How to add noise to the first index in nump array

Question

How to add noise to the first index of array,with the number of iterations the max number iteration is 2.

I want to add a noise in range (1, max_iterator) to rows in order, for example:

add 0.788 to row 1st
add 1.233 to row 2nd
add 0.788 to row 3rd
add 1.233 to row 4th
and 0.788 to row 5th
and so on

I’ve tried to use this code but it doesn’t work

a = np.array([[858,833,123],
             [323,542,927],
             [938,361,271],
             [679,272,451]])
noise = np.random.normal(loc=0, scale=0.1, size=1)
max_iter = 2
for i in range(max_iter):
    for j in range(len(a)):
        a[i][0] = np.clip(a[i][0] + noise, a_min=0.0, a_max=None)
print(a)

Asked By: stack offer

||

Source

Answer 1

I can’t tell from your code or question if this is what you want or not:

a = np.array([[858,833,123],
             [323,542,927],
             [938,361,271],
             [679,272,451]])
a[:, 0] += np.random.normal(loc=0, scale=0.1, size=len(a))
a[:, 0] = np.clip(a[:, 0], a_min=0.0, a_max=None)

This adds noise to the first column

Answered By: Robin De Schepper

Answer 2

You can do:

# you'll need to cast the array as floats rather ints to add other floats
a = np.array([[858,833,123],
             [323,542,927],
             [938,361,271],
             [679,272,451]]).astype(float)

n = 2

# create n random values
noise = np.random.normal(loc=0, scale=0.1, size=n)

# loop through rows in steps of n
for i in range(0, a.shape[0], n):
    a[i:i + n, 0] = np.clip(a[i:i + 1, 0] + noise, a_min=0.0, a_max=None)

Answered By: Matt Pitkin

How to add noise to the first index in nump array

Question:

Answers: