목록구현 (1)

soyeonland

SSD - Pytorch

The purpose of the document is study class PriorBox(object): """ Compute priorbox coordinates in center-offset form for each source feature map """ def __init__(self, cfg): super(PriorBox, self).__init__() self.image_size = cfg['min_dim'] #300 self.num_priors = len(cfg['aspect_ratios']) #6 self.variance = cfg['variance'] or [0.1] self.feature_maps = cfg['feature_maps'] #[38, 19, 10, 5, 3, 1] sel..
Study/Code Review 2020. 2. 28. 22:00
이전 Prev 1 Next 다음