Scribd
Upload
3 views1 page

Understanding Random State in Splits

The document explains the use of the train_test_split function in predictive analytics, particularly focusing on the random_state parameter. By default, the function generates different random splits unless a specific random_state value is provided, which ensures consistent results across multiple runs. This feature is useful for reproducibility in documentation and examples.

Uploaded by

rs9154040
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views1 page

Understanding Random State in Splits

The document explains the use of the train_test_split function in predictive analytics, particularly focusing on the random_state parameter. By default, the function generates different random splits unless a specific random_state value is provided, which ensures consistent results across multiple runs. This feature is useful for reproducibility in documentation and examples.

Uploaded by

rs9154040
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Predictive Analytics: Regression

Using Random State


The train_test_split function divides arrays or matrices into random training
and test subsets. By default, if the random_state parameter is not specified,
the function will generate different random splits each time it is executed.
This is the expected behavior of the function. For example:

a, b = [Link](10).reshape((5, 2)), range(5)


train_test_split(a, b)

[array([[6, 7],
[8, 9],
[2, 3]]), array([[0, 1],
[4, 5]]), [3, 4, 1], [0, 2]]

train_test_split(a, b)
[array([[0, 1],
[6, 7],
[2, 3]]), array([[8, 9],
[4, 5]]), [0, 3, 1], [4, 2]]

In contrast, when you specify a value for random_state (e.g., random_state =


some_number), the train_test_split function ensures that the output of the
split remains consistent across multiple runs. This means that the result
obtained from the first run will be identical to the result obtained from the
second run and subsequent runs. The specific value chosen for random_state
(e.g., 42, 0, 21, etc.) is not important as long as it is consistently used. This
feature is beneficial when you desire reproducible outcomes, such as in
documentation or when you want others to consistently observe the same
results when running the examples.

You might also like