YuhangJi

Results 3 issues of YuhangJi

i made a reimplement with tensorflow2, but it is not very well. In MIT dataset, the model loss value just decline to 2.3.

测试集图像没有框,训练集图像有框都是乱的,请问这个原因可能是什么

作者你好,训练一轮需要四十分钟左右这个正常吗?另外跑得过程中提示我RuntimeWarning: overflow encountered in exp decode_bbox_height = np.exp(mbox_loc[:, 3] /4)。