Set values in pandas series to a random value based on condition

I have a pandas series, and I’d like to set all the values that are 0 to a random number between 100 and 1000. Is there a direct way of doing that? I’ve only seen a roundabout way like so:

nums = np.random.choice(range(100, 1100), size = 924)

df.loc[df['Target']==0] = nums

The error I get back here is "Must have equal len keys and value when setting with an iterable".

Size is 924 because I know that’s how many values I want to change.

>Solution :

you can use the .loc accessor with a boolean mask to select the rows where the value is 0, and then assign a random value between 100 and 1000 to those rows.

import pandas as pd
import numpy as np

# create a sample series
s = pd.Series([1, 2, 3, 0, 4, 0, 5, 6, 0])

# create a boolean mask to select the rows where the value is 0
mask = s == 0

# set the values to a random number between 100 and 1000
s.loc[mask] = np.random.randint(100, 1000, size=mask.sum())

print(s)

output –

0      1
1      2
2      3
3    784
4      4
5    208
6      5
7      6
8    923