CODEWITHSUNDEEP
pythonmatrixmatrix-multiplication
Macter
Macter

Reputation: 132

Multiply rows of matrix by particular columns in another matrix

I have two matrices of the following dimensions:

matrix(W): (7, 15)
matrix(X): (268, 7)

I want to multiply each row in X by the first column of W, and then sum up the result. I then want to do this for all columns in W, so that I will have 15 values at the end. The 15 values will be the sum of each of the 268 rows multiplied by a particular column in W.

To be clear:

z1 = (X1 * Wcol1) + (X2 * Wcol1) + .... + (X268 * Wcol1)
z2 = (X1 * Wcol2) + (X2 * Wcol2) + .... + (X268 * Wcol2)
...
z15 = (X1 * Wcol15) + (X2 * Wcol15) + .... + (X268 * Wcol15)

I am currently working with the following:

samples=268
for j in range(samples):
    zs = np.array([])
    z = X[j,:] * w[:,0]
    zs = np.append(zs, z)
    print(zs)

I believe this returns the 268 components of the first "z1" (which then need to be summed). I am having trouble generalising this, to do it for all the 15 columns.

Upvotes: 0

Views: 1095

Answers (2)

Z3D__
Z3D__

Reputation: 71

It sounds like you would like to calculate the sum of the columns of the matrix product X*W and store the values in an array. Assuming X and W are of class matrix, you can achieve this directly as follows:

Z=np.array(sum(X*W))

Upvotes: 2

Siladittya
Siladittya

Reputation: 1195

import numpy as np

Then take input for the matrices X and W

xmultw = np.matmul(X,W)
res = np.sum(xmultw,axis=0)

Hope this solves your problem

Upvotes: 1

Related Questions