591

randrange(start, stop) only takes integer arguments. So how would I get a random number between two float values?

martineau
  • 112,593
  • 23
  • 157
  • 280
Mantis Toboggan
  • 6,505
  • 3
  • 18
  • 10
  • 8
    If you wanted numpy it's `np.random.uniform(start, stop)` or `np.random.uniform(start, stop, samples)` if you wanted multiple samples. Otherwise below answers are best. – sachinruk Dec 04 '19 at 00:33

5 Answers5

882

Use random.uniform(a, b):

>>> random.uniform(1.5, 1.9)
1.8733202628557872
Yuri Stuken
  • 11,890
  • 1
  • 25
  • 23
  • 8
    could this theoretically produce 1.5 and 1.9? or would it only ever produce 1.50~1 and 1.89~? – Musixauce3000 Apr 29 '16 at 13:39
  • 28
    @Musixauce3000 Short Answer: Yes. Longer answer: If you look at the documentation it states `Returns a random floating point number N such that a <= N <= b for a <= b and b <= N <= a for b < a` In other words the output `N` can equal either input `a` and `b`. In this case `1.5` and `1.9`. – Dan Mar 01 '17 at 13:37
  • Is there another way to do this without using the `.uniform` function, but instead with either `.random` or `randrange`? – EnigmaTech Mar 27 '19 at 10:12
  • 1
    @DerryckDX `1.5 + random.random() * (1.9 - 1.5)` should do it, even though according to the specs this will never return exactly `1.9` (even in theory). – Yonatan N Mar 29 '19 at 00:03
  • 2
    @Musixauce3000 it seems `uniform(a, b)` is implemented as `a + (b-a) * random()` and returns *a random number in the range [a, b) or [a, b] depending on rounding* https://github.com/python/cpython/blob/963eb0f4738456455b9bef7eb531b46805415208/Lib/random.py#L415 – Pavel Jun 04 '19 at 08:47
124

if you want generate a random float with N digits to the right of point, you can make this : 

round(random.uniform(1,2), N)

the second argument is the number of decimals.

Baurin Leza
  • 1,796
  • 1
  • 12
  • 14
86

random.uniform(a, b) appears to be what your looking for. From the docs:

Return a random floating point number N such that a <= N <= b for a <= b and b <= N <= a for b < a.

See here.

tzot
  • 87,612
  • 28
  • 135
  • 198
garnertb
  • 9,250
  • 33
  • 38
14

From my experience dealing with python, I can only say that the random function can help in generating random float numbers. Take the example below;

import random

# Random float number between range 15.5 to 80.5
print(random.uniform(15.5, 80.5))

# between 10 and 100
print(random.uniform(10, 100))
The random.uniform() function returns a random floating-point number between a given range in Python

The two sets of code generates random float numbers. You can try experimenting with it to give you what you want.

DrosnickX
  • 387
  • 4
  • 9
11

Most commonly, you'd use:

import random
random.uniform(a, b) # range [a, b) or [a, b] depending on floating-point rounding

Python provides other distributions if you need.

If you have numpy imported already, you can used its equivalent:

import numpy as np
np.random.uniform(a, b) # range [a, b)

Again, if you need another distribution, numpy provides the same distributions as python, as well as many additional ones.

stwykd
  • 2,365
  • 2
  • 15
  • 19