def main(_):
opt = {
"config": YOLO_CONFIG,
"model": YOLO_MODEL,
"load": YOLO_WEIGHTS,
"threshold": YOLO_THRES
}
tfnet = TFNet(opt)
channel = grpc.insecure_channel(FLAGS.server)
stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
image = cv2.imread(FLAGS.image)
image = tfnet.framework.resize_input(image)
image = np.expand_dims(image, 0)
# # generate batched input
# batch_size = 4
# batched_im = image
# for i in range(batch_size - 1):
# batched_im = np.append(batched_im, image, axis = 0)
# debug only
image2 = cv2.imread("/home/yitao/Documents/edge/D2-system/data/dog.jpg")
image2 = tfnet.framework.resize_input(image2)
image2 = np.expand_dims(image2, 0)
batched_im = image
batched_im = np.append(batched_im, image2, axis=0)
print(batched_im.shape)
# print(batched_im[0].shape)
for i in range(10):
start = time.time()
batched_result = tfnet.return_batched_predict(batched_im,
"traffic_yolo", stub)
end = time.time()
print("duration = %.3f sec" % (end - start))
# print(batched_result)
batched_ouput = []
for dets in batched_result:
output = ""
for d in dets:
output += "%s|%s|%s|%s|%s|%s-" % (
str(d['topleft']['x']), str(d['topleft']['y']),
str(d['bottomright']['x']), str(d['bottomright']['y']),
str(d['confidence']), str(d['label']))
batched_ouput.append(output[:-1])
print(batched_ouput)
