forked from affctivai/coglier
-
Notifications
You must be signed in to change notification settings - Fork 0
/
0.0 GAMEEMO preprocessing.py
142 lines (118 loc) · 5.85 KB
/
0.0 GAMEEMO preprocessing.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
132
133
134
135
136
137
138
139
140
141
142
import os
from os.path import join
import pandas as pd
import numpy as np
from utils.constant import *
import argparse
def save_datas_seg(src, window, stride, emotions, channels ,sublist, save_path):
print('Segmentation x: (samples, channels, winodw size), y: (samples, 2) # [label, subID]')
for subnum in sublist:
print('sub ID:',subnum, end=' ')
sub_x, sub_y = [], []
sub_v, sub_a = [], []
for emo in emotions:
EEG_path = join(src, f'(S{subnum})','Preprocessed EEG Data','.csv format',f'S{subnum}{emo}'+'AllChannels.csv')
csv = pd.read_csv(EEG_path, usecols=channels)
data = csv.to_numpy()
SAM_path = join(src, f'(S{subnum})','SAM Ratings',f'{emo}.txt')
label = open(SAM_path, 'r')
label = label.readline().strip()
valence, arousal = int(label[1]), int(label[-1])
n = len(data)
idx = 0
while idx + window < n:
seg = data[idx : idx + window]
seg = seg.swapaxes(0, 1)
sub_x.append(seg)
sub_y.append([label, int(subnum)])
sub_v.append([valence-1, int(subnum)])
sub_a.append([arousal-1, int(subnum)])
idx += stride
sub_x = np.array(sub_x); sub_y = np.array(sub_y); sub_v = np.array(sub_v); sub_a = np.array(sub_a);
print(f'EEG:{sub_x.shape} aro:{sub_a.shape} val:{sub_v.shape}')
os.makedirs(save_path, exist_ok=True)
np.savez(join(save_path, subnum), x=sub_x, y=sub_y, v=sub_v, a=sub_a)
print(f'saved in {save_path}')
def save_datas_seg_DE(src, window, stride, emotions, channels ,sublist, save_path):
from utils.transform import BandDifferentialEntropy
print('Segmentation with DE x: (samples, channels, 4 freq), y: (samples, 2) # [label, subID]')
for subnum in sublist:
print('sub ID:',subnum, end=' ')
sub_x, sub_y = [], []
sub_v, sub_a = [], []
for emo in emotions:
EEG_path = join(src, f'(S{subnum})','Preprocessed EEG Data','.csv format',f'S{subnum}{emo}'+'AllChannels.csv')
csv = pd.read_csv(EEG_path, usecols=channels)
data = csv.to_numpy()
SAM_path = join(src, f'(S{subnum})','SAM Ratings',f'{emo}.txt')
label = open(SAM_path, 'r')
label = label.readline().strip()
valence, arousal = int(label[1]), int(label[-1])
n = len(data)
idx = 0
while idx + window < n:
seg = data[idx : idx + window]
seg = seg.swapaxes(0, 1)
bde = BandDifferentialEntropy()
sub_x.append(bde.apply(seg))
sub_y.append([label, int(subnum)])
sub_v.append([valence-1, int(subnum)])
sub_a.append([arousal-1, int(subnum)])
idx += stride
sub_x = np.array(sub_x); sub_y = np.array(sub_y); sub_v = np.array(sub_v); sub_a = np.array(sub_a);
print(f'EEG:{sub_x.shape} aro:{sub_a.shape} val:{sub_v.shape}')
os.makedirs(save_path, exist_ok=True)
np.savez(join(save_path, subnum), x=sub_x, y=sub_y, v=sub_v, a=sub_a)
print(f'saved in {save_path}')
def save_datas_seg_PSD(src, window, stride, emotions, channels ,sublist, save_path):
from utils.transform import BandPowerSpectralDensity
print('Segmentation with PSD x: (samples, channels, 4 freq), y: (samples, 2) # [label, subID]')
for subnum in sublist:
print('sub ID:',subnum, end=' ')
sub_x, sub_y = [], []
sub_v, sub_a = [], []
for emo in emotions:
EEG_path = join(src, f'(S{subnum})','Preprocessed EEG Data','.csv format',f'S{subnum}{emo}'+'AllChannels.csv')
csv = pd.read_csv(EEG_path, usecols=channels)
data = csv.to_numpy()
SAM_path = join(src, f'(S{subnum})','SAM Ratings',f'{emo}.txt')
label = open(SAM_path, 'r')
label = label.readline().strip()
valence, arousal = int(label[1]), int(label[-1])
n = len(data)
idx = 0
while idx + window < n:
seg = data[idx : idx + window]
seg = seg.swapaxes(0, 1)
psd = BandPowerSpectralDensity()
sub_x.append(psd.apply(seg))
sub_y.append([label, int(subnum)])
sub_v.append([valence-1, int(subnum)])
sub_a.append([arousal-1, int(subnum)])
idx += stride
sub_x = np.array(sub_x); sub_y = np.array(sub_y); sub_v = np.array(sub_v); sub_a = np.array(sub_a);
print(f'EEG:{sub_x.shape} aro:{sub_a.shape} val:{sub_v.shape}')
os.makedirs(save_path, exist_ok=True)
np.savez(join(save_path, subnum), x=sub_x, y=sub_y, v=sub_v, a=sub_a)
print(f'saved in {save_path}')
# -----------------------------------------main---------------------------------------------------
parser = argparse.ArgumentParser()
parser.add_argument("--src_dir", type=str, default="/mnt/data")
parser.add_argument("--window", type=int, default=256)
parser.add_argument("--stride", type=int, default=128)
parser.add_argument("--method", type=str, default="seg", help='seg, PSD, DE')
args = parser.parse_args()
SRC = args.src_dir
WINDOW = args.window
STRIDE = args.stride
METHOD = args.method
SUBLIST = [str(i).zfill(2) for i in range(1, GAMEEMO_SUBNUM + 1)]
EMOS = ['G1', 'G2', 'G3', 'G4']
src_dir = join(SRC, 'GAMEEMO')
save_dir = join(os.getcwd(), 'datasets', 'GAMEEMO', 'npz', 'Preprocessed')
if METHOD == 'seg':
save_datas_seg(src_dir, WINDOW, STRIDE, EMOS, GAMEEMO_CHLS, SUBLIST, join(save_dir, 'seg'))
elif METHOD == 'PSD':
save_datas_seg_PSD(src_dir, WINDOW, STRIDE, EMOS, GAMEEMO_CHLS, SUBLIST, join(save_dir, 'seg_PSD'))
elif METHOD == 'DE':
save_datas_seg_DE(src_dir, WINDOW, STRIDE, EMOS, GAMEEMO_CHLS, SUBLIST, join(save_dir, 'seg_DE'))