I have a big script in Python. I inspired myself in other people’s code so I ended up using the numpy.random module for some things (for example for creating an array of random numbers taken from a binomial distribution) and in other places I use the module random.random.

Can someone please tell me the major differences between the two? Looking at the doc webpage for each of the two it seems to me that numpy.random just has more methods, but I am unclear about how the generation of the random numbers is different.

The reason why I am asking is because I need to seed my main program for debugging purposes. But it doesn’t work unless I use the same random number generator in all the modules that I am importing, is this correct?

Also, I read here, in another post, a discussion about NOT using numpy.random.seed(), but I didn’t really understand why this was such a bad idea. I would really appreciate if someone explain me why this is the case.

You have made many correct observations already!

Unless you’d like to seed both of the random generators, it’s probably simpler in the long run to choose one generator or the other. But if you do need to use both, then yes, you’ll also need to seed them both, because they generate random numbers independently of each other.

For numpy.random.seed(), the main difficulty is that it is not thread-safe – that is, it’s not safe to use if you have many different threads of execution, because it’s not guaranteed to work if two different threads are executing the function at the same time. If you’re not using threads, and if you can reasonably expect that you won’t need to rewrite your program this way in the future, numpy.random.seed() should be fine. If there’s any reason to suspect that you may need threads in the future, it’s much safer in the long run to do as suggested, and to make a local instance of the numpy.random.Random class. As far as I can tell, random.random.seed() is thread-safe (or at least, I haven’t found any evidence to the contrary).

The numpy.random library contains a few extra probability distributions commonly used in scientific research, as well as a couple of convenience functions for generating arrays of random data. The random.random library is a little more lightweight, and should be fine if you’re not doing scientific research or other kinds of work in statistics.

Otherwise, they both use the Mersenne twister sequence to generate their random numbers, and they’re both completely deterministic – that is, if you know a few key bits of information, it’s possible to predict with absolute certainty what number will come next. For this reason, neither numpy.random nor random.random is suitable for any serious cryptographic uses. But because the sequence is so very very long, both are fine for generating random numbers in cases where you aren’t worried about people trying to reverse-engineer your data. This is also the reason for the necessity to seed the random value – if you start in the same place each time, you’ll always get the same sequence of random numbers!

As a side note, if you do need cryptographic level randomness, you should use the secrets module, or something like Crypto.Random if you’re using a Python version earlier than Python 3.6.

From Python for Data Analysis, the module numpy.random supplements the Python random with functions for efficiently generating whole arrays of sample values from many kinds of probability distributions.

By contrast, Python’s built-in random module only samples one value at a time, while numpy.random can generate very large sample faster. Using IPython magic function %timeit one can see which module performs faster:

In [1]: from random import normalvariate
In [2]: N = 1000000

In [3]: %timeit samples = [normalvariate(0, 1) for _ in xrange(N)]
1 loop, best of 3: 963 ms per loop

In [4]: %timeit np.random.normal(size=N)
10 loops, best of 3: 38.5 ms per loop

The source of the seed and the distribution profile used are going to affect the outputs – if you are looking for cryptgraphic randomness, seeding from os.urandom() will get nearly real random bytes from device chatter (ie ethernet or disk) (ie /dev/random on BSD)

this will avoid you giving a seed and so generating determinisitic random numbers. However the random calls then allow you to fit the numbers to a distribution (what I call scientific random ness – eventually all you want is a bell curve distribution of random numbers, numpy is best at delviering this.

SO yes, stick with one generator, but decide what random you want – random, but defitniely from a distrubtuion curve, or as random as you can get without a quantum device.

I am a bit confused on what random.seed() does in Python. For example, why does the below trials do what they do (consistently)?

>>> import random
>>> random.seed(9001)
>>> random.randint(1, 10)
1
>>> random.randint(1, 10)
3
>>> random.randint(1, 10)
6
>>> random.randint(1, 10)
6
>>> random.randint(1, 10)
7

I couldn’t find good documentation on this.

Pseudo-random number generators work by performing some operation on a value. Generally this value is the previous number generated by the generator. However, the first time you use the generator, there is no previous value.

Seeding a pseudo-random number generator gives it its first “previous” value. Each seed value will correspond to a sequence of generated values for a given random number generator. That is, if you provide the same seed twice, you get the same sequence of numbers twice.

Generally, you want to seed your random number generator with some value that will change each execution of the program. For instance, the current time is a frequently-used seed. The reason why this doesn’t happen automatically is so that if you want, you can provide a specific seed to get a known sequence of numbers.

All the other answers don’t seem to explain the use of random.seed(). Here is a simple example (source):

import random
random.seed( 3 )
print "Random number with seed 3 : ", random.random() #will generate a random number 
#if you want to use the same random number once again in your program
random.seed( 3 )
random.random()   # same random number as before

>>> random.seed(9001)   
>>> random.randint(1, 10)  
1     
>>> random.seed(9001)     
>>> random.randint(1, 10)    
1           
>>> random.seed(9001)          
>>> random.randint(1, 10)                 
1                  
>>> random.seed(9001)         
>>> random.randint(1, 10)          
1     
>>> random.seed(9002)                
>>> random.randint(1, 10)             
3

You try this.

Let’s say ‘random.seed’ gives a value to random value generator (‘random.randint()’) which generates these values on the basis of this seed. One of the must properties of random numbers is that they should be reproducible. When you put same seed, you get the same pattern of random numbers. This way you are generating them right from the start. You give a different seed- it starts with a different initial (above 3).

