CODEWITHSUNDEEP
pythonnumpymatrix
Johnny Metz
Johnny Metz

Reputation: 5965

python numpy: calculate across row of a matrix

I'm trying to calculate across the row of a matrix and store that value in a different matrix. Is this the most efficient way to do this or are there any built in functions I should be aware of.

import numpy as np
a = np.array([ [1, 2, 3], [4, 5, 6], [7, 8, 9] ])

def calc_across(matrix):
    frame = []
    for row in matrix:
        frame.append( [row[0] * row[1]/2. * row[2]/3] )  # period present to generate floats
    return np.array(frame)

b = calc_across(a)

If I do print b I get the following matrix:

b = [ [1.], [20.], [84.] ]

If a is 3x3, b must be 3x1 (3 rows, 1 column). If a is 10x3, b must be 10x1, etc.

Upvotes: 0

Views: 915

Answers (1)

Abdou
Abdou

Reputation: 13274

Try:

b = np.prod(a / [1.0,2.0,3.0],axis=1, keepdims=True)
b

# # array([[  1.],
       # [ 20.],
       # [ 84.]])

I hope this helps.

Upvotes: 2

Related Questions