def test():
os.system('rm -rf tmp/pos/*')
os.system('rm -rf tmp/neg/*')
os.system('rm -rf tmp/part/*')
logger.info('Load WIDER')
train_data, val_data = load_wider()
print("TRAIN DATA: ", len(train_data))
img_path, bboxes = train_data[np.random.choice(len(train_data))]
bboxes = np.asarray(bboxes)
#pdb.set_trace()
img = cv2.imread(img_path, cv2.IMREAD_COLOR)
detector = JfdaDetector(cfg.PROPOSAL_NETS['r'])
#print("HERE??>> 529")
negatives, positives, part = proposal(img, bboxes, detector)
logger.info('%d gt_bboxes', len(bboxes))
logger.info('%d negatives, %d positives, %d part', len(negatives),
len(positives), len(part))
for i, (data[1], bbox_target) in enumerate(positives):
cv2.imwrite(longPath + 'tmp/pos/%03d.png' % i, data)
for i, (data) in enumerate(negatives):
cv2.imwrite(longPath + 'tmp/neg/%03d.png' % i, data[0])
#pdb.set_trace()
for i, (data[1], bbox_target) in enumerate(part):
cv2.imwrite('tmp/part/%03d.png' % i, data[0])
cv2.imwrite('tmp/test.png', img)
def gen_wider():
logger.info('loading WIDER')
train_data, val_data = load_wider()
#train_data, val_data = load_scutbrainwashcheat()
#train_data, val_data = load_cheat()
logger.info('total images, train: %d, val: %d', len(train_data),
len(val_data))
train_faces = functools.reduce(lambda acc, x: acc + len(x[1]), train_data,
0)
val_faces = functools.reduce(lambda acc, x: acc + len(x[1]), val_data, 0)
logger.info('total faces, train: %d, val: %d', train_faces, val_faces)
def gen(data, db_names):
for db_name in db_names:
remove_if_exists(db_name)
logger.info('fill queues')
q_in = [multiprocessing.Queue() for i in range(cfg.WORKER_N)]
q_out = multiprocessing.Queue(1024)
fill_queues(data, q_in)
readers = [multiprocessing.Process(target=wider_reader_func, args=(q_in[i], q_out)) \
for i in range(cfg.WORKER_N)]
#pdb.set_trace()
for p in readers:
p.start()
writer = multiprocessing.Process(target=wider_writer_func,
args=(q_out, db_names))
writer.start()
for p in readers:
p.join()
time.sleep(1)
q_out.put(('finish', []))
writer.join()
logger.info('writing train data, %d images', len(train_data))
db_names = [
'data/%snet_positive_train' % cfg.NET_TYPE,
'data/%snet_negative_train' % cfg.NET_TYPE,
'data/%snet_part_train' % cfg.NET_TYPE
]
gen(train_data, db_names)
logger.info('writing val data, %d images', len(val_data))
db_names = [
'data/%snet_positive_val' % cfg.NET_TYPE,
'data/%snet_negative_val' % cfg.NET_TYPE,
'data/%snet_part_val' % cfg.NET_TYPE
]
gen(val_data, db_names)