rhett-chen
代码的注释是不是有问题
如下所示： seed_xyz = end_points['point_clouds'] # use all sampled point cloud, BNs3 B, point_num, _ = seed_xyz.shape # batch _size coordinates_batch = end_points['coors'] features_batch = end_points['feats'] mink_input = ME.SparseTensor(features_batch, coordinates=coordinates_batch) seed_features = self.backbone(mink_input).F seed_features = seed_features[end_points['quantize2original']].view(B, point_num, -1).transpose(1, 2) end_points = self.graspable(seed_features, end_points) seed_features_flipped = seed_features.transpose(1, 2) # BNsfeat_dim objectness_score = end_points['objectness_score'] graspness_score = end_points['graspness_score'].squeeze(1) objectness_pred = torch.argmax(objectness_score, 1) objectness_mask = (objectness_pred == 1) graspness_mask = graspness_score > GRASPNESS_THRESHOLD graspable_mask = objectness_mask & graspness_mask seed_features_graspable = [] seed_xyz_graspable = [] graspable_num_batch = 0. for i in range(B): cur_mask = graspable_mask[i] graspable_num_batch += cur_mask.sum() cur_feat = seed_features_flipped[i][cur_mask] # Nsfeat_dim cur_seed_xyz = seed_xyz[i][cur_mask] # Ns3 seed_xyz [i]大小是Ns3 cur_seed_xyz = seed_xyz[i][cur_mask]的大小还是Ns3？
