def transform_to_2D(self, method, x_train):
if method == 'gasf':
gasf = GASF(image_size=x_train.shape[1] // 2,
overlapping=False,
scale=-1)
x_tr = gasf.fit_transform(x_train)
print('applying GASF')
elif method == 'mtf':
mtf = MTF(image_size=x_train.shape[1],
n_bins=4,
quantiles='empirical',
overlapping=False)
x_tr = mtf.fit_transform(x_train)
print('applying MTF')
elif method == 'rp':
rp = RecurrencePlots(dimension=1,
epsilon='percentage_points',
percentage=10)
x_tr = rp.fit_transform(x_train)
print('applying RP')
else:
print('wrong method')
x_tr = []
return x_tr
def recurrence_plot(dataset, n_channels, seq_length, img_size=32, downsample=False):
"""
function to transform the given dataset into recurrence plot images
:param dataset: the raw data read from .csv or .npy file
:type dataset: ndarray
:param n_channels: number of channels
:type n_channels: int
:param seq_length: length of one data stream of a single sensor
:type seq_length: int
:param img_size: size of the recurrence plot image (if downsampling is enabled)
:type img_size: int
:param downsample: whether downsampling is applied or not
:type downsample: bool
:return: the flattened images, tupel holding the image size
"""
rp = RecurrencePlots()
if downsample:
size_flat = img_size * img_size
else:
size_flat = seq_length * seq_length
pics = np.zeros([dataset.shape[0], n_channels * size_flat]) # initialize array
for ch in range(n_channels): # loop over all channels
pics_ch = rp.fit_transform(dataset[:, ch * seq_length:(ch + 1) * seq_length]) # calculate recurrence plot
pics_ch = np.moveaxis(pics_ch, 0, 2) # [samples, heigth, width] --> [heigth,width,samples]
if downsample:
pics_ch = transform.resize(pics_ch, [img_size, img_size], mode='constant', anti_aliasing=0.15,
preserve_range=True) # rescale (downsample)
else:
img_size = seq_length
pics_ch = np.reshape(pics_ch, [size_flat, -1]) # [signal, samples]
pics_ch = np.moveaxis(pics_ch, 0, 1) # swap axes --> [samples, signal]
pics[:, ch * size_flat:(ch + 1) * size_flat] = pics_ch
return pics, [img_size, img_size]
def transform_ECG(x, method):
# transform ECG sequence(s) to binary image(s)
if method == 'gasf':
gasf = GASF(image_size=x.shape[1] // 2, overlapping=False, scale=-1)
x = gasf.fit_transform(x)
# print('applying GASF')
elif method == 'mtf':
mtf = MTF(image_size=x.shape[1], n_bins=4, quantiles='empirical', overlapping=False)
x = mtf.fit_transform(x)
# print('applying MTF')
elif method == 'rp':
rp = RecurrencePlots(dimension=1, epsilon='percentage_points', percentage=10)
x = rp.fit_transform(x)
# print('applying RP')
else:
raise ValueError("Invalid method: " + str(method))
return x
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from pyts.image import GASF, GADF, RecurrencePlots, MTF
sig = pd.read_csv('sample_1.csv').iloc[0:10000, 4:7]
n_samples, n_features = 100, 1
rng = np.random.RandomState(41)
X = rng.randn(n_samples, n_features)
# Recurrence plot transformation
rp = RecurrencePlots(dimension=1,
epsilon='percentage_points',
percentage=30)
X_rp = rp.fit_transform(X)
plt.figure(figsize=(8, 8))
plt.imshow(X_rp[0], cmap='binary', origin='lower')
plt.show()
# MTF transformation
image_size = 1
mtf = MTF(image_size)
X_mtf = mtf.fit_transform(sig)
y, sr = librosa.load(path)
time = np.arange(0,
len(y) / sr, 1 / sr)
total_data_train.append(y)
total_data_time_train.append(time)
if len(y) > max_size:
max_size = len(y)
if path.find('undir') < 0:
data_out_train.append('Directed')
else:
data_out_train.append('Undirected')
X_train = np.empty([len(total_data_train), max_size])
rp = RecurrencePlots(dimension=1, epsilon='percentage_points', percentage=30)
folder = 'F:\data_for_avishek\LSTMGeneric2\RP_Images2\Train\\'
#folderD = 'F:\data_for_avishek\LSTMGeneric2\RP_Images2\Train\Directed\\'
#folderU = r'F:\data_for_avishek\LSTMGeneric2\RP_Images2\Train\Undirected\\'
#num = 78;
#fig = plt.figure(frameon=False)
#ax = plt.gca()
#im = ax.imshow(data_list[0],...)
#X_train[idx,:len(val)] = val
#X_rp = rp.fit_transform(val.reshape(1,len(val)))
for idx, val in enumerate(total_data_train):
if idx < 10: