MCPcopy
hub / github.com/stumpy-dev/stumpy / _stump

Function _stump

stumpy/stump.py:252–506  ·  view source on GitHub ↗

A Numba JIT-compiled version of STOMPopt with Pearson correlations for parallel computation of the (top-k) matrix profile, the (top-k) matrix profile indices, the top-1 left matrix profile and its matrix profile index, and the top-1 right matrix profile and its matrix profile index.

(
    T_A,
    T_B,
    m,
    μ_Q,
    M_T,
    σ_Q_inverse,
    Σ_T_inverse,
    μ_Q_m_1,
    M_T_m_1,
    T_A_subseq_isfinite,
    T_B_subseq_isfinite,
    T_A_subseq_isconstant,
    T_B_subseq_isconstant,
    diags,
    ignore_trivial,
    k,
)

Source from the content-addressed store, hash-verified

250 fastmath=config.STUMPY_FASTMATH_FLAGS,
251)
252def _stump(
253 T_A,
254 T_B,
255 m,
256 μ_Q,
257 M_T,
258 σ_Q_inverse,
259 Σ_T_inverse,
260 μ_Q_m_1,
261 M_T_m_1,
262 T_A_subseq_isfinite,
263 T_B_subseq_isfinite,
264 T_A_subseq_isconstant,
265 T_B_subseq_isconstant,
266 diags,
267 ignore_trivial,
268 k,
269):
270 """
271 A Numba JIT-compiled version of STOMPopt with Pearson correlations for parallel
272 computation of the (top-k) matrix profile, the (top-k) matrix profile indices,
273 the top-1 left matrix profile and its matrix profile index, and the top-1 right
274 matrix profile and its matrix profile index.
275
276 Parameters
277 ----------
278 T_A : numpy.ndarray
279 The time series or sequence for which to compute the matrix profile
280
281 T_B : numpy.ndarray
282 The time series or sequence that will be used to annotate `T_A`. For every
283 subsequence in `T_A`, its nearest neighbor in `T_B` will be recorded.
284
285 m : int
286 Window size
287
288 μ_Q : numpy.ndarray
289 Mean of the query sequence, `Q`, relative to the current sliding window
290
291 M_T : numpy.ndarray
292 Sliding mean of time series, `T`
293
294 σ_Q_inverse : numpy.ndarray
295 Inverse standard deviation of the query sequence, `Q`, relative to the current
296 sliding window
297
298 Σ_T_inverse : numpy.ndarray
299 Inverse sliding standard deviation of time series, `T`
300
301 μ_Q_m_1 : numpy.ndarray
302 Mean of the query sequence, `Q`, relative to the current sliding window and
303 using a window size of `m-1`
304
305 M_T_m_1 : numpy.ndarray
306 Sliding mean of time series, `T`, using a window size of `m-1`
307
308 T_A_subseq_isfinite : numpy.ndarray
309 A boolean array that indicates whether a subsequence in `T_A` contains a

Callers 2

stumpFunction · 0.85
updateMethod · 0.85

Calls 1

_compute_diagonalFunction · 0.70

Tested by

no test coverage detected