1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

""" 

Functions which are common and require SciPy Base and Level 1 SciPy 

(special, linalg) 

""" 

 

from __future__ import division, print_function, absolute_import 

 

from numpy import arange, newaxis, hstack, product, array, frombuffer, load 

 

__all__ = ['central_diff_weights', 'derivative', 'ascent', 'face', 

'electrocardiogram'] 

 

 

def central_diff_weights(Np, ndiv=1): 

""" 

Return weights for an Np-point central derivative. 

 

Assumes equally-spaced function points. 

 

If weights are in the vector w, then 

derivative is w[0] * f(x-ho*dx) + ... + w[-1] * f(x+h0*dx) 

 

Parameters 

---------- 

Np : int 

Number of points for the central derivative. 

ndiv : int, optional 

Number of divisions. Default is 1. 

 

Notes 

----- 

Can be inaccurate for large number of points. 

 

""" 

if Np < ndiv + 1: 

raise ValueError("Number of points must be at least the derivative order + 1.") 

if Np % 2 == 0: 

raise ValueError("The number of points must be odd.") 

from scipy import linalg 

ho = Np >> 1 

x = arange(-ho,ho+1.0) 

x = x[:,newaxis] 

X = x**0.0 

for k in range(1,Np): 

X = hstack([X,x**k]) 

w = product(arange(1,ndiv+1),axis=0)*linalg.inv(X)[ndiv] 

return w 

 

 

def derivative(func, x0, dx=1.0, n=1, args=(), order=3): 

""" 

Find the n-th derivative of a function at a point. 

 

Given a function, use a central difference formula with spacing `dx` to 

compute the `n`-th derivative at `x0`. 

 

Parameters 

---------- 

func : function 

Input function. 

x0 : float 

The point at which `n`-th derivative is found. 

dx : float, optional 

Spacing. 

n : int, optional 

Order of the derivative. Default is 1. 

args : tuple, optional 

Arguments 

order : int, optional 

Number of points to use, must be odd. 

 

Notes 

----- 

Decreasing the step size too small can result in round-off error. 

 

Examples 

-------- 

>>> from scipy.misc import derivative 

>>> def f(x): 

... return x**3 + x**2 

>>> derivative(f, 1.0, dx=1e-6) 

4.9999999999217337 

 

""" 

if order < n + 1: 

raise ValueError("'order' (the number of points used to compute the derivative), " 

"must be at least the derivative order 'n' + 1.") 

if order % 2 == 0: 

raise ValueError("'order' (the number of points used to compute the derivative) " 

"must be odd.") 

# pre-computed for n=1 and 2 and low-order for speed. 

if n == 1: 

if order == 3: 

weights = array([-1,0,1])/2.0 

elif order == 5: 

weights = array([1,-8,0,8,-1])/12.0 

elif order == 7: 

weights = array([-1,9,-45,0,45,-9,1])/60.0 

elif order == 9: 

weights = array([3,-32,168,-672,0,672,-168,32,-3])/840.0 

else: 

weights = central_diff_weights(order,1) 

elif n == 2: 

if order == 3: 

weights = array([1,-2.0,1]) 

elif order == 5: 

weights = array([-1,16,-30,16,-1])/12.0 

elif order == 7: 

weights = array([2,-27,270,-490,270,-27,2])/180.0 

elif order == 9: 

weights = array([-9,128,-1008,8064,-14350,8064,-1008,128,-9])/5040.0 

else: 

weights = central_diff_weights(order,2) 

else: 

weights = central_diff_weights(order, n) 

val = 0.0 

ho = order >> 1 

for k in range(order): 

val += weights[k]*func(x0+(k-ho)*dx,*args) 

return val / product((dx,)*n,axis=0) 

 

 

def ascent(): 

""" 

Get an 8-bit grayscale bit-depth, 512 x 512 derived image for easy use in demos 

 

The image is derived from accent-to-the-top.jpg at 

http://www.public-domain-image.com/people-public-domain-images-pictures/ 

 

Parameters 

---------- 

None 

 

Returns 

------- 

ascent : ndarray 

convenient image to use for testing and demonstration 

 

Examples 

-------- 

>>> import scipy.misc 

>>> ascent = scipy.misc.ascent() 

>>> ascent.shape 

(512, 512) 

>>> ascent.max() 

255 

 

>>> import matplotlib.pyplot as plt 

>>> plt.gray() 

>>> plt.imshow(ascent) 

>>> plt.show() 

 

""" 

import pickle 

import os 

fname = os.path.join(os.path.dirname(__file__),'ascent.dat') 

with open(fname, 'rb') as f: 

ascent = array(pickle.load(f)) 

return ascent 

 

 

def face(gray=False): 

""" 

Get a 1024 x 768, color image of a raccoon face. 

