CODEWITHSUNDEEP
arrayspython-3.xnumpy-ndarray
turaran32
turaran32

Reputation: 93

change elementes of array

I have an array that I have saved it in the MC variable:

with open(PATH_FICHERO1, newline='',  encoding='latin-1') as csvfile:
    data = list(csv.reader(csvfile))
MC=numpy.array([numpy.array(datai) for datai in data])
type(MC)
print(MC)

enter image description here

In the case that there is a letter B I want to change the value for a random value that is between 1/3 and 1. Applying the following code does not give me an error, nevertheless when viewing the matrix the random values are always the same .

B= np.random.uniform(1/3,2/3)
MC = np.where( MC == 'B',B , MC)
print(B)
print(MC)

The values should be the same, in the case of B it is generated randomly but then I don't know why it is not applied in the matrix.

enter image description here

Upvotes: 1

Views: 54

Answers (2)

singh
singh

Reputation: 79

It looks like the problem is in code.

B= np.random.uniform(1/3,2/3)
MC = np.where( MC == 'B',B , MC)

Above you have assigned the random value to variable B and then mapping the same value to each element in the matrix. e.g. if the random value is 0.55 assigned to variable B then the same value will be mapped to each character in the matrix, that's why you are seeing the same value everywhere.

Solution :-

import numpy as np
MC = [['1', 'B','B','B','B'],['B', '1','B','B','B'],['B', 'B','1','B','B'],['B', 'B','B','1','B'],['B', 'B','B','B','1']]

for i in range(len(MC)):
  for j in range(len(MC[i])):
    if MC[i][j]== "B":
      MC[i][j]=str(np.random.uniform(1/3,2/3))

Result:

[['1',
  '0.4133872881475391',
  '0.388606739011768',
  '0.632669589057151',
  '0.555652072409681'],
 ['0.5964448593391674',
  '1',
  '0.5916592530420399',
  '0.6004173979328635',
  '0.45989982907152327'],
 ['0.49491629056763076',
  '0.5814889531351302',
  '1',
  '0.42177756298000196',
  '0.44166060842279004'],
 ['0.43908913157941704',
  '0.5630605854105761',
  '0.5867510079174598',
  '1',
  '0.5518789342365693'],
 ['0.6008453194037271',
  '0.5921925074432766',
  '0.344618130286966',
  '0.5767303673801853',
  '1']]

Upvotes: 1

DYZ
DYZ

Reputation: 57033

By default, np.random.uniform generates ONE random variable. You must pass the shape of the output array if you want an array of random variables:

B = np.random.uniform(1/3, 2/3, MC.shape)

Upvotes: 2

Related Questions