def write_csv_files(self):
files = glob(os.path.join(self.image_dir, "*.jpg"))
files.sort()
if self.val_frac != 1.0:
filemap, idmap, x1map, y1map, x2map, y2map = read_labels(
self.image_dir, self.points1_file, self.points2_file, self.target_size
)
if self.id_label == 1:
self.label_names = ["id"]
else:
self.label_names = ["x1", "y1", "x2", "y2"]
indexes = range(len(self.label_names))
self.label_dict = {k: v for k, v in zip(self.label_names, indexes)}
tlines = []
vlines = []
np.random.shuffle(files)
v_idx = int(self.val_frac * len(files))
tfiles = files[v_idx:]
vfiles = files[:v_idx]
vfiles.sort()
if self.id_label == 1:
if self.val_frac == 1.0:
vlines = [(f, 0) for f in vfiles]
else:
tlines = [(f, idmap[filemap[f]]) for f in tfiles]
else:
if self.val_frac == 1.0:
vlines = [(f, 0, 0, 0, 0) for f in vfiles]
else:
tlines = [(f, x1map[f], y1map[f], x2map[f], y2map[f]) for f in tfiles]
np.random.shuffle(tlines)
if not os.path.exists(self.out_dir):
os.makedirs(self.out_dir)
for ff, ll in zip([self.train_file, self.val_file], [tlines, vlines]):
with open(ff, "wb") as f:
if self.id_label == 1:
f.write("filename,id\n")
for tup in ll:
f.write("{},{}\n".format(*tup))
else:
f.write("filename,x,y\n")
for tup in ll:
f.write("{},{},{},{},{}\n".format(*tup))
self.train_nrec = len(tlines)
self.ntrain = -(-self.train_nrec // self.macro_size)
self.train_start = 0
self.val_nrec = len(vlines)
self.nval = -(-self.val_nrec // self.macro_size)
if self.ntrain == 0:
self.val_start = 100
else:
self.val_start = 10 ** int(np.log10(self.ntrain * 10))
def __init__(self, callback_data, model, dataset, traindir, imwidth,
epochs, datadir, pointnum):
super(Evaluator, self).__init__()
self.model = model
self.dataset = dataset
self.callback_data = callback_data
self.imwidth = imwidth
self.epochs = epochs
self.min_dist = 4 * imwidth
traindir = os.path.join(os.path.dirname(datadir), traindir)
_, _, x1map, y1map, x2map, y2map = read_labels(
traindir, 'points1.json', 'points2.json', imwidth)
xmap, ymap = (x1map, y1map) if pointnum == 1 else (x2map, y2map)
self.xymean = np.array([np.mean(xmap.values()),
np.mean(ymap.values())])
def __init__(self, callback_data, model, dataset, imwidth, epochs, datadir,
pointnum):
super(Evaluator, self).__init__()
self.model = model
self.dataset = dataset
self.callback_data = callback_data
self.imwidth = imwidth
self.epochs = epochs
self.min_dist = 4 * imwidth
traindir = os.path.join(os.path.dirname(datadir), 'train')
_, _, x1map, y1map, x2map, y2map = read_labels(traindir,
'points1.json',
'points2.json', imwidth)
xmap, ymap = (x1map, y1map) if pointnum == 1 else (x2map, y2map)
self.xymean = np.array(
[np.mean(xmap.values()),
np.mean(ymap.values())])
def get_features_labels(data_set, wavelet_transform=0):
parent_dir = 'heartbeat-sounds'
training_set = data_set + '_training'
testing_set = data_set + '_testing'
CSV = data_set + '.csv'
if wavelet_transform == 1:
print('Performing Wavelet Transformation')
tr_sub_dirs = [training_set]
ts_sub_dirs = [testing_set]
new_labels, label_array = labels.read_labels(CSV)
tr_features, tr_labels = parse_audio_files(parent_dir, tr_sub_dirs,
new_labels, wavelet_transform)
ts_features, ts_labels = parse_audio_files(parent_dir, ts_sub_dirs,
new_labels, wavelet_transform)
tr_labels = one_hot_encode(tr_labels)
ts_labels = one_hot_encode(ts_labels)
return tr_features, tr_labels, ts_features, ts_labels
from keras.optimizers import Adam
from keras.callbacks import ModelCheckpoint, LearningRateScheduler, EarlyStopping
from keras.utils import np_utils
from keras.regularizers import l2
import labels
import wavelet_transformation
directory = 'set_'+sys.argv[1]
wavelet_transform = int(sys.argv[2])
INPUT_LIB = 'heartbeat-sounds/'
SAMPLE_RATE = 4000
CSV = directory+'.csv'
DATASET = directory+'/'
new_labels,label_array = labels.read_labels(CSV)
def clean_filename(fname, string):
file_name = fname.split('/')[1]
if file_name[:2] == '__':
file_name = string + file_name
return file_name
def load_wav_file(name, path, wavelet_transform=1):
_, audio_data = wavfile.read(path + name)
#print(_)
#assert _ == SAMPLE_RATE
if wavelet_transform == 1:
audio_data = wavelet_transformation.wavelet_transformation(audio_data)
return audio_data
import audios
import coders
import labels
import videos
import youtube
if args.loop or args.play:
coders.write_video_batch()
sys.exit()
with multiprocessing.Manager() as manager:
labels.labels = manager.list()
videos.videos = manager.dict()
random.seed(int.from_bytes(os.getrandom(4), 'big'))
if not args.labels_reinit:
labels.read_labels()
else:
if args.input_labels:
labels.read_labels(args.input_labels)
else:
if args.labels_from_input:
assert args.input
created_labels = videos.labels_from_video(args.input)
labels.update_labels(created_labels)
labels.write_labels()
if args.audios:
args.audios[:] = audios.read_audios()
if not labels.labels:
if args.labels_public:
created_labels = youtube.obtain_labels(args.audios[0])
if not args.labels_public or not created_labels:
