A Numba JIT-compiled version of mSTOMP for parallel computation of the multi-dimensional distance profile Parameters ---------- d : int The total number of dimensions in `T` k : int The total number of sliding windows to iterate over idx : int
(
d,
k,
idx,
D,
T,
m,
excl_zone,
M_T,
Σ_T,
QT_even,
QT_odd,
QT_first,
μ_Q,
σ_Q,
Q_subseq_isconstant,
T_subseq_isconstant,
)
| 814 | fastmath=config.STUMPY_FASTMATH_FLAGS, |
| 815 | ) |
| 816 | def _compute_multi_D( |
| 817 | d, |
| 818 | k, |
| 819 | idx, |
| 820 | D, |
| 821 | T, |
| 822 | m, |
| 823 | excl_zone, |
| 824 | M_T, |
| 825 | Σ_T, |
| 826 | QT_even, |
| 827 | QT_odd, |
| 828 | QT_first, |
| 829 | μ_Q, |
| 830 | σ_Q, |
| 831 | Q_subseq_isconstant, |
| 832 | T_subseq_isconstant, |
| 833 | ): |
| 834 | """ |
| 835 | A Numba JIT-compiled version of mSTOMP for parallel computation of the |
| 836 | multi-dimensional distance profile |
| 837 | |
| 838 | Parameters |
| 839 | ---------- |
| 840 | d : int |
| 841 | The total number of dimensions in `T` |
| 842 | |
| 843 | k : int |
| 844 | The total number of sliding windows to iterate over |
| 845 | |
| 846 | idx : int |
| 847 | The subsequence index for the i-th time series, `T[i]` |
| 848 | |
| 849 | D : numpy.ndarray |
| 850 | The output distance profile |
| 851 | |
| 852 | T : numpy.ndarray |
| 853 | The time series or sequence for which to compute the matrix profile |
| 854 | |
| 855 | m : int |
| 856 | Window size |
| 857 | |
| 858 | excl_zone : int |
| 859 | The half width for the exclusion zone relative to the current |
| 860 | sliding window |
| 861 | |
| 862 | M_T : numpy.ndarray |
| 863 | Sliding mean of time series, `T` |
| 864 | |
| 865 | Σ_T : numpy.ndarray |
| 866 | Sliding standard deviation of time series, `T` |
| 867 | |
| 868 | QT_even : numpy.ndarray |
| 869 | Dot product between some query sequence,`Q`, and time series, `T` |
| 870 | |
| 871 | QT_odd : numpy.ndarray |
| 872 | Dot product between some query sequence,`Q`, and time series, `T` |
| 873 |