问题遇到的现象和发生背景
Faster—Rcnn的ROIhead部分代码理解
问题相关代码，请勿粘贴截图
for i in range(n):
# ------------------------------------------------------ #
# 根据建议框的种类，取出对应的回归预测结果
# ------------------------------------------------------ #
n_sample = roi_cls_locs.size()[1]
roi_cls_loc = roi_cls_locs[i]
roi_score = roi_scores[i]
gt_roi_loc = gt_roi_locs[i]
gt_roi_label = gt_roi_labels[i]
roi_cls_loc = roi_cls_loc.view(n_sample, -1, 4)
roi_loc = roi_cls_loc[torch.arange(0, n_sample), gt_roi_label]
我想要达到的结果
不知道最后两行如何理解，谢谢
