More explanation on how the variables work

[casaltenburg]

Hi there,

Are you guys still active? I did something similair to your example (see code below).
But my result(image below) is not what I want. I only want to detect the big change.
I think I have to change the variables, but I have no idea why and how.
Can you guys help me and explain it in an easy way?

bc = bocd.BayesianOnlineChangePointDetection(bocd.ConstantHazard(250), bocd.StudentT(mu=0, kappa=1, alpha=1, beta=1))

rt_mle = np.empty(raw_signal.shape)
for i, d in enumerate(raw_signal):
    bc.update(d)
    rt_mle[i] = bc.rt

plt.figure(figsize=(20, 10))
plt.plot(raw_signal, alpha=0.5, label="observation")
index_changes = np.where(np.diff(rt_mle)<0)[0]
plt.scatter(index_changes, raw_signal[index_changes], c='green', label="change point")