scikit-learn random state in splitting dataset

Question

Can anyone tell me why we set random state to zero in splitting train and test set.

X_train, X_test, y_train, y_test = \
    train_test_split(X, y, test_size=0.30, random_state=0)

I have seen situations like this where random state is set to 1!

X_train, X_test, y_train, y_test = \
    train_test_split(X, y, test_size=0.30, random_state=1)

What is the consequence of this random state in cross validation as well?

@Scott Hunter it comes from sklearn.cross_validation. But what is the effect of random state being zero and one on the the train and test split? — Shelly, Feb 13 '17 at 00:44
@Scott Hunter http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html — Shelly, Feb 13 '17 at 01:13
That doesn't say much about "random state" and its effect that's why i am asking this question — Shelly, Feb 13 '17 at 01:15
It's just to make sure that you obtain the same split everytime you run your script. Read up a bit on Pseudo-random-number-generators. (a number like 32525352 would have the same effect as 0 or 1; it's just a constant which is mapped to some internal state) If you don't do this, it's seeded based on time, resulting in different results in most of your runs. — sascha, Feb 13 '17 at 05:19
@Shelly But i have seen it come from sklearn.model_selection — Ganesh Chowdhary Sadanala, Mar 15 '19 at 17:03
Possible duplicate of [Random state (Pseudo-random number) in Scikit learn](https://stackoverflow.com/questions/28064634/random-state-pseudo-random-number-in-scikit-learn) — Ricky Geng, Apr 28 '19 at 04:36

score 68 · Answer 1 · answered Feb 13 '17 at 06:00

68

It doesn't matter if the random_state is 0 or 1 or any other integer. What matters is that it should be set the same value, if you want to validate your processing over multiple runs of the code. By the way I have seen random_state=42 used in many official examples of scikit as well as elsewhere also.

random_state as the name suggests, is used for initializing the internal random number generator, which will decide the splitting of data into train and test indices in your case. In the documentation, it is stated that:

If random_state is None or np.random, then a randomly-initialized RandomState object is returned.

If random_state is an integer, then it is used to seed a new RandomState object.

If random_state is a RandomState object, then it is passed through.

This is to check and validate the data when running the code multiple times. Setting random_state a fixed value will guarantee that same sequence of random numbers are generated each time you run the code. And unless there is some other randomness present in the process, the results produced will be same as always. This helps in verifying the output.

answered Feb 13 '17 at 06:00

Vivek Kumar

35,217
8
109
132

it's strange, every time I seem to rerun my classification metrics such as Specificity and Sensitivity, etc, I get a variation in my score despite the fact that I have a set_seed. Any idea why that might be? Is there anywhere else I need to set_seed aside from the train _test split, for example under .fit() or .score or .predict()? I don't believe I have any other sources of randomness anywhere. – bernando_vialli May 04 '18 at 14:49
1

@mathlover I also observed the same randomness in my output as well. All i found is that when you set some value to `random_state` then the output like `mean_absolute_error` in my case get fixed( i mean every time i run it.it outputs same) – 0xPrateek Aug 10 '19 at 13:34
when the value itself doesn't matter, why isn't it just a boolean? – Ben Sep 20 '19 at 13:27
@Ben Because internally the value supplied in `random_state` will act as a seed for pseudo-random number generator used in `numpy`. When its not set, then most implementations will use the current system time as the seed. So its not proper to just set it to boolean. – Vivek Kumar Sep 20 '19 at 13:42
1

Random seed is often set to 42 as "The Answer to the Ultimate Question of Life, the Universe, and Everything is 42" from The Hitchhiker's Guide to the Galaxy. But think most people know this. Just in case it is of interest see... https://en.wikipedia.org/wiki/Phrases_from_The_Hitchhiker%27s_Guide_to_the_Galaxy#Answer_to_the_Ultimate_Question_of_Life,_the_Universe,_and_Everything_(42) – Joey Jul 26 '20 at 05:01
If I have written a line of shuffled in code and next time I run the code , data is shuffled now , Should it again be splitted same if random state given ? . my question is regarding shuffling. – Hamza Mar 19 '21 at 06:03
@OfficialDeveloper, no it will change. The random state in case of train_test_split will guarantee that the indices in train and test are same in each run. But if the actual data points are shuffled, so the same indices will have different data – Vivek Kumar Mar 19 '21 at 09:05
Okay Means If we don't shuffle the data every time , then splits remains same – Hamza Mar 19 '21 at 09:08
so is this thing has the same use as the k fold validation, the difference is that u dont split the data into 10 parts (fold = 10) and split them to training and testing data? – Anggia Bagaskara Sep 13 '21 at 20:34

score 18 · Answer 2 · answered Dec 06 '19 at 10:52

when random_state set to an integer, train_test_split will return same results for each execution.

when random_state set to an None, train_test_split will return different results for each execution.

see below example:

from sklearn.model_selection import train_test_split

X_data = range(10)
y_data = range(10)

for i in range(5):
    X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size = 0.3,random_state = 0) # zero or any other integer
    print(y_test)

print("*"*30)

for i in range(5): 
    X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size = 0.3,random_state = None)
    print(y_test)

Output:

[2, 8, 4]

[4, 7, 6]

[4, 3, 7]

[8, 1, 4]

[9, 5, 8]

[6, 4, 5]

score 14 · Answer 3 · answered Dec 13 '18 at 03:58

If you don't mention the random_state in the code, then whenever you execute your code a new random value is generated and the train and test datasets would have different values each time.

However, if you use a particular value for random_state(random_state = 1 or any other value) everytime the result will be same,i.e, same values in train and test datasets.

score 7 · Answer 4 · answered Dec 12 '18 at 20:31

The random_state splits a randomly selected data but with a twist. And the twist is the order of the data will be same for a particular value of random_state.You need to understand that it's not a bool accpeted value. starting from 0 to any integer no, if you pass as random_state,it'll be a permanent order for it. Ex: the order you will get in random_state=0 remain same. After that if you execuit random_state=5 and again come back to random_state=0 you'll get the same order. And like 0 for all integer will go same. How ever random_state=None splits randomly each time.

If still having doubt watch this

score 4 · Answer 5 · answered Jan 22 '20 at 12:06

If you don't specify the random_state in your code, then every time you run(execute) your code a new random value is generated and the train and test datasets would have different values each time.

However, if a fixed value is assigned like random_state = 0 or 1 or 42 then no matter how many times you execute your code the result would be the same .i.e, same values in train and test datasets.

score 4 · Answer 6 · answered Mar 28 '20 at 13:56

random_state is None by default which means every time when you run your program you will get different output because of splitting between train and test varies within.

random_state = any int value means every time when you run your program you will get tehe same output because of splitting between train and test does not varies within.

score 3 · Answer 7 · answered Jul 11 '20 at 22:22

The random_state is an integer value which implies the selection of a random combination of train and test. When you set the test_size as 1/4 the there is a set generated of permutation and combination of train and test and each combination has one state. Suppose you have a dataset---> [1,2,3,4]

Train   |  Test   | State
[1,2,3]    [4]      **0**
[1,3,4]    [2]      **1**
[4,2,3]    [1]      **2**
[2,4,1]    [3]      **3**

We need it because while param tuning of model same state will considered again and again. So that there won't be any inference with the accuracy.

But in case of Random forest there is also similar story but in a different way w.r.t the variables.

score 2 · Answer 8 · answered Feb 12 '19 at 15:13

2

We used the random_state parameter for reproducibility of the initial shuffling of training datasets after each epoch.

answered Feb 12 '19 at 15:13

Debasish Bhol

19
4

score 1 · Answer 9 · edited Sep 05 '20 at 06:57

1

For multiple times of execution of our model, random state make sure that data values will be same for training and testing data sets. It fixes the order of data for train_test_split

edited Sep 05 '20 at 06:57

Smart Manoj

5,230
4
34
59

answered Jan 12 '20 at 09:40

hari

71
1
1

how does it make sure the same split? I mean can you explain the algorithm? – RafiO Jul 09 '23 at 22:36

score 0 · Answer 10 · answered Jul 11 '21 at 08:06

Lets say our dataset is having one feature and 10data points. X=[0,1,2,3,4,5,6,7,8,9] and lets say 0.3(30% is testset) is specified as test data percentage then we are going to have 10C3=120 different combinations of data.[Refer picture in link for tabular explanation]: https://i.stack.imgur.com/FZm4a.png

Based on the random number specified system will pick random state and assigns train and test data

scikit-learn random state in splitting dataset

10 Answers10

Linked

Related