/
inference_time.py
131 lines (110 loc) · 4.2 KB
/
inference_time.py
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
import helpermethods as helper
import numpy as np
import sys
import edgeml_pytorch.utils as utils
from edgeml_pytorch.graph.protoNN import ProtoNN
import torch
import time
import scipy
from antropy.antropy import entropy
from scipy.signal import periodogram, welch
import pandas as pd
# HyperParams
hyperParam = {}
fs = 32
window = np.load('1window.npy')
window_length = fs * 8
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# dataDimension = window.shape[0]
dataDimension = 20
PROJECTION_DIM = 10
NUM_PROTOTYPES = 20
numClasses = 2
times = list()
# Feature Methods
def hMob(x):
row = np.array(x)
return (np.sqrt(np.var(np.gradient(x)) / np.var(x)))
def feature_mean(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(np.mean(x, axis = 0) for x in row)
def feature_stddev(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(np.std(x, axis = 0) for x in row)
def kurtosis(row):
row = np.array(row)
# annotation = row[-1]
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(scipy.stats.kurtosis(x, axis = 0, bias = False) for x in row)
def skewness(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(scipy.stats.skew(x, axis = 0, bias = False) for x in row)
def spectral_entropy(row, sf = 32, nperseg = window_length, axis = 1):
row = np.array(row)
# annotation = row[-1]
# row = row[:-1]
row = np.reshape(row, (21, window_length))
_, psd = welch(row, sf, nperseg=nperseg, axis=axis)
psd_norm = psd / psd.sum(axis=axis, keepdims=True)
se = - np.where(psd_norm == 0, 0, psd_norm * np.log(psd_norm) / np.log(2)).sum(axis=axis)
return pd.Series(se)
def hjorthActivity(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(np.var(x, axis = 0) for x in row)
def hjorthMobility(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(np.sqrt(np.var(np.gradient(x)) / np.var(x)) for x in row)
def hjorthComplexity(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series((hMob(np.gradient(x)) / hMob(x)) for x in row)
def permutation_entropy(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(entropy.perm_entropy(x) for x in row)
def sample_entropy(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.array(entropy.sample_entropy(x) for x in row)
def approximate_entropy(row):
row = np.array(row)
# row = row[:-1]
row = np.reshape(row, (21, window_length))
return pd.Series(entropy.app_entropy(x) for x in row)
def getFeatures(window):
list_of_feature_methods = [feature_mean, feature_stddev, kurtosis, skewness, spectral_entropy, hjorthActivity, hjorthMobility,
hjorthComplexity, permutation_entropy, sample_entropy, approximate_entropy]
return np.array([i(window) for i in list_of_feature_methods])
def loadModel():
hyperParam['B'] = np.load('WEIGHTS/1sec/output_pca_20_test_20_lr_0.0001_e_1000/B.npy')
hyperParam['gamma'] = np.load('WEIGHTS/1sec/output_pca_20_test_20_lr_0.0001_e_1000/gamma.npy')
hyperParam['W'] = np.load('WEIGHTS/1sec/output_pca_20_test_20_lr_0.0001_e_1000/W.npy')
hyperParam['Z'] = np.load('WEIGHTS/1sec/output_pca_20_test_20_lr_0.0001_e_1000/Z.npy')
return
loadModel()
# print(hyperParam['W'].shape, dataDimension, PROJECTION_DIM)
protoNN = ProtoNN(dataDimension, PROJECTION_DIM, NUM_PROTOTYPES, numClasses, hyperParam['gamma'], W = hyperParam['W'], B = hyperParam['B']).to(device)
for i in range(int(sys.argv[1])):
start = time.time()
features = getFeatures(window)
features = features.flatten()[:200].reshape((10, 20))
logits = protoNN.forward(torch.Tensor(features).to(device))
_, _ = torch.max(logits, dim=1)
end = time.time()
times.append(end - start)
# print(times)
print("%.5f milli seconds!" % (np.mean(times) * 1000))