(gsr_data["GSR_Timestamp_Unix_CAL"].values >= start)
& (gsr_data["GSR_Timestamp_Unix_CAL"].values <= end)]
eda = gsr_split["GSR_GSR_Skin_Conductance_CAL"].values.flatten()
ppg = gsr_split["GSR_PPG_A13_CAL"].values.flatten()
status = 0
# extract eda features
#scr_features = eda_features_exct.extractSCRFeatures(eda)
#extract cvx of eda and time domain of ppg to check whether the inputs are not disorted
cvx_features = eda_features_exct.extractCVXEDA(eda)
ppg_time = ppg_features_exct.extractTimeDomain(ppg)
if (cvx_features.shape[0] != 0) & (ppg_time.shape[0] != 0):
eda_features = np.concatenate([
cvx_features,
eda_features_exct.extractMFCCFeatures(eda, min_len=min_len)
])
# extract PPG features
ppg_features = np.concatenate([
ppg_time,
ppg_features_exct.extractFrequencyDomain(ppg),
ppg_features_exct.extractNonLinear(ppg)
])
if (np.sum(np.isinf(ppg_features)) ==
0 | np.sum(np.isnan(ppg_features)) == 0):
# print(eda_features.shape)
# save features
np.save(path_results + "eda_" + str(idx) + ".npy",
eda_features)
from GSR.GSRFeatures import PPGFeatures, EDAFeatures
from Conf import Settings as set
import pandas as pd
data = pd.read_csv("..\\Data\\Dummy\\GSR_sample.csv", header=[0, 1])
#PPG
ppg = data["GSR_PPG_A13_CAL"].iloc[0:900].values
ppgExtract = PPGFeatures(set.FS_GSR)
print(ppgExtract.extractTimeDomain(ppg.flatten()))
print(ppgExtract.extractFrequencyDomain(ppg.flatten()))
print(ppgExtract.extractNonLinear(ppg.flatten()))
#EDA
eda = data["GSR_GSR_Skin_Conductance_CAL"].iloc[0:900].values
edaExtract = EDAFeatures(set.FS_GSR)
print(edaExtract.extractMFCCFeatures(ppg.flatten()))
print(edaExtract.extractSCRFeatures(ppg.flatten()))