clustering for a single timeseries

Question

I have a single array numpy array(x) and i want to cluster it in unsupervised way using DBSCAN and hierarchial clustering using scikitlearn. Is the clustering possible for single array data? Additionally i need to plot the clusters and its corresponding representation on the input data.

I tried

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.cluster import DBSCAN
from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler
from scipy import stats
import scipy.cluster.hierarchy as hac
#my data

x = np.linspace(0, 500, 10000)
x = 1.5 * np.sin(x)
#dbscan
clustering = DBSCAN(eps=3).fit(x)
# here i am facing problem
# hierarchial

Does `clustering = DBSCAN(eps=3).fit(x.reshape(-1,1))` solve the immediate error? — rickhg12hs, Dec 17 '22 at 23:54
When I run the code you've shown, there is an error at `clustering = DBSCAN(eps=3).fit(x)`. By changing it to `clustering = DBSCAN(eps=3).fit(x.reshape(-1,1))` the error does not occur. — rickhg12hs, Dec 18 '22 at 04:46
@rickhg12hr the clustered for the 1d timeseries is plotted along a line by using this code line plt.scatter(x,np.zeros(len(x)), c=clusters). How can we plot the clustered in 2d image as depicted here https://scikit-learn.org/stable/auto_examples/cluster/plot_dbscan.html#sphx-glr-auto-examples-cluster-plot-dbscan-py or it cannot be possible for 1d timeseries data. can you please suggest. — pro, Dec 19 '22 at 15:09
See the bottom of my answer below. You can do it, but I don't see the value in clustering when there are no features other than time and amplitude. But perhaps your application does see some value. — rickhg12hs, Dec 19 '22 at 15:12
caanot we do it for plt.scatter(x,np.zeros(len(x)), c=clusters) — pro, Dec 19 '22 at 15:17
If you want to plot on a 2D-plane, one dimension is amplitude ... what is the other dimension with your data? — rickhg12hs, Dec 19 '22 at 15:31
actually, i have time on x-axis and amplitude on the y-axis that plot a waveform/timeseries. Actually in my question i didnot mentioned the time which is len(amplitude) — pro, Dec 19 '22 at 15:34
See the last example in my answer. It uses time (X-axis) and signal amplitude (Y-axis). — rickhg12hs, Dec 19 '22 at 15:36
however in your last answer scattered plot should come like this https://scikit-learn.org/stable/auto_examples/cluster/plot_dbscan.html#sphx-glr-auto-examples-cluster-plot-dbscan-py but it gives waveform as clustered — pro, Dec 19 '22 at 15:38
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/250530/discussion-between-rickhg12hs-and-pro). — rickhg12hs, Dec 19 '22 at 15:39

rickhg12hs · Accepted Answer · 2022-12-18T11:31:48.057

Yes, DBSCAN can cluster "1-D" arrays. See time series below, although I don't know the significance of clustering just the waveform.

For example,

import numpy as np

rng =np.random.default_rng(42)
x=rng.normal(loc=[-10,0,0,0,10], size=(200,5)).reshape(-1,1)
rng.shuffle(x)
print(x[:10])
# [[-10.54349551]
#  [ -0.32626201]
#  [  0.22359555]
#  [ -0.05841124]
#  [ -0.11761086]
#  [ -1.0824272 ]
#  [  0.43476607]
#  [ 11.40382139]
#  [  0.70166365]
#  [  9.79889535]]

from sklearn.cluster import DBSCAN

dbs=DBSCAN()
clusters = dbs.fit_predict(x)

import matplotlib.pyplot as plt
plt.scatter(x,np.zeros(len(x)), c=clusters)

You can use AgglomerativeClustering for hierarchical clustering.

Here's an example using the data from above.

from sklearn.cluster import AgglomerativeClustering

aggC = AgglomerativeClustering(n_clusters=None, distance_threshold=1.0, linkage="single")
clusters = aggC.fit_predict(x)

plt.scatter(x,np.zeros(len(x)), c=clusters)

Time Series / Waveform (no other features)

You can do it, but with no features other than time and signal amplitude, I don't know if this has any meaning.

import numpy as np
from scipy import signal

y = np.hstack((np.zeros(100), signal.square(2*np.pi*np.linspace(0,2,200, endpoint=False)), np.zeros(100), signal.sawtooth(2*np.pi*np.linspace(0,2,200, endpoint=False)+np.pi/2,width=0.5), np.zeros(100), np.sin(2*np.pi*np.linspace(0,2,200,endpoint=False)), np.zeros(100)))

import datetime

start = datetime.datetime.fromisoformat("2022-12-01T12:00:00.000000")

times = np.array([(start+datetime.timedelta(microseconds=_)).timestamp() for _ in range(1000)])

my_sig = np.hstack((times.reshape(-1,1),y.reshape(-1,1)))

print(my_sig[:5,:])
# [[1.6698924e+09 0.0000000e+00]
#  [1.6698924e+09 0.0000000e+00]
#  [1.6698924e+09 0.0000000e+00]
#  [1.6698924e+09 0.0000000e+00]
#  [1.6698924e+09 0.0000000e+00]]

from sklearn.cluster import AgglomerativeClustering

aggC = AgglomerativeClustering(n_clusters=None, distance_threshold=4.0)

clusters = aggC.fit_predict(my_sig)

import matplotlib.pyplot as plt
plt.scatter(my_sig[:,0], my_sig[:,1], c=clusters)

@pro I don't know if it means anything, but I added a times series clustering example. — rickhg12hs, Dec 18 '22 at 11:32

clustering for a single timeseries

1 Answers1