numpy 3D点积
[英]numpy 3D dot product
问题描述
I have two 3dim numpy matrices and I want to do a dot product according to one axis without using a loop: 
我有两个3dim numpy矩阵，我想根据一个轴做一个点积而不使用循环：
a=[ [[ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0]],
[[ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0]],
[ [ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0]],
[ [ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0]],
[[ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0]],
[[ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0.]],
[[ 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0],
[ 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0],
[ 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1],
[ 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0]]]
b=[[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]],
[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]],
[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]],
[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]],
[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]],
[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]],
[[ 0, 0, 1, 0, 0.],
[ 1, 0, 0, 0, 0.],
[ 0, 0, 0, 0, 0.],
[ 0, 1, 0, 0, 0.]]]
dt = np.dtype(np.float32)
a=np.asarray(a,dtype=dt)
b=np.asarray(b,dtype=dt)
print(a.shape)
print(b.shape)
a has the shape of (7, 4, 15) and b has the shape of (7, 4, 5). a具有（7，4，15）的形状，b具有（7，4，5）的形状。 I want the c=np.dot(a,b) be in the size of (7,5,15) as below: 我希望c = np.dot（a，b）的大小为（7,5,15），如下所示：
c = np.zeros((7,15,5))
for i in range(7):
c[i,:,:] = np.dot(a[i,:,:].T , b[i,:,:])
But I am looking for a solution without a for-loop. 但是我正在寻找没有for循环的解决方案。 something like: 就像是：
c = np.tensordot(a.reshape(4,7,5),b.reshape(7,4,15),axes=([1,0],[0,1]))
but this one doesn't work as expected. 但这并没有达到预期效果。
I also tried this: 我也试过这个：
newaxes_a=[2,0,1]
newaxes_b=[1,0,2]
newshape_a=(-1,28)
newshape_b=(28,-1)
a_t = a.transpose(newaxes_a).reshape(newshape_a)
b_t = b.transpose(newaxes_b).reshape(newshape_b)
c = np.dot(a_t, b_t)
which didn't work as expected. 这没有按预期工作。
Any ideas? 有任何想法吗？
1 个解决方案
解决方案1
4 已采纳 2015-11-30 16:47:19
You can use np.einsum - 您可以使用np.einsum
#to match the given example
c2 = np.einsum('ijk,ijl->ikl',a,b)
print np.allclose(c, c2)
Another one using broadcasting - 另一个使用broadcasting -
c = (a[:,:,None,:]*b[...,None]).sum(1)
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.