def main():
args = parse_args()
cfg, model_name = _trim(get_config(args.config, show=False), args)
print(f"Building model({model_name})...")
model = build_model(cfg)
assert osp.isfile(
args.pretrained_params
), f"pretrained params ({args.pretrained_params} is not a file path.)"
if not os.path.isdir(args.output_path):
os.makedirs(args.output_path)
print(f"Loading params from ({args.pretrained_params})...")
params = paddle.load(args.pretrained_params)
model.set_dict(params)
model.eval()
model = to_static(model,
input_spec=[
paddle.static.InputSpec(shape=[
None, args.num_seg, 3, args.img_size,
args.img_size
],
dtype='float32'),
])
paddle.jit.save(model, osp.join(args.output_path, model_name))
print(
f"model ({model_name}) has been already saved in ({args.output_path}).")
def main():
args = parse_args()
cfg, model_name = _trim(get_config(args.config, show=False))
print(f"Building model({model_name})...")
model = build_model(cfg)
params_info = paddle.summary(model, (1, 8, 3, 224, 224))
print(params_info)
def main():
args = parse_args()
cfg = get_config(args.config, overrides=args.override)
dataset = build_dataset((cfg.DATASET.test, cfg.PIPELINE.test))
_, world_size = get_dist_info()
parallel = world_size != 1
if parallel:
paddle.distributed.init_parallel_env()
model = build_model(cfg.MODEL)
test_model(model, dataset, cfg, args.weights, world_size)
def main():
args = parse_args()
cfg, model_name = _trim(get_config(args.config, show=False), args)
print(f"Building model({model_name})...")
model = build_model(cfg)
img_size = args.img_size
num_seg = args.num_seg
#NOTE: only support tsm now, will refine soon
params_info = paddle.summary(model, (1, num_seg, 3, img_size, img_size))
print(params_info)
if args.FLOPs:
flops_info = paddle.flops(model, [1, num_seg, 3, img_size, img_size],
print_detail=True)
print(flops_info)
def main():
args = parse_args()
cfg, model_name = trim_config(get_config(args.config, show=False))
print(f"Building model({model_name})...")
model = build_model(cfg.MODEL)
assert osp.isfile(
args.pretrained_params
), f"pretrained params ({args.pretrained_params} is not a file path.)"
if not os.path.isdir(args.output_path):
os.makedirs(args.output_path)
print(f"Loading params from ({args.pretrained_params})...")
params = paddle.load(args.pretrained_params)
model.set_dict(params)
model.eval()
input_spec = get_input_spec(cfg.INFERENCE, model_name)
model = to_static(model, input_spec=input_spec)
paddle.jit.save(model, osp.join(args.output_path, model_name))
print(
f"model ({model_name}) has been already saved in ({args.output_path}).")