 

raccoon-procyon-lotor.jpg at http://www.public-domain-image.com 

 

Parameters 

---------- 

gray : bool, optional 

If True return 8-bit grey-scale image, otherwise return a color image 

 

Returns 

------- 

face : ndarray 

image of a racoon face 

 

Examples 

-------- 

>>> import scipy.misc 

>>> face = scipy.misc.face() 

>>> face.shape 

(768, 1024, 3) 

>>> face.max() 

255 

>>> face.dtype 

dtype('uint8') 

 

>>> import matplotlib.pyplot as plt 

>>> plt.gray() 

>>> plt.imshow(face) 

>>> plt.show() 

 

""" 

import bz2 

import os 

with open(os.path.join(os.path.dirname(__file__), 'face.dat'), 'rb') as f: 

rawdata = f.read() 

data = bz2.decompress(rawdata) 

face = frombuffer(data, dtype='uint8') 

face.shape = (768, 1024, 3) 

if gray is True: 

face = (0.21 * face[:,:,0] + 0.71 * face[:,:,1] + 0.07 * face[:,:,2]).astype('uint8') 

return face 

 

 

def electrocardiogram(): 

""" 

Load an electrocardiogram as an example for a one-dimensional signal. 

 

The returned signal is a 5 minute long electrocardiogram (ECG), a medical 

recording of the heart's electrical activity, sampled at 360 Hz. 

 

Returns 

------- 

ecg : ndarray 

The electrocardiogram in millivolt (mV) sampled at 360 Hz. 

 

Notes 

----- 

The provided signal is an excerpt (19:35 to 24:35) from the `record 208`_ 

(lead MLII) provided by the MIT-BIH Arrhythmia Database [1]_ on 

PhysioNet [2]_. The excerpt includes noise induced artifacts, typical 

heartbeats as well as pathological changes. 

 

.. _record 208: https://physionet.org/physiobank/database/html/mitdbdir/records.htm#208 

 

.. versionadded:: 1.1.0 

 

References 

---------- 

.. [1] Moody GB, Mark RG. The impact of the MIT-BIH Arrhythmia Database. 

IEEE Eng in Med and Biol 20(3):45-50 (May-June 2001). 

(PMID: 11446209); https://doi.org/10.13026/C2F305 

.. [2] Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, 

Mark RG, Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, 

PhysioToolkit, and PhysioNet: Components of a New Research Resource 

for Complex Physiologic Signals. Circulation 101(23):e215-e220; 

https://doi.org/10.1161/01.CIR.101.23.e215 

 

Examples 

-------- 

>>> from scipy.misc import electrocardiogram 

>>> ecg = electrocardiogram() 

>>> ecg 

array([-0.245, -0.215, -0.185, ..., -0.405, -0.395, -0.385]) 

>>> ecg.shape, ecg.mean(), ecg.std() 

((108000,), -0.16510875, 0.5992473991177294) 

 

As stated the signal features several areas with a different morphology. 

E.g. the first few seconds show the electrical activity of a heart in 

normal sinus rhythm as seen below. 

 

>>> import matplotlib.pyplot as plt 

>>> fs = 360 

>>> time = np.arange(ecg.size) / fs 

>>> plt.plot(time, ecg) 

>>> plt.xlabel("time in s") 

>>> plt.ylabel("ECG in mV") 

>>> plt.xlim(9, 10.2) 

>>> plt.ylim(-1, 1.5) 

>>> plt.show() 

 

After second 16 however, the first premature ventricular contractions, also 

called extrasystoles, appear. These have a different morphology compared to 

typical heartbeats. The difference can easily be observed in the following 

plot. 

 

>>> plt.plot(time, ecg) 

>>> plt.xlabel("time in s") 

>>> plt.ylabel("ECG in mV") 

>>> plt.xlim(46.5, 50) 

>>> plt.ylim(-2, 1.5) 

>>> plt.show() 

 

At several points large artifacts disturb the recording, e.g.: 

 

>>> plt.plot(time, ecg) 

>>> plt.xlabel("time in s") 

>>> plt.ylabel("ECG in mV") 

>>> plt.xlim(207, 215) 

>>> plt.ylim(-2, 3.5) 

>>> plt.show() 

 

Finally, examining the power spectrum reveals that most of the biosignal is 

made up of lower frequencies. At 60 Hz the noise induced by the mains 

electricity can be clearly observed. 

 

>>> from scipy.signal import welch 

>>> f, Pxx = welch(ecg, fs=fs, nperseg=2048, scaling="spectrum") 

>>> plt.semilogy(f, Pxx) 

>>> plt.xlabel("Frequency in Hz") 

>>> plt.ylabel("Power spectrum of the ECG in mV**2") 

>>> plt.xlim(f[[0, -1]]) 

>>> plt.show() 

""" 

import os 

file_path = os.path.join(os.path.dirname(__file__), "ecg.dat") 

with load(file_path) as file: 

ecg = file["ecg"].astype(int) # np.uint16 -> int 

# Convert raw output of ADC to mV: (ecg - adc_zero) / adc_gain 

ecg = (ecg - 1024) / 200.0 

return ecg