如何在Keras中计算矢量明智的点积? [英] How to calculate vector-wise dot product in Keras?

问题描述

我希望在Keras中计算矢量明智的点积. 详细地说，我的意思是，如果我有两个张量A和B，它们的形状均为(None, 30, 100)，我想计算形状为(None, 30, 1)的结果C，它将满足

I hope to calculate a vector wise dot product in Keras. In detail, I mean if I have two tensor A and B, both with shape (None, 30, 100), I want to calculate the result C with shape (None, 30, 1) which would satisfy

C[:,:,i] = dot(A[:,:,i], B[:,:,i]).

我想知道是否可行，因为后端中的batch_dot()函数将仅返回形状(None, 30, 30)并且仅具有关系

I wonder if that is possible, since the batch_dot() function in the backend would only return the shape (None, 30, 30) and only have the relationship

C[:,i,j] = dot(A[:,:,i], B[:,:,j])

但这不是我想要的.

谢谢！

推荐答案

您可以尝试以下方法:

import keras.backend as K

C = K.sum(A * B,axis=-1,keepdims=True)

这篇关于如何在Keras中计算矢量明智的点积?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！