Merge branch 'main' into main

braniii · web-flow · commit 174d645ed12b · 2023-09-01T12:11:42.000+02:00
diff --git a/.circleci/config.yml b/.circleci/config.yml
diff --git a/.coveragerc b/.coveragerc
@@ -18,4 +18,4 @@ exclude_lines =
     if 0:
     if __name__ == .__main__.:
     if self.verbose:
-show_missing = True
+show_missing = True
diff --git a/.github/workflows/build-wheels.yml b/.github/workflows/build-wheels.yml
@@ -1,14 +1,14 @@
 name: build_wheels
 
-on: # [push, pull_request]
-  release:
-    types:
-      - created
-  workflow_dispatch:
-    inputs:
-      version:
-        description: 'Manually trigger wheel build in Github UI'
-        required: true
+on: #[push, pull_request]
+   release:
+     types:
+       - created
+   workflow_dispatch:
+     inputs:
+       version:
+         description: 'Manually trigger wheel build in Github UI'
+         required: true
 
 
 jobs:
@@ -17,7 +17,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [ubuntu-18.04, windows-latest, macos-latest]
+        os: [ubuntu-22.04, windows-latest, macos-latest]
 
     steps:
       - uses: actions/checkout@v2
@@ -28,12 +28,13 @@ jobs:
           python-version: '3.8'
       - name: Install cibuildwheel
         run: |
-          python -m pip install cibuildwheel==1.10.0
+          python -m pip install cibuildwheel==2.12.1
       - name: Build wheels
         env:
           # We only build for Python 3.6+. On Linux manylinux2010 is used.
           # Skipping pypy wheels for now since scipy & scikit-learn haven't build them yet.
-          CIBW_SKIP: "pp* *p27* *p35*"
+          # Skip python3.11 for 32bit.
+          CIBW_SKIP: "pp* *-win32 *-manylinux_i686 *musllinux*"
           CIBW_TEST_REQUIRES: "pytest pandas scikit-learn"
           CIBW_TEST_COMMAND: "pytest --pyargs sklearn_extra"
         run: |
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -1,16 +1,16 @@
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.4.0
+    rev: v4.4.0
     hooks:
     -   id: check-yaml
     -   id: end-of-file-fixer
     -   id: trailing-whitespace
 -   repo: https://github.com/psf/black
-    rev: 20.8b1
+    rev: 23.1.0
     hooks:
     -   id: black
--   repo: https://gitlab.com/pycqa/flake8
-    rev: 3.9.0
+-   repo: https://github.com/pycqa/flake8
+    rev: 6.0.0
     hooks:
     -   id: flake8
         types: [file, python]
diff --git a/azure-pipelines.yml b/azure-pipelines.yml
@@ -66,9 +66,9 @@ jobs:
     env:
       CODECOV_TOKEN: $(CODECOV_TOKEN)
 
-- job: 'macOS1014'
+- job: 'macOS'
   pool:
-    vmImage: 'macOS-10.14'
+    vmImage: 'macOS-latest'
   strategy:
     matrix:
       Python37:
@@ -122,9 +122,9 @@ jobs:
       CODECOV_TOKEN: $(CODECOV_TOKEN)
 
 
-- job: 'win2016'
+- job: 'win'
   pool:
-    vmImage: 'vs2017-win2016'
+    vmImage: 'windows-latest'
   strategy:
     matrix:
       Python38:
diff --git a/conftest.py b/conftest.py
@@ -7,7 +7,6 @@
 
 
 def pytest_collection_modifyitems(config, items):
-
     # numpy changed the str/repr formatting of numpy arrays in 1.14. We want to
     # run doctests only for numpy >= 1.14.
     skip_doctests = False
diff --git a/doc/changelog.rst b/doc/changelog.rst
@@ -1,17 +1,19 @@
 Changelog
 =========
 
-Unreleased
+Version 0.3.0
 ----------
 
+- Fix bug with random initialization of KMedoids [`#129 <https://github.com/scikit-learn-contrib/scikit-learn-extra/pull/129>`_].
+- `KMedoids` supports array-like init method [`#137 <https://github.com/scikit-learn-contrib/scikit-learn-extra/pull/137>`_].
 - Add a stopping criterion and parameter tuning heuristic for Huber robust mean
     estimator.
 - Add `CLARA` (Clustering for Large Applications) which extends k-medoids to
     be more scalable using a sampling approach.
     [`#83 <https://github.com/scikit-learn-contrib/scikit-learn-extra/pull/83>`_].
 - Fix `_estimator_type` for :class:`~sklearn_extra.robust` estimators. Fix
   misbehavior of scikit-learn's :class:`~sklearn.model_selection.cross_val_score` and
-  :class:`~sklearn.grid_search.GridSearchCV` for :class:`~sklearn_extra.robust.RobustWeightedClassifier`
+  :class:`~sklearn.grid_search.GridSearchCV` for :class:`~sklearn_extra.robust.RobustWeightedClassifier`.
 
