Convert EEG signal to pure spectrogram image

eeg_to_spectrogram.py

# eeg_to_spectrogram

# steps

# 1. read eeg with pyedflib
# 2. convert eeg signal to spectrogram using matplotlib
# 3. remove unneccesary staff from matplotlib
# 4. save a pure spectrogram image for further analysis or 
# machine learning image classification task

# reference
# https://dsp.stackexchange.com/questions/25115/python-time-frequency-spectrogram
# https://stackoverflow.com/questions/11837979/removing-white-space-around-a-saved-image-in-matplotlib

import pyedflib
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
from matplotlib.pyplot import specgram

f = pyedflib.EdfReader("data/shhs1-200001.edf")
a = (f.readSignal(2))
print(a.shape)

# use scipy.signal spectrogram to extract raw data
from scipy.signal import spectrogram
raw = spectrogram(a[:7500], fs=125, noverlap=1)[2]

im = specgram(a[:7500], Fs=125, noverlap=1)[3]
plt.gca().set_axis_off()
plt.subplots_adjust(top = 1, bottom = 0, right = 1, left = 0, 
            hspace = 0, wspace = 0)
plt.margins(0,0)
plt.gca().xaxis.set_major_locator(matplotlib.ticker.NullLocator())
plt.gca().yaxis.set_major_locator(matplotlib.ticker.NullLocator())
plt.savefig("image.png", bbox_inches = 'tight',
    pad_inches = 0)
f._close()

Can you only use .edf file conversion? Can you use.mat, .dat and other files conversion