MNT Add author/license note where missing and add the linter (scikit-…

…learn#29477)

Co-authored-by: Jérémie du Boisberranger <[email protected]>

.pre-commit-config.yaml

@@ -7,7 +7,7 @@ repos:
     -   id: trailing-whitespace
 -   repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.2.1
+    rev: v0.5.1
     hooks:
     -   id: ruff
         args: ["--fix", "--output-format=full"]

benchmarks/bench_kernel_pca_solvers_time_vs_n_components.py

@@ -36,8 +36,6 @@
 of components (this takes more time).
 """
 
-# Authors: Sylvain MARIE, Schneider Electric
-
 import time
 
 import matplotlib.pyplot as plt

benchmarks/bench_plot_parallel_pairwise.py

@@ -1,5 +1,6 @@
 # Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
+
 import time
 
 import matplotlib.pyplot as plt

examples/applications/plot_cyclical_feature_engineering.py

@@ -13,6 +13,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Data exploration on the Bike Sharing Demand dataset
 # ---------------------------------------------------

examples/applications/plot_digits_denoising.py

@@ -21,8 +21,8 @@
 
 """
 
-# Authors: Guillaume Lemaitre <[email protected]>
-# Licence: BSD 3 clause
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
 
 # %%
 # Load the dataset via OpenML

examples/applications/plot_face_recognition.py

@@ -12,6 +12,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 from time import time
 

examples/applications/plot_time_series_lagged_features.py

@@ -15,6 +15,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Analyzing the Bike Sharing Demand dataset
 # -----------------------------------------

examples/bicluster/plot_bicluster_newsgroups.py

@@ -23,6 +23,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import operator
 from collections import defaultdict
 from time import time

examples/calibration/plot_calibration_curve.py

@@ -14,6 +14,7 @@
 
 # Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Dataset
 # -------

examples/calibration/plot_compare_calibration.py

@@ -19,6 +19,7 @@
 # %%
 # Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
+
 #
 # Dataset
 # -------

examples/classification/plot_classifier_comparison.py

@@ -19,9 +19,7 @@
 
 """
 
-# Code source: Gaël Varoquaux
-#              Andreas Müller
-# Modified for documentation by Jaques Grobler
+# Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
 
 import matplotlib.pyplot as plt

examples/classification/plot_lda.py

@@ -8,6 +8,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.pyplot as plt
 import numpy as np
 

examples/classification/plot_lda_qda.py

@@ -11,6 +11,9 @@
 deviation with QDA.
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Data generation
 # ---------------

examples/cluster/plot_affinity_propagation.py

@@ -9,6 +9,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import numpy as np
 
 from sklearn import metrics

examples/cluster/plot_agglomerative_dendrogram.py

@@ -1,5 +1,6 @@
 # Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
+
 """
 =========================================
 Plot Hierarchical Clustering Dendrogram

examples/cluster/plot_bisect_kmeans.py

@@ -14,6 +14,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.pyplot as plt
 
 from sklearn.cluster import BisectingKMeans, KMeans

examples/cluster/plot_cluster_comparison.py

@@ -24,6 +24,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import time
 import warnings
 from itertools import cycle, islice

examples/cluster/plot_cluster_iris.py

@@ -18,8 +18,7 @@
 
 """
 
-# Code source: Gaël Varoquaux
-# Modified for documentation by Jaques Grobler
+# Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
 
 import matplotlib.pyplot as plt

examples/cluster/plot_dbscan.py

@@ -12,6 +12,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Data generation
 # ---------------

examples/cluster/plot_dict_face_patches.py

@@ -21,6 +21,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Load the data
 # -------------

examples/cluster/plot_digits_agglomeration.py

@@ -8,8 +8,7 @@
 
 """
 
-# Code source: Gaël Varoquaux
-# Modified for documentation by Jaques Grobler
+# Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
 
 import matplotlib.pyplot as plt

examples/cluster/plot_hdbscan.py

@@ -12,6 +12,10 @@
 
 We first define a couple utility functions for convenience.
 """
+
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 import matplotlib.pyplot as plt
 import numpy as np

examples/cluster/plot_inductive_clustering.py

@@ -20,8 +20,8 @@
 
 """
 
-# Authors: Chirag Nagpal
-#          Christos Aridas
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
 
 import matplotlib.pyplot as plt
 

examples/cluster/plot_kmeans_digits.py

@@ -25,6 +25,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Load the dataset
 # ----------------

examples/cluster/plot_kmeans_plusplus.py

@@ -10,6 +10,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.pyplot as plt
 
 from sklearn.cluster import kmeans_plusplus

examples/cluster/plot_kmeans_silhouette_analysis.py

@@ -31,6 +31,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.cm as cm
 import matplotlib.pyplot as plt
 import numpy as np

examples/cluster/plot_linkage_comparison.py

@@ -23,6 +23,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import time
 import warnings
 from itertools import cycle, islice

examples/cluster/plot_mean_shift.py

@@ -11,6 +11,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import numpy as np
 
 from sklearn.cluster import MeanShift, estimate_bandwidth

examples/cluster/plot_mini_batch_kmeans.py

@@ -14,6 +14,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Generate the data
 # -----------------

examples/compose/plot_compare_reduction.py

@@ -20,9 +20,8 @@
 
 """
 
-# Authors: Robert McGibbon
-#          Joel Nothman
-#          Guillaume Lemaitre
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
 
 # %%
 # Illustration of ``Pipeline`` and ``GridSearchCV``

examples/compose/plot_digits_pipe.py

@@ -10,8 +10,7 @@
 
 """
 
-# Code source: Gaël Varoquaux
-# Modified for documentation by Jaques Grobler
+# Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
 
 import matplotlib.pyplot as plt

examples/covariance/plot_covariance_estimation.py

@@ -15,6 +15,9 @@
 trade-off.
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Generate sample data
 # --------------------

examples/covariance/plot_lw_vs_oas.py

@@ -21,6 +21,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.pyplot as plt
 import numpy as np
 from scipy.linalg import cholesky, toeplitz

examples/covariance/plot_mahalanobis_distances.py

@@ -62,6 +62,9 @@
 
 """  # noqa: E501
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Generate data
 # --------------

examples/covariance/plot_robust_vs_empirical_covariance.py

@@ -53,6 +53,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.font_manager
 import matplotlib.pyplot as plt
 import numpy as np

examples/cross_decomposition/plot_compare_cross_decomposition.py

@@ -21,6 +21,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # Dataset based latent variables model
 # ------------------------------------

examples/cross_decomposition/plot_pcr_vs_pls.py

@@ -33,6 +33,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 # %%
 # The data
 # --------

examples/datasets/plot_digits_last_image.py

@@ -14,8 +14,7 @@
 
 """
 
-# Code source: Gaël Varoquaux
-# Modified for documentation by Jaques Grobler
+# Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
 
 import matplotlib.pyplot as plt

examples/datasets/plot_iris_dataset.py

@@ -15,8 +15,7 @@
 
 """
 
-# Code source: Gaël Varoquaux
-# Modified for documentation by Jaques Grobler
+# Authors: The scikit-learn developers
 # SPDX-License-Identifier: BSD-3-Clause
 
 # %%

examples/datasets/plot_random_dataset.py

@@ -14,6 +14,9 @@
 
 """
 
+# Authors: The scikit-learn developers
+# SPDX-License-Identifier: BSD-3-Clause
+
 import matplotlib.pyplot as plt
 
 from sklearn.datasets import make_blobs, make_classification, make_gaussian_quantiles