A Numba JIT-compiled version of non-normailzed (i.e., without z-normalization) mSTOMP, a variant of mSTAMP, for parallel computation of the multi-dimensional matrix profile and multi-dimensional matrix profile indices. Note that only self-joins are supported. Parameters ---
(
T,
m,
range_stop,
excl_zone,
T_A_subseq_isfinite,
T_B_subseq_isfinite,
p,
p_norm,
p_norm_first,
k,
range_start=1,
include=None,
discords=False,
)
| 715 | |
| 716 | |
| 717 | def _maamp( |
| 718 | T, |
| 719 | m, |
| 720 | range_stop, |
| 721 | excl_zone, |
| 722 | T_A_subseq_isfinite, |
| 723 | T_B_subseq_isfinite, |
| 724 | p, |
| 725 | p_norm, |
| 726 | p_norm_first, |
| 727 | k, |
| 728 | range_start=1, |
| 729 | include=None, |
| 730 | discords=False, |
| 731 | ): |
| 732 | """ |
| 733 | A Numba JIT-compiled version of non-normailzed (i.e., without z-normalization) |
| 734 | mSTOMP, a variant of mSTAMP, for parallel computation of the multi-dimensional |
| 735 | matrix profile and multi-dimensional matrix profile indices. Note that only |
| 736 | self-joins are supported. |
| 737 | |
| 738 | Parameters |
| 739 | ---------- |
| 740 | T : numpy.ndarray |
| 741 | The time series or sequence for which to compute the multi-dimensional |
| 742 | matrix profile |
| 743 | |
| 744 | m : int |
| 745 | Window size |
| 746 | |
| 747 | range_stop : int |
| 748 | The index value along T for which to stop the matrix profile |
| 749 | calculation. This parameter is here for consistency with the |
| 750 | distributed `mstumped` algorithm. |
| 751 | |
| 752 | excl_zone : int |
| 753 | The half width for the exclusion zone relative to the current |
| 754 | sliding window |
| 755 | |
| 756 | T_A_subseq_isfinite : numpy.ndarray |
| 757 | A boolean array that indicates whether a subsequence in `T_A` contains a |
| 758 | `np.nan`/`np.inf` value (False) |
| 759 | |
| 760 | T_B_subseq_isfinite : numpy.ndarray |
| 761 | A boolean array that indicates whether a subsequence in `T_B` contains a |
| 762 | `np.nan`/`np.inf` value (False) |
| 763 | |
| 764 | p : float |
| 765 | The p-norm to apply for computing the Minkowski distance. Minkowski distance is |
| 766 | typically used with `p` being 1 or 2, which correspond to the Manhattan distance |
| 767 | and the Euclidean distance, respectively. |
| 768 | |
| 769 | p_norm : numpy.ndarray |
| 770 | The input p-norm array between some query sequence,`Q`, and time series, `T` |
| 771 | |
| 772 | p_norm_first : numpy.ndarray |
| 773 | The p-norm between the first query sequence,`Q`, and time series, `T` |
| 774 |
no test coverage detected