Given a seed, it will generate random numbers between 1 and 10 one after another. So you assume one set of numbers for one seed value.

A random number is generated by some operation on previous value.

If there is no previous value then the current time is taken as previous value automatically. We can provide this previous value by own using random.seed(x) where x could be any number or string etc.

Hence random.random() is not actually perfect random number, it could be predicted via random.seed(x).

import random 
random.seed(45)            #seed=45  
random.random()            #1st rand value=0.2718754143840908
0.2718754143840908  
random.random()            #2nd rand value=0.48802820785090784
0.48802820785090784  
random.seed(45)            # again reasign seed=45  
random.random()
0.2718754143840908         #matching with 1st rand value  
random.random()
0.48802820785090784        #matching with 2nd rand value

Hence, generating a random number is not actually random, because it runs on algorithms. Algorithms always give the same output based on the same input. This means, it depends on the value of the seed. So, in order to make it more random, time is automatically assigned to seed().

Seed() can be used for later use ---

Example:
>>> import numpy as np
>>> np.random.seed(12)
>>> np.random.rand(4)
array([0.15416284, 0.7400497 , 0.26331502, 0.53373939])
>>>
>>>
>>> np.random.seed(10)
>>> np.random.rand(4)
array([0.77132064, 0.02075195, 0.63364823, 0.74880388])
>>>
>>>
>>> np.random.seed(12) # When you use same seed as before you will get same random output as before
>>> np.random.rand(4)
array([0.15416284, 0.7400497 , 0.26331502, 0.53373939])
>>>
>>>
>>> np.random.seed(10)
>>> np.random.rand(4)
array([0.77132064, 0.02075195, 0.63364823, 0.74880388])
>>>

# Simple Python program to understand random.seed() importance

import random

random.seed(10)

for i in range(5):    
    print(random.randint(1, 100))

Execute the above program multiple times…

1st attempt: prints 5 random integers in the range of 1 – 100

2nd attempt: prints same 5 random numbers appeared in the above execution.

3rd attempt: same

…..So on

Explanation: Every time we are running the above program we are setting seed to 10, then random generator takes this as a reference variable. And then by doing some predefined formula, it generates a random number.

Hence setting seed to 10 in the next execution again sets reference number to 10 and again the same behavior starts…

As soon as we reset the seed value it gives the same plants.

Note: Change the seed value and run the program, you’ll see a different random sequence than the previous one.

In this case, random is actually pseudo-random. Given a seed, it will generate numbers with an equal distribution. But with the same seed, it will generate the same number sequence every time. If you want it to change, you’ll have to change your seed. A lot of people like to generate a seed based on the current time or something.

Imho, it is used to generate same random course result when you use random.seed(samedigit) again.

In [47]: random.randint(7,10)

Out[47]: 9


In [48]: random.randint(7,10)

Out[48]: 9


In [49]: random.randint(7,10)

Out[49]: 7


In [50]: random.randint(7,10)

Out[50]: 10


In [51]: random.seed(5)


In [52]: random.randint(7,10)

Out[52]: 9


In [53]: random.seed(5)


In [54]: random.randint(7,10)

Out[54]: 9

Set the seed(x) before generating a set of random numbers and use the same seed to generate the same set of random numbers. Useful in case of reproducing the issues.

>>> from random import *
>>> seed(20)
>>> randint(1,100)
93
>>> randint(1,100)
88
>>> randint(1,100)
99
>>> seed(20)
>>> randint(1,100)
93
>>> randint(1,100)
88
>>> randint(1,100)
99
>>> 

Here is my understanding. Every time we set a seed value, a “label” or ” reference” is generated. The next random.function call is attached to this “label”, so next time you call the same seed value and random.function, it will give you the same result.

np.random.seed( 3 )
print(np.random.randn()) # output: 1.7886284734303186

np.random.seed( 3 )
print(np.random.rand()) # different function. output: 0.5507979025745755

np.random.seed( 5 )
print(np.random.rand()) # different seed value. output: 0.22199317108973948

Here is a small test that demonstrates that feeding the seed() method with the same argument will cause the same pseudo-random result:

# testing random.seed()

import random

def equalityCheck(l):
    state=None
    x=l[0]
    for i in l:
        if i!=x:
            state=False
            break
        else:
            state=True
    return state


l=[]

for i in range(1000):
    random.seed(10)
    l.append(random.random())

print "All elements in l are equal?",equalityCheck(l)

random.seed(a, version) in python is used to initialize the pseudo-random number generator (PRNG).

PRNG is algorithm that generates sequence of numbers approximating the properties of random numbers. These random numbers can be reproduced using the seed value. So, if you provide seed value, PRNG starts from an arbitrary starting state using a seed.

Argument a is the seed value. If the a value is None, then by default, current system time is used.

and version is An integer specifying how to convert the a parameter into a integer. Default value is 2.

import random
random.seed(9001)
random.randint(1, 10) #this gives output of 1
# 1

If you want the same random number to be reproduced then provide the same seed again

random.seed(9001)
random.randint(1, 10) # this will give the same output of 1
# 1

If you don’t provide the seed, then it generate different number and not 1 as before

random.randint(1, 10) # this gives 7 without providing seed
# 7

If you provide different seed than before, then it will give you a different random number

random.seed(9002)
random.randint(1, 10) # this gives you 5 not 1
# 5

So, in summary, if you want the same random number to be reproduced, provide the seed. Specifically, the same seed.