def GetAllData(fe_fd, csv_file, agg_num, hop):
"""
Input: Features folder(String), CSV file(String), agg_num(Integer), hop(Integer).
Output: Loaded features(Numpy Array) and labels(Numpy Array).
Loads all the features saved as pickle files.
"""
# read csv
with open( csv_file, 'rb') as f:
reader = csv.reader(f)
lis = list(reader)
# init list
X3d_all = []
y_all = []
i=0
for li in lis:
# load data
[na, lb] = li[0].split('\t')
na = na.split('/')[1][0:-4]
path = fe_fd + '/' + na + '.f'
X = aud_feature.load(path)
# reshape data to (n_block, n_time, n_freq)
i+=1
X3d = aud_utils.mat_2d_to_3d( X, agg_num, hop )
X3d_all.append( X3d )
y_all += [ lb_to_id[lb] ] * len( X3d )
print "Features loaded",i
print 'All files loaded successfully'
# concatenate list to array
X3d_all = np.concatenate( X3d_all )
y_all = np.array( y_all )
return X3d_all, y_all
def get_train_data(self):
"""
Loads all the features saved as pickle files.
Input: Features folder(str), CSV file(str)
Output: Loaded features(np array) and labels(np array).
"""
feature = self.feature
fe_fd = cfg.dev_fd + '/' + feature
csv_file = cfg.meta_train_csv
agg_num = self.agg_num
hop = self.hop
# read csv
with open(csv_file, 'rb') as f:
reader = csv.reader(f)
lis = list(reader)
# init list
X3d_all = []
y_all = []
i = 0
for li in lis:
# load data
na = li[1]
path = fe_fd + '/' + na + '.f'
info_path = cfg.label_csv + '/' + na + '.csv'
with open(info_path, 'rb') as g:
reader2 = csv.reader(g)
lis2 = list(reader2)
tags = lis2[-2][1]
y = np.zeros(len(cfg.labels))
for ch in tags:
y[cfg.lb_to_id[ch]] = 1
X = aud_feature.load(path)
# reshape data to (n_block, n_time, n_freq)
i += 1
X3d = aud_utils.mat_2d_to_3d(X, agg_num, hop)
X3d_all.append(X3d)
y_all += [y] * len(X3d)
print "Features loaded", i
print 'All files loaded successfully'
# concatenate list to array
X3d_all = np.concatenate(X3d_all)
y_all = np.array(y_all)
return X3d_all, y_all
def GetAllData(fe_fd, csv_file, agg_num, hop):
"""
Input: Features folder(String), CSV file(String), agg_num(Integer), hop(Integer).
Output: Loaded features(Numpy Array) and labels(Numpy Array).
Loads all the features saved as pickle files.
"""
# read csv
with open(csv_file, 'rb') as f:
reader = csv.reader(f)
lis = list(reader)
# init list
X3d_all = []
y_all = []
i = 0
for li in lis:
# load data
na = li[1]
path = fe_fd + '/' + na + '.f'
info_path = label_csv + '/' + na + '.csv'
with open(info_path, 'rb') as g:
reader2 = csv.reader(g)
lis2 = list(reader2)
tags = lis2[-2][1]
y = np.zeros(len(labels))
for ch in tags:
y[lb_to_id[ch]] = 1
X = aud_feature.load(path)
# reshape data to (n_block, n_time, n_freq)
i += 1
X3d = aud_utils.mat_2d_to_3d(X, agg_num, hop)
X3d_all.append(X3d)
y_all += [y] * len(X3d)
print "Features loaded", i
print 'All files loaded successfully'
# concatenate list to array
X3d_all = np.concatenate(X3d_all)
y_all = np.array(y_all)
return X3d_all, y_all
def get_train_data(self):
"""
Loads all the features saved as pickle files.
Input: Features folder(str), CSV file(str)
Output: Loaded features(np array) and labels(np array).
"""
feature = self.feature
fe_fd = cfg.dev_fd + '/' + feature
csv_file = cfg.label_csv
agg_num = self.agg_num
hop = self.hop
# read csv
with open(csv_file, 'rb') as f:
reader = csv.reader(f)
lis = list(reader)
# init list
X3d_all = []
y_all = []
i = 0
for li in lis:
# load data
[na, lb] = li[0].split('\t')
na = na.split('/')[1][0:-4]
path = fe_fd + '/' + na + '.f'
X = aud_feature.load(path)
# reshape data to (n_block, n_time, n_freq)
i += 1
X3d = aud_utils.mat_2d_to_3d(X, agg_num, hop)
X3d_all.append(X3d)
y_all += [cfg.lb_to_id[lb]] * len(X3d)
print "Features loaded", i
print 'All files loaded successfully'
# concatenate list to array
X3d_all = np.concatenate(X3d_all)
y_all = np.array(y_all)
return X3d_all, y_all
