Opt CPD algorithm (Mahalanobis metric)

Reported on 6 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Time Series6 results

Change Point DetectiononSKAB
NAB (LowFN)
23.37
best: 31.59 (LSTMCaps)
Change Point DetectiononSKAB
NAB (lowFP)
19.9
best: 21.69 (BinSeg CPD algorithm (Mahalanobis metric))
Change Point DetectiononSKAB
NAB (standard)
22.37
best: 27.77 (LSTMCaps)
Change Point DetectiononTEP
NAB (LowFN)
37.29
best: 42.16 (OptEnsemble CPDE algorithm (Min+MinMax/Rank))
Change Point DetectiononTEP
NAB (lowFP)
35.82
best: 41 (OptEnsemble CPDE algorithm (Min+MinMax/Rank))
Change Point DetectiononTEP
NAB (standard)
36.88
best: 41.81 (OptEnsemble CPDE algorithm (Min+MinMax/Rank))