label_id = self.classes.index(name)
bdnbox.append(label_id)
ret += [bdnbox]
return np.array(ret)
if __name__ == "__main__":
classes = [
"aeroplane", "bicycle", "bird", "boat", "bottle", "bus", "car", "cat",
"chair", "cow", "diningtable", "dog", "horse", "motorbike", "person",
"pottedplant", "sheep", "sofa", "train", "tvmonitor"
]
anno_xml = Anno_xml(classes)
root_path = "./data/VOCdevkit/VOC2012/"
train_img_list, train_annotation_list, val_img_list, val_annotation_list = make_datapath_list(
root_path)
idx = 1
img_file_path = val_img_list[idx]
# print(img_file_path)
img = cv2.imread(img_file_path)
height, width, channels = img.shape
# cv2.imshow("img", img)
# cv2.waitKey()
annotation_infor = anno_xml(val_annotation_list[idx], width, height)
print(annotation_infor)
parser = argparse.ArgumentParser()
parser.add_argument('--batch-size', type=int, default=32)
parser.add_argument('--epochs', type=int, default=100)
FLAGS = parser.parse_args()
batch_size = FLAGS.batch_size
num_epochs = FLAGS.epochs
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
print("device:", device)
torch.backends.cudnn.benchmark = True
# dataloader
root_path = "../stereo_datasets/training"
train_img_list, train_lp_list, train_anno_list, val_img_list, val_lp_list, val_anno_list \
= make_datapath_list(root_path)
classes = [
'Car', 'Van', 'Truck', 'Pedestrian', 'Person_sitting', 'Cyclist', 'Tram',
'Misc', 'DontCare'
]
color_mean = (104, 117, 123)
input_size = 300
# img_list, anno_list, phase, transform, anno_xml
train_dataset = MyDataset(train_img_list,
train_lp_list,
train_anno_list,
phase="train",
transform=DataTransform(input_size, color_mean),