-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathGet_data.py
94 lines (85 loc) · 3.95 KB
/
Get_data.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
# -*- coding: utf-8 -*-
"""
Created on Fri Nov 6 15:09:09 2020
@author: Administrator
"""
import tensorflow as tf
import numpy as np
import os
from scipy.io import wavfile
import matplotlib.pyplot as plt
import Shannon_th
from sklearn.cluster import MeanShift
from scipy import interpolate
from python_speech_features import mfcc
def txt_to_matrix(filename,fs):
file=open(filename)
lines=file.readlines()
rows=len(lines)#文件行数
datamat=np.zeros((rows,1))#初始化矩阵
row=0
for line in lines:
line=line.strip().split('\t')#strip()默认移除字符串首尾空格或换行符
#datamat[row,:]=line[:]####自己数据集，pascal数据集用这个
datamat[row,:]=float(line[0])####PhysioNet 2016 数据集用这个
row+=1
datamat=datamat.reshape([int(len(datamat)/2),2])
return np.int32(datamat*fs)####自己数据集，PhysioNet数据集为np.int32(datamat*fs)
def get_wavlist(filepath_wav,filepath_text):
fs, audio = wavfile.read(filepath_wav)
audio = audio/max(abs(audio))
labellist = txt_to_matrix(filepath_text,fs)
return audio, fs, labellist
def get_wav(audio,wav_flat):
wav = []
length = len(wav_flat)
for i in range(length):
if (wav_flat[i][1]-wav_flat[i][0])<4800:
y = interpolate.interp1d(range(wav_flat[i][1]-wav_flat[i][0]), audio[wav_flat[i][0]:wav_flat[i][1]], kind='cubic')
xint = np.linspace(0,wav_flat[i][1]-wav_flat[i][0]-1,4800)
wav_same_length = y(xint)
for j in range(len(wav_same_length)):
wav.append(wav_same_length[j])
elif (wav_flat[i][1]-wav_flat[i][0])>4800:#############这里有个大问题，直接裁剪，会把边边的给去掉
n = np.int32((wav_flat[i][1]-wav_flat[i][0]-4800)/2)
wav_same_length = audio[wav_flat[i][0]+n:wav_flat[i][1]-n]
for j in range(len(wav_same_length)):
wav.append(wav_same_length[j])
else:
for j in range(wav_flat[i][0],wav_flat[i][1],1):
wav.append(audio[j])
return wav[0:int(len(wav)/4800)*4800]#????????????
def get_normal(wav_x):
for i in range (len(wav_x)):
wav_x[i]=wav_x[i]/max(abs(wav_x[i]))
return wav_x
def get_label(test_x_flat,labellist):
label = np.zeros((len(test_x_flat),2))
labellist_cover = np.zeros((len(test_x_flat),3))
for i in range(len(test_x_flat)):
label[i,0]=1
for j in range(len(labellist)):
if (test_x_flat[i][0] <= labellist[j][0]) and (test_x_flat[i][1] >= labellist[j][1]) and (j ==0 or (test_x_flat[i][0] > labellist[j-1][1])) and (j ==len(labellist)-1 or (test_x_flat[i][1] < labellist[j+1][0])):
label[i,0]=0
label[i,1]=1
# labellist_cover[i,0]=labellist[j,0]
# labellist_cover[i,1]=labellist[j,1]
GTx = (labellist[j][0] + labellist[j][1]) / 2
GTw = labellist[j][1] - labellist[j][0]
anchorx = (test_x_flat[i][0] + test_x_flat[i][1]) / 2
anchorw = test_x_flat[i][1] - test_x_flat[i][0]
labellist_cover[i,0] = (GTx - anchorx) / anchorw
labellist_cover[i,1] = np.log(GTw / anchorw)
labellist_cover[i,2]=1
break
return label,labellist_cover
def get(filepath_wav,filepath_text):
audio,fs,labellist = get_wavlist(filepath_wav,filepath_text)
test_x_flat = Shannon_th.segment(audio,fs)
label, labellist_cover = get_label(test_x_flat,labellist)
test_ = get_wav(audio,test_x_flat)#得到数据，包括裁剪与差值
test_x = np.reshape(test_,[-1,4800])
test_x = get_normal(test_x)
# test_x是具体每一段的数据，test_x_flat是每一段的左右边界值，label是类别标签，
# labellist是具体的心音短标签，labellist_cover是与label对应的对应位置的真实位置标签
return test_x, test_x_flat, label, labellist , fs , labellist_cover ,audio