spectrogram for one x-axis accelerometer signal

Asked Dec 06 '17 at 11:38

Active Dec 07 '17 at 11:58

Viewed 326 times

trying to convert accelerometer signal into spectrogram instead of raw data i try the following code

import matplotlib.pyplot as plt
from scipy import signal
import numpy as np 
sample_rate, samples = np.loadtxt("trainingdataset.txt", delimiter=",", dtype=np.int32)

frequencies, times, spectogram = signal.spectrogram(samples, sample_rate)

plt.imshow(spectogram)
plt.pcolormesh(times, frequencies, spectogram)
plt.ylabel('Frequency [Hz]')
plt.xlabel('Time [sec]')
plt.show()

but it gets me the following error

Traceback (most recent call last):
  File "C:/Users/hadeer.elziaat/Desktop/r.py", line 5, in <module>
    sample_rate, samples = np.loadtxt("trainingdataset.txt", delimiter=",", dtype=np.int32)
ValueError: too many values to unpack (expected 2)

the dataset text file only contains the accelerometer x-axis with 2947 rows and one column

this is my dataset

edited Dec 07 '17 at 11:58

asked Dec 06 '17 at 11:38

Hadeer El-Zayat

show your .txt- – eyllanesc Dec 06 '17 at 23:42
done .. i have add a sample of my dataset – Hadeer El-Zayat Dec 07 '17 at 11:56
If you want to do a spectrogram you need the time intervals in which the sample was taken, you only provide the value of the sample with which it is impossible. – eyllanesc Dec 07 '17 at 13:02
is the time interval will be added in the dataset or in the code? – Hadeer El-Zayat Dec 08 '17 at 09:23

spectrogram for one x-axis accelerometer signal

0 Answers0