Matching random seed in R for code coming from Python - python

I am given some python codes which depends on random numbers. This python code use a random_seed = 300.
Now I am trying to replicate this Python code in R. To make sure that the replication is perfect, I need to compare the end results between R and Python. Given that, the code depends on the random numbers, is there any way to know the equivalent random seed to be used in R?
I had a look into Creating same random number sequence in Python, NumPy and R, but it appears to be opposite way implementation i.e. from R to Python.
There is also a R library called reticulate where I could run python code in R, but could not figure out if I could fetch the R-equivalent random seed using this library
Any pointer will be very helpful.
Many thanks,

Related

python random number why isn't it same on all computers

One friend that I don't have any communication with anymore once told me the following:
Using that library(python's random) you can select a seed If you give it at seed it
means the randomly generated number will always be the same No matter
what computer you run it on
So I tried to test this, because this is what I need so it's the same on all computers and everytime someone calls this(this is important, as I am working on a blockchain NFT and trust is important here)
So I found this: https://machinelearningmastery.com/how-to-generate-random-numbers-in-python/
on that link, there's example:
from random import seed
from random import random
# seed random number generator
seed(1)
# generate some random numbers
print(random(), random(), random())
# reset the seed
seed(1)
# generate some random numbers
print(random(), random(), random())
running above in playground of python, I get
(0.417022004703, 0.720324493442, 0.000114374817345) (0.417022004703,
0.720324493442, 0.000114374817345)
But as you can see, on that website, the creator of that post got the following:
0.13436424411240122 0.8474337369372327 0.763774618976614
0.13436424411240122 0.8474337369372327 0.763774618976614
Why aren't they same on all computers then ? I am using the same seed. and how can I ensure that they will be the same ?
The ANSWER is that, even though you have tagged Python-3.x, you are actually using Python 2, and the random number algorithm changed between 2 and 3.
I can tell that because your print statement printed the values as a tuple, with parentheses. That wouldn't happen with Python 3's print function.
It's odd to rely on a particular implementation of a random number algorithm for financial purposes. If you really need reproducibility, then you should embed your own algorithm. There are several RNGs that are not difficult to code. But if the algorithm needs to be predictable, why not just use incrementing numbers? If you don't need randomness, then don't use randomness.
I get the post author's values both in Python 2.3.0 (over 18 years old) and in Python 3.10 (the newest, just a month old).
I get your values if I use numpy.random instead of Python's random.
So I suspect you're not telling the truth about your code or that that "playground of python" you're using is weird.

Can someone tell me the internal working of random function in python?

I am using python for more than 1 year. One question strike in my mind, how the random values are generated, there is some specific mechanism through which particular value should be selected by system that appears to be random to user. In all programming language how it is generated? Does all have the same mechanism?
Usually the OS provides a source of (pseudo) random data. For example on Linux there are /dev/urandom and /dev/random.
Some random functions in programming languages use that, other are based on a seed value and generate more or less reproducible values from it.
The random module in Python used a seeded approach. For use-cases that require more randomness use the secrets module. I uses the OS random sources.

Replicate a T-SQL random function in Python

I have a T-SQL code and I want to run a few simulations in Python. There is a code that includes random functions and I am not sure how I can replicate it.
When I have RAND() in SQL, I just use this in Python:
import random as random
print random.random()
But, I have also this code: RAND(CHECKSUM(NEWID()))
I guess, it is used for some kind of seed in the RAND function. But, how I can replicate the same thing in Python to have as much closer results I could?
in Python you need to first call
random.seed() in you program (once only)
then
random.random()
each time you want a random number
to give a pseudo random number from 0 to 1
yes RAND(CHECKSUM(NEWID())) appears to be a good trick to get a random value to use as a seed - it could be argued that a NEWID has a greater amount of randomness than using time as a seed. It depends on your application, time is considered insufficient as a seed for cryptography without adding other entropy - the python randomize uses time as a seed

Compare Numpy and Matlab code that uses random permutation

I'm having problems to compare the output of two code because of random number state.
I'm comparing the MATLAB randperm function with the output of the equivalent numpy.random.permutation function but, even if I've set the seed to the same value with a MATLAB rand('twister',0) and a python numpy.random.seed(0) I'm obtaining different permutations.
I've to say that the result of MATLAB's rand and numpy numpy.random.rand are the same if the seed are set like above.
This is a common issue. While the random number generator is identical, the function which converts your random number stream into a random permutation is different. There is no specified standard algorithm which describes the expected result.
To solve this issue, you have to use the same library in both tools.

python (SimPy) generate random numbers that follow the erlang distribution

I am using Python (SimPy package mostly, but it is irrelevant to the question I think), modeling some systems and running simulations. For this purpose I need to produce random numbers that follow distributions. I have done alright so far with some distributions like exponential and normal by importing the random (eg from random import *) and using the expovariate or normalvariate methods. However I cannot find any method in random that produce numbers that follow the Erlang distribution. So:
Is there some method that I overlooked?
Do I have to import some other library?
Can I make some workaround? (In think that I can use the Exponential distribution to produce random “Erlang” numbers but I am not sure how. A piece of code might help me.
Thank you in advance!
Erlang distribution is a special case of the gamma distribution, which exists as numpy.random.gamma (reference). Just use an integer value for the k ("shape") argument. See also about scipy.stats.gamma for functions with the PDF, CDF etc.
As the previous answer stated, the erlang distribution is a special case of the gamma distribution. As far as I know, you do not, however, need the numpy package. Random numbers from a gamma distribution can be generated in python using random.gammavariate(alpha, beta).
Usage:
import random
print random.gammavariate(3,1)

Categories

Resources