ishangsf/rspec
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Modify rule S6709: Add how to fix it for Scikit-learn (#3883)

Browse Source
This commit is contained in:
David Kunzmann 2024-05-07 14:21:34 +02:00 committed by GitHub
parent 621b7ce90e
commit 86d6b7c75b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 64 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/header_names/allowed_framework_names.adoc
View File

@ -91,6 +91,7 @@
* Jinja
* lxml
* MySQL Connector/Python
* Numpy
* Paramiko
* pyca
* PyCrypto
@ -105,6 +106,7 @@
* PyYAML
* Requests
* Scrypt
* Scikit-Learn
* SignXML
* SQLAlchemy
* ssl

27
rules/S6709/python/how-to-fix-it/numpy.adoc Normal file
View File

@ -0,0 +1,27 @@
== How to fix it in Numpy
To fix this issue, provide a predictable seed to the random number generator.
=== Code examples
==== Noncompliant code example
[source,python,diff-id=1,diff-type=noncompliant]
----
import numpy as np
def foo():
generator = np.random.default_rng() # Noncompliant: no seed parameter is provided
x = generator.uniform()
----
==== Compliant solution
[source,python,diff-id=1,diff-type=compliant]
----
import numpy as np
def foo():
generator = np.random.default_rng(42) # Compliant
x = generator.uniform()
----

29
rules/S6709/python/how-to-fix-it/sklearn.adoc Normal file
View File

@ -0,0 +1,29 @@
== How to fix it in Scikit-Learn
To fix this issue, provide a predictable seed to the estimator or the utility function.
=== Code examples
==== Noncompliant code example
[source,python,diff-id=2,diff-type=noncompliant]
----
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_iris
X, y = load_iris(return_X_y=True)
X_train, _, y_train, _ = train_test_split(X, y) # Noncompliant: no seed parameter is provided
----
==== Compliant solution
[source,python,diff-id=2,diff-type=compliant]
----
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_iris
import numpy as np
rng = np.random.default_rng(42)
X, y = load_iris(return_X_y=True)
X_train, _, y_train, _ = train_test_split(X, y, random_state=rng.integers(1)) # Compliant
----

29
rules/S6709/python/rule.adoc
View File

@ -26,38 +26,20 @@ Note that a global seed for `RandomState` can be set using `numpy.random.seed` o
In contexts that are not related to data science and machine learning, having a predictable seed may not be the desired behavior. Therefore, this rule will only raise issues if machine learning and data science libraries are being used.
== How to fix it
To fix this issue, provide a predictable seed to the random number generator.
// How to fix it section
=== Code examples
include::how-to-fix-it/numpy.adoc[]
==== Noncompliant code example
include::how-to-fix-it/sklearn.adoc[]
[source,python,diff-id=1,diff-type=noncompliant]
----
import numpy as np
def foo():
generator = np.random.default_rng() # Noncompliant: no seed parameter is provided
x = generator.uniform()
----
==== Compliant solution
[source,python,diff-id=1,diff-type=compliant]
----
import numpy as np
def foo():
generator = np.random.default_rng(42) # Compliant
x = generator.uniform()
----
== Resources
=== Documentation
* NumPy documentation - https://numpy.org/neps/nep-0019-rng-policy.html[NEP 19 RNG Policy]
* Scikit-learn documentation - https://scikit-learn.org/stable/glossary.html#term-random_state[Glossary random_state]
* Scikit-learn documentation - https://scikit-learn.org/stable/common_pitfalls.html#controlling-randomness[Controlling randomness]
=== Standards
@ -66,3 +48,4 @@ def foo():
=== Related rules
* S6711 - `numpy.random.Generator` should be preferred to `numpy.random.RandomState`