Usage

To use OADDS in a project:

from outlierdenstream import Sample, OADDS

Initialize OADDS object:

oadds = OADDS(lamb=0.125, epsilon='dynamic', beta=0.04, mu='auto', startingBuffer=bufferDf)
oadds.runInitialization()

Fit each sample of the dataset with:

for row in dataset:
    sample = Sample(row, timestamp)
    result = oadds.runOnNewSample(sample)

The algorithm returns True (outlier) if it is not able to merge the new sample to an existing core-micro-cluster or merges the sample to an existing outlier-micro-cluster. Returns False (normal) otherwise.