def main():
args = parse_args()
ctx = mx.gpu(args.gpu)
symbol = get_resnet_test(num_classes=config.NUM_CLASSES,
num_anchors=config.NUM_ANCHORS)
predictor = get_net(symbol, args.prefix, args.epoch, ctx)
demo_net(predictor, args.image, args.vis)
def main():
args = parse_args()
ctx = mx.gpu(args.gpu)
sym = get_resnet_test(num_classes=config.NUM_CLASSES, num_anchors=config.NUM_ANCHORS)
data, image_names, im_scales = load_data()
predictor = get_net(data, sym, args.prefix, args.epoch, ctx)
demo_net(predictor, data, image_names, im_scales)
def create_net(configs):
logger.info("[Python create_net] load configs: %s",
configs,
extra={"reqid": ""})
tar_files_name = 'model_files'
# load tar files
if tar_files_name not in configs:
return None, 400, {"code": 400, "message": 'no field "tar_files"'}
tar_files = configs[tar_files_name]
conf, err = net.parse_infer_config(tar_files)
if err:
return None, 400, {"code": 400, "message": err}
params_file, sym_file, label_file = (conf.weight, conf.deploy_sym,
conf.labels)
use_device_name = 'use_device'
if use_device_name not in configs:
return None, 400, {"code": 400, "message": 'no field "use_device"'}
use_device = configs[use_device_name]
threshold = CONF_THRESH
if 'custom_params' in configs:
custom_values = configs['custom_params']
if 'threshold' in custom_values:
threshold = custom_values["threshold"]
ctx = mx.gpu() if use_device == 'GPU' else mx.cpu() # TODO set the gpu/cpu
classes = _load_cls(label_file)
symbol = get_resnet_test(num_classes=len(classes),
num_anchors=config.NUM_ANCHORS)
os.rename(sym_file, sym_file + '-symbol.json')
os.rename(params_file, sym_file + '-0000.params')
logger.info("params_file: %s, sym_file:%s,label_file:%s",
params_file,
sym_file,
label_file,
extra={"reqid": ""})
logger.info("use_device: %s, threshold:%s,classes:%s,symbol:%s",
use_device,
threshold,
classes,
symbol,
extra={"reqid": ""})
return dict(error='',
predictor=get_net(symbol, sym_file, 0, ctx),
classes=classes,
threshold=threshold), 0, None
help='saved model prefix',
default='model/e2e',
type=str)
parser.add_argument('--epoch',
help='epoch of pretrained model',
default=50,
type=int)
parser.add_argument('--gpu', help='GPU device to use', default=0, type=int)
parser.add_argument('--vis', help='display result', action='store_true')
args = parser.parse_args()
return args
if __name__ == '__main__':
args = parse_args()
ctx = mx.gpu(0)
if (Global.prefix_value == 'model/e2e'):
symbol = get_resnet_test(num_classes=Global.num_class_value,
num_anchors=config.NUM_ANCHORS)
if (Global.prefix_value == 'model/final'):
symbol = get_vgg_test(num_classes=Global.num_class_value,
num_anchors=config.NUM_ANCHORS)
predictor = get_net(symbol, args.prefix, ctx)
from glob import glob
res = glob("/root/mx-rcnn/testimage/bird/*.JPG")
cnt = 0
for i in res:
demo_net(predictor, i, False)
# if cnt>10:
# break