ResourceExhaustedError (see above for traceback): OOM when allocating tensor

[HuangCongQing]

https://blog.csdn.net/lics999/article/details/78517638

[HuangCongQing]

https://blog.csdn.net/feifei884431/article/details/51429829

[HuangCongQing]

https://blog.csdn.net/muyiyushan/article/details/61201380

[HuangCongQing]

ResourceExhaustedError (see above for traceback): OOM when allocating tensor with shape[10,256,400,528]
[10,256,400,528]的第一个参数表示batch_size的大小，第二个参数表示某层卷积核的个数，第三个参数表示图像的高，第四个参数表示图像的长
这里出现这种错误的原因时超出内存了，因此可以适当减小batch_size的大小即可解决