theano:按类标签求和 [英] theano: summation by class label

问题描述

我有一个矩阵，表示到一组点的k个近邻的距离， 并且有一个最近邻居的类别标签矩阵. (都是N-k矩阵)

I have a matrix which represents a distances to the k-nearest neighbour of a set of points, and there is a matrix of class labels of the nearest neighbours. (both N-by-k matrix)

在theano中构建(N-by-#classes)矩阵的最佳方法是什么，该矩阵的(i，j)元素将是从第i个点到其带有类标签的k-NN个点的距离之和'j'?

What is the best way in theano to build a (N-by-#classes) matrix whose (i,j) element will be the sum of distances from i-th point to its k-NN points with the class label 'j'?

示例:

# N = 2
# k = 5
# number of classes = 3

K_val  = [[1,2,3,4,6],
          [2,4,5,5,7]]

l_val  = [[0,1,2,0,1],
          [2,0,1,2,0]]

result = [[5,8,3],
          [11,5,7]]

这个任务在theano吗?

this task in theano?

K = theano.tensor.matrix()
l = theano.tensor.matrix()
result = <..some code..>

f = theano.function(inputs=[K,l], outputs=result)

推荐答案

看看这个仓库可能会很有趣: https://github.com/erogol/KLP_KMEANS/blob/master/klp_kmeans.py

You might be interesting in having a look to this repo: https://github.com/erogol/KLP_KMEANS/blob/master/klp_kmeans.py

是使用theano(func kpl_kmeans)的K-Means实现.我相信您想要的是函数find_bmu中使用的矩阵W.

Is a K-Means implementation using theano (func kpl_kmeans). I believe what you want is the matrix W used in the function find_bmu.

希望您发现它有用.

这篇关于theano:按类标签求和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！