InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error: Assign requires shapes of both tensors to match. lhs shape= [1,1,1024,1024] rhs shape= [1,1,4096,4096] [[node save/Assign_29 (defined at E:\Infrared_Deeplearning\SSD-AbsoluteCoord\utility\scaffolds.py:86) ]]

[navy63]

InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error: Assign requires shapes of both tensors to match. lhs shape= [1,1,1024,1024] rhs shape= [1,1,4096,4096] [[node save/Assign_29 (defined at E:\Infrared_Deeplearning\SSD-AbsoluteCoord\utility\scaffolds.py:86) ]]

[navy63]

我知道原因了，这个版本SSD和balancap版本的SSD使用ckpt不一样，你这个版本的不用改什么东西，可以直接测试，所有的东西国内可以直接在百度云下载，你提供了链接，非常感谢，balancap版本的loss不会下降，一直在振荡，mAP也非常小，期待你这个效果

[chenjjjjuu]

我知道原因了，这个版本SSD和balancap版本的SSD使用ckpt不一样，你这个版本的不用改什么东西，可以直接测试，所有的东西国内可以直接在百度云下载，你提供了链接，非常感谢，balancap版本的loss不会下降，一直在振荡，mAP也非常小，期待你这个效果
啥意思？我改了模型然后eval的时候也会出现这个问题，是因为不能改网络模型吗