args = arg_parse()
images = args.images
outputs_names = args.outputs
batch_size = int(args.bs)
confidence = float(args.confidence)
nms_thesh = float(args.nms_thresh)
start = 0
CUDA = torch.cuda.is_available()
classes = load_classes("data/coco.names")
counter = 0
# Set up the neural network
print("Loading network.....")
model = Darknet(args.cfgfile)
model.load_weights(args.weightsfile)
print("Network successfully loaded")
model.hyperparams["height"] = args.reso
inp_dim = int(model.hyperparams["height"])
assert inp_dim % 32 == 0
assert inp_dim > 32
num_classes = model.num_classes
# If there's a GPU availible, put the model on GPU
if CUDA:
model.cuda()
# Set the model in evaluation mode
classes = load_classes(args.class_path)
# Get data configuration
data_config = parse_data_config(args.data_config_path)
train_path = data_config["train"]
# Get hyper parameters
hyperparams = parse_model_configuration(args.model_config_path)[0]
learning_rate = float(hyperparams["learning_rate"])
momentum = float(hyperparams["momentum"])
decay = float(hyperparams["decay"])
burn_in = int(hyperparams["burn_in"])
# Initiate model
model = Darknet(args.model_config_path)
if args.weights_path:
model.load_weights(args.weights_path)
else:
model.apply(weights_init_normal)
if cuda:
model = model.cuda()
model.train()
# Get dataloader
dataloader = DataLoader(
ListDataset(train_path), batch_size=args.batch_size, shuffle=False, num_workers=args.n_cpu
)