[英]How to reverse batch normalization
In a regression network, I would like to use batch normalization on the objective y
to obtain y_norm
to fit.在回归网络中,我想对目标
y
使用批量归一化以获得适合的y_norm
。 Because y_norm
is well distributed.因为
y_norm
分布良好。
In testing stage after training, I need to "undo" a batch normalization on the predicted y_norm
.在训练后的测试阶段,我需要“撤消”对预测的
y_norm
的批量归一化。 Is there any elegant way in tensorflow/keras in which I can construct an "undo" layer from the origin BN layer?在 tensorflow/keras 中有什么优雅的方法可以从原始 BN 层构造一个“撤消”层吗?
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.