def __data_generation(self, list_IDs_temp):
'Generates data containing batch_size samples' # X : (n_samples, *dim, n_channels)
# Initialization
X = np.empty((self.batch_size, *self.dim, self.n_channels), dtype = float)
y = np.empty((self.batch_size), dtype = int)
# Generate data
for i, ID in enumerate(list_IDs_temp):
data = extend_ts(self.h5file[ID]['ecgdata'][:, 0], self.sequence_length)
data = np.reshape(data, (1, len(data)))
if self.augment:
# dropout bursts
data = zero_filter(data, threshold = 2, depth = 10)
# random resampling
data = random_resample(data)
# Generate spectrogram
data_spectrogram = spectrogram(data, nperseg = self.nperseg, noverlap = self.noverlap)[2]
# Normalize
data_transformed = norm_float(data_spectrogram, self.data_mean, self.data_std)
X[i,] = np.expand_dims(data_transformed, axis = 3)
# Assuming that the dataset names are unique (only 1 per label)
y[i] = self.labels[ID]
return X, keras.utils.to_categorical(y, num_classes=self.n_classes)
def __data_generation(self, list_IDs_temp):
'Generates data containing batch_size samples' # X : (n_samples, *dim, n_channels)
# Initialization
X = np.empty((self.batch_size, *self.dim, self.n_channels), dtype=float)
y = np.empty((self.batch_size), dtype=int)
# Generate data
for i, ID in enumerate(list_IDs_temp):
data = extend_ts(self.h5file[ID]['ecgdata'][:, 0],
self.sequence_length)
data = np.reshape(data, (1, len(data)))
if self.augment:
# dropout bursts
data = zero_filter(data, threshold=2, depth=10)
# random resampling
data = random_resample(data)
# Generate spectrogram
data_spectrogram = spectrogram(data,
nperseg=self.nperseg,
noverlap=self.noverlap)[2]
# Normalize spectrogram
#data_transformed = norm_float(data_spectrogram, self.data_mean, self.data_std)
data_norm = (data_spectrogram -
np.mean(data_spectrogram)) / np.std(data_spectrogram)
X[i, ] = np.expand_dims(data_norm, axis=3)
# Assuming that the dataset names are unique (only 1 per label)
y[i] = self.labels[ID]
return X, keras.utils.to_categorical(y, num_classes=self.n_classes)
plt.tight_layout()
plt.show()
fig.savefig('physionet_ECG_spectrogram.png', bbox_inches='tight', dpi=150)
from physionet_processing import zero_filter, random_resample
data_zero = [] # collect the zero burst data here
data_resampled = [] # list for resampled data
runs = 4
for i in range(runs):
data_zero.append(
zero_filter(np.expand_dims(ts_extended, axis=0), threshold=2,
depth=10)[0])
data_resampled.append(
random_resample(np.expand_dims(ts_extended, axis=0))[0])
# Plot the result
fig, axarr = plt.subplots(runs, 2, figsize=(15, 8))
x_lim = [6, 8] # zoom in on x axis
x_ticks = np.arange(x_lim[0], x_lim[1] + 0.5, 0.5) # ticks
x_label = 'Time [s]'
y_label = 'Potential [mV]'
for row in range(runs):
# Plot the dropout burst data
ax = axarr[row, 0]
ax.plot(time, ts_extended, 'k')
ax.plot(time, data_zero[row], 'r')
ax.set(xlim=x_lim, xticks=x_ticks)