

from .stamp import _mass_PI

def _stomp(T_A, m, T_B=None, ignore_trivial=True):

"""

Compute "Scalable Time series Ordered-search Matrix Profile" (STOMP)

Parameters

----------

T_A : numpy.ndarray

The time series or sequence for which the matrix profile index will

be returned

m : int

Window size

T_B : numpy.ndarray, default None

The time series or sequence that will be used to annotate T_A. For every

subsequence in T_A, its nearest neighbor in T_B will be recorded.

ignore_trivial : bool, default True

`True` if this is a self join and `False` otherwise (i.e., AB-join).

Returns

-------

out : numpy.ndarray

A four column numpy array where the first column is the matrix profile,

the second column is the matrix profile indices. The third and fourth

columns are the left and right matrix profile indices, respectively.

Notes

-----

`DOI: 10.1109/ICDM.2016.0085 \

<https://www.cs.ucr.edu/~eamonn/STOMP_GPU_final_submission_camera_ready.pdf>`__

See Table II

Timeseries, T_A, will be annotated with the distance location

(or index) of all its subsequences in another times series, T_B.

For every subsequence, Q, in T_A, you will get a distance

and index for the closest subsequence in T_A. Thus, the array

returned will have length T_A.shape[0]-m+1. Additionally, the

left and right matrix profiles are also returned.

Note: Unlike in the Table II where T_A.shape is expected to be equal

to T_B.shape, this implementation is generalized so that the shapes of

T_A and T_B can be different. In the case where T_A.shape == T_B.shape,

then our algorithm reduces down to the same algorithm found in Table II.

Additionally, unlike STAMP where the exclusion zone is m/2, the default

exclusion zone for STOMP is m/4 (See Definition 3 and Figure 3).

For self-joins, set `ignore_trivial = True` in order to avoid the

trivial match.

Note that left and right matrix profiles are only available for self-joins.

"""

msg = "stumpy.stomp._stomp is not supported and only provided for reference.\n"

msg += "Please use the Numba JIT-compiled stumpy.stump or stumpy.gpu_stump instead."

warnings.warn(msg)

if T_B is None:

T_B = T_A

core.check_self_join(ignore_trivial)

ignore_trivial = True

T_A, μ_Q, σ_Q, Q_subseq_isconstant = core.preprocess(T_A, m)

T_B, M_T, Σ_T, T_subseq_isconstant = core.preprocess(T_B, m)

if T_A.ndim != 1: # pragma: no cover

raise ValueError(f"T_A is {T_A.ndim}-dimensional and must be 1-dimensional. ")

if T_B.ndim != 1: # pragma: no cover

raise ValueError(f"T_B is {T_B.ndim}-dimensional and must be 1-dimensional. ")

ignore_trivial = core.check_ignore_trivial(T_A, T_B, ignore_trivial)

if ignore_trivial: # self-join

core.check_window_size(

m, max_size=min(T_A.shape[0], T_B.shape[0]), n=T_A.shape[0]

)

else: # AB-join

core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))

n = T_A.shape[0]

l = n - m + 1

excl_zone = int(

np.ceil(m / config.STUMPY_EXCL_ZONE_DENOM)

) # See Definition 3 and Figure 3

out = np.empty((l, 4), dtype=object)

# Handle first subsequence, add exclusionary zone

if np.isinf(μ_Q[0]):

P = np.inf

I = -1

IR = -1

else:

if ignore_trivial:

P, I = _mass_PI(

T_A[:m],

T_B,

M_T,

Σ_T,

excl_zone,

T_subseq_isconstant=T_subseq_isconstant,

Q_subseq_isconstant=Q_subseq_isconstant[[0]],

)

PR, IR = _mass_PI(

T_A[:m],

T_B,

M_T,

Σ_T,

excl_zone,

right=True,

T_subseq_isconstant=T_subseq_isconstant,

Q_subseq_isconstant=Q_subseq_isconstant[[0]],

)

else:

P, I = _mass_PI(

T_A[:m],

T_B,

M_T,

Σ_T,

T_subseq_isconstant=T_subseq_isconstant,

Q_subseq_isconstant=Q_subseq_isconstant[[0]],

)

IR = -1 # No left and right matrix profile available

out[0] = P, I, -1, IR

QT = core.sliding_dot_product(T_A[:m], T_B)

QT_first = core.sliding_dot_product(T_B[:m], T_A)

w = T_B.shape[0] - m + 1

for i in range(1, l):

QT[1:] = (

QT[: w - 1] - T_A[i - 1] * T_B[: w - 1] + T_A[i - 1 + m] * T_B[-(w - 1) :]

)

QT[0] = QT_first[i]

D = core._calculate_squared_distance_profile(

QT,

μ_Q[i],

σ_Q[i],

M_T,

Σ_T,

Q_subseq_isconstant[i],

T_subseq_isconstant,

)

if ignore_trivial:

core.apply_exclusion_zone(D, i, excl_zone, np.inf)

I = np.argmin(D)

P = np.sqrt(D[I])

if P == np.inf:

I = -1

# Get left and right matrix profiles

IL = -1

PL = np.inf

if ignore_trivial and i > 0:

IL = np.argmin(D[:i])

PL = D[IL]

if PL == np.inf:

IL = -1

IR = -1

PR = np.inf

if ignore_trivial and i + 1 < D.shape[0]:

IR = i + 1 + np.argmin(D[i + 1 :])

PR = D[IR]

if PR == np.inf:

IR = -1

out[i] = P, I, IL, IR

core._check_P(out[:, 0])

return out

stumpy/stumpy/stomp.py at main · stumpy-dev/stumpy · GitHub

Uh oh!

Expand file tree

Breadcrumbs

stomp.py

Latest commit

History

Breadcrumbs

stomp.py

File metadata and controls