How To Set Random Seed In Numpy

What Does random.seed Practice in NumPy

Agreement how to create reproducible results when generating pseudo-random constructs with NumPy in Python

Photograph by Andrew Seaman on Unsplash

Introduction

Randomness is a central mathematical concept that is usually used in the context of programming as well. Sometimes, nosotros may demand to introduce some randomness when creating some toy information or when we demand to perform some specific calculations that will be dependent on some random event.

In today's commodity, nosotros are first going to hash out near the concepts of pseudo random numbers and true randomness. Additionally, we will hash out most numpy's random.seed and how to use it in order to create reproducible results. Finally, we will showcase how to ensure that the same seed is sustained throughout the code.

Understanding Pseudorandomness

A sequence of pseudo random numbers is one that has been generated using a deterministic procedure but appears to be statistically random.

Even though the properties of sequences generated by pseudo random generators (also known equally Deterministic Random Bit Generators) approximate the backdrop of random number sequences, in reality they are not truly random. This is considering the generated sequence is determined by an initial value which is known as the seed.

You can view seed as the bodily starting point of the sequence. Pseudo-random number generators generate each number based on some processes and operations performed on the previously generated value. Since the outset value to exist generated has no previous value on which the generated tin can perform these operations, the seed acts as the "previous value" for the beginning number to be generated.

Creating reproducible results using random.seed

In the aforementioned way, NumPy's random number routines generate sequences of pseudo random numbers. This is achieved by creating a sequence with the employ of BitGenerators (objects that generate random numbers) and Generators that make use of the created sequences to sample from dissimilar probability distributions such every bit Normal, Uniform or Binomial.

Now in guild to generate reproducible sequences of pseudo random numbers, the BitGenerator object accepts a seed that is used to set the initial state. This can be achieved past setting numpy.random.seed as shown below:

                import numpy as np                                  np.random.seed(123)                              

Creating reproducible results is a common requirement in different apply cases. For case, when testing some slice of functionality, you may need to create reproducible results past configuring the seed to a specific value so that the generated results can exist compared against the expected results.

Additionally, the creation of reproducible results is common in the wider field of enquiry. For instance, if you work with a model that uses randomness (a random forest for example) and want to publish the results (say in a paper) then you may want (and probably have) to ensure that other people and users tin reproduce the results you present.

Local Effect of seed

It is also important to mention that the random seed in NumPy too affects other methods, such as numpy.random.permutation and information technology likewise has a local consequence.

This means that if yous specify numpy.random.seed only once but call numpy.random.permutation multiple times, the results that you'll get won't be identical (since they won't depend on the aforementioned seed). To showcase the trouble, permit's consider the post-obit code:

                import numpy as np                np.random.seed(123)                print(np.random.permutation(10))
                                      array([4, 0, seven, 5, 8, 3, one, 6, 9, 2])                                                  print(np.random.permutation(10))
                                      assortment([3, five, 4, 2, 8, vii, 6, 9, 0, ane])                                                

As you tin can run across, the results are not reproducible fifty-fifty though we have ready the random seed. This happens because random.seed has only 'local consequence'. In social club to reproduce the results, you'd have to specify the same random seed just before every call of np.random.permutation as shown below.

                import numpy as np                np.random.seed(123)
print(np.random.permutation(10))
                                      array([iv, 0, vii, v, 8, 3, ane, 6, 9, 2])                                                  np.random.seed(123)
print(np.random.permutation(10))
                                      array([iv, 0, vii, 5, eight, 3, 1, 6, 9, two])                                                

Final Thoughts

In today'south commodity nosotros discussed nigh the concepts of true or pseudo randomness and purpose of random.seed in NumPy and Python. Additionally, we showcased how to create reproducible results every fourth dimension we execute the same piece of code, even when the results are dependent on some (pseudo)randomness. Finally, we explored how to ensure that the issue of random seed will be sustained throughout the lawmaking when this is required.

Become a member and read every story on Medium. Your membership fee direct supports me and other writers you read.

Source: https://towardsdatascience.com/random-seed-numpy-786cf7876a5f

Share this post