Dataset Usage

Univariate time series

Detection, evaluation and visualization. Note that the gray line is the ground truth of anomalies.

from streamad.util import StreamGenerator, UnivariateDS, plot
from streamad.model import SpotDetector

ds = UnivariateDS()
stream = StreamGenerator(ds.data)
model = SpotDetector()

scores = []

for x in stream.iter_item():
    score = model.fit_score(x)
    scores.append(score)

data, label, date, features = ds.data, ds.label, ds.date, ds.features
plot(data=data,scores=scores,date=date,features=features,label=label)

Multivariate time series

An example of multivariate time series detection.

from streamad.util import StreamGenerator, MultivariateDS, plot
from streamad.model import xStreamDetector

ds = MultivariateDS()
stream = StreamGenerator(ds.data)
model = xStreamDetector()

scores = []

for x in stream.iter_item():
    score = model.fit_score(x)
    scores.append(score)

data, label, date, features = ds.data, ds.label, ds.date, ds.features
plot(data=data,scores=scores,date=date,features=features,label=label)

Customized time series

import numpy as np
from streamad.util import StreamGenerator, CustomDS, plot
from streamad.model import SpotDetector

# Customized data
n, A, center, phi = 730, 50, 100, 30
T = 2 * np.pi / 100
t = np.arange(n)
ds = A * np.sin(T * t - phi * T) + center
ds[235:255] = 80
label = np.array([0] * n)
label[235:255] = 1

ds = CustomDS(ds, label)  # You can also use a file path here
stream = StreamGenerator(ds.data)
model = SpotDetector()

scores = []

for x in stream.iter_item():
    score = model.fit_score(x)
    scores.append(score)

data, label, date, features = ds.data, ds.label, ds.date, ds.features
plot(data=data, scores=scores, date=date, features=features, label=label)