 Version 0.2.0
 -------------
diff --git a/doc/docs.md b/doc/docs.md
@@ -5,6 +5,5 @@
 - scikit-learn(>=0.21)
 - Cython (>0.28)
 ### User Installation:
-You can install scikit-learn-extra using this command: 
+You can install scikit-learn-extra using this command:
 `pip install https://github.com/scikit-learn-contrib/scikit-learn-extra/archive/master.zip`
-
diff --git a/doc/index.rst b/doc/index.rst
@@ -28,4 +28,3 @@ scikit-learn-extra is a Python module for machine learning that extends scikit-l
 
    contributing
    changelog
-
diff --git a/doc/modules/kernel_approximation.rst b/doc/modules/kernel_approximation.rst
@@ -6,29 +6,29 @@ Kernel map approximation for faster kernel methods
 
 .. currentmodule:: sklearn_extra.kernel_approximation
 
-Kernel methods, which are among the most flexible and influential tools in 
-machine learning with applications in virtually all areas of the field, rely 
+Kernel methods, which are among the most flexible and influential tools in
+machine learning with applications in virtually all areas of the field, rely
 on high-dimensional feature spaces in order to construct powerfull classifiers or
 regressors or clustering algorithms. The main drawback of kernel methods
 is their prohibitive computational complexity. Both spatial and temporal complexity
  is at least quadratic because we have to compute the whole kernel matrix.
 
 One of the popular way to improve the computational scalability of kernel methods is
 to approximate the feature map impicit behind the kernel method. In practice,
-this means that we will compute a low dimensional approximation of the 
+this means that we will compute a low dimensional approximation of the
 the otherwise high-dimensional embedding used to define the kernel method.
 
 :class:`Fastfood` approximates feature map of an RBF kernel by Monte Carlo approximation
 of its Fourier transform.
 
-Fastfood replaces the random matrix of Random Kitchen Sinks 
+Fastfood replaces the random matrix of Random Kitchen Sinks
 (`RBFSampler <https://scikit-learn.org/stable/modules/generated/sklearn.kernel_approximation.RBFSampler.html#sklearn.kernel_approximation.RBFSampler>`_)
 with an approximation that uses the Walsh-Hadamard transformation to gain
 significant speed and storage advantages.  The computational complexity for
 mapping a single example is O(n_components log d).  The space complexity is
-O(n_components).  
+O(n_components).
 
 See `scikit-learn User-guide <https://scikit-learn.org/stable/modules/kernel_approximation.html#kernel-approximation>`_ for more general informations on kernel approximations.
 
-See also :class:`EigenProRegressor <sklearn_extra.kernel_methods.EigenProRegressor>` and :class:`EigenProClassifier <sklearn_extra.kernel_methods.EigenProClassifier>` for another 
+See also :class:`EigenProRegressor <sklearn_extra.kernel_methods.EigenProRegressor>` and :class:`EigenProClassifier <sklearn_extra.kernel_methods.EigenProClassifier>` for another
 way to compute fast kernel methods algorithms.
diff --git a/examples/cluster/plot_clustering.py b/examples/cluster/plot_clustering.py
@@ -70,7 +70,6 @@
     # Define two other clustering algorithms
     kmeans_rob = RobustWeightedKMeans(
         n_clusters,
-        eta0=0.01,
         weighting="mom",
         max_iter=100,
         k=int(n_samples / 20),
diff --git a/examples/cluster/plot_kmedoids_digits.py b/examples/cluster/plot_kmedoids_digits.py
@@ -63,7 +63,6 @@
 plot_cols = 2
 
 for i, (model, description) in enumerate(selected_models):
-
     # Obtain labels for each point in mesh. Use last trained model.
     model.fit(reduced_data)
     Z = model.predict(np.c_[xx.ravel(), yy.ravel()])
diff --git a/examples/plot_clara_digits.py b/examples/plot_clara_digits.py
@@ -79,7 +79,6 @@
 plot_cols = 2
 
 for i, (model, description) in enumerate(selected_models):
-
     # Obtain labels for each point in mesh. Use last trained model.
     init_time = time.time()
     model.fit(reduced_data)
diff --git a/examples/plot_kmedoids.py b/examples/plot_kmedoids.py
@@ -37,7 +37,6 @@
     plt.cm.Spectral(each) for each in np.linspace(0, 1, len(unique_labels))
 ]
 for k, col in zip(unique_labels, colors):
-
     class_member_mask = labels == k
 
     xy = X[class_member_mask]
diff --git a/examples/robust/plot_robust_classification_diabete.py b/examples/robust/plot_robust_classification_diabete.py
@@ -42,7 +42,6 @@
     weighting="huber",
     loss="hinge",
     c=1.35,
-    eta0=1e-3,
     max_iter=300,
 )
 
@@ -60,7 +59,6 @@
         weighting="huber",
         loss="hinge",
         c=1.35,
-        eta0=1e-3,
         max_iter=300,
         random_state=rng,
     )
diff --git a/examples/robust/plot_robust_regression_california_houses.py b/examples/robust/plot_robust_regression_california_houses.py
@@ -48,14 +48,13 @@ def quadratic_loss(est, X, y, X_test, y_test):
 estimators = [
     (
         "SGD",
-        SGDRegressor(learning_rate="adaptive", eta0=1e-2),
+        SGDRegressor(learning_rate="adaptive"),
     ),
     (
         "RobustWeightedRegressor",
         RobustWeightedRegressor(
             weighting="huber",
             c=0.1,
-            eta0=1e-2,
             sgd_args={
                 "learning_rate": "invscaling",
             },
diff --git a/pyproject.toml b/pyproject.toml
@@ -14,3 +14,6 @@ requires = [
 
 [tool.black]
 line-length = 79
+
+[tool.cibuildwheel]
+manylinux-i686-image = "manylinux2010"
diff --git a/sklearn_extra/cluster/_k_medoids.py b/sklearn_extra/cluster/_k_medoids.py
@@ -228,6 +228,7 @@ def fit(self, X, y=None):
         X = check_array(
             X, accept_sparse=["csr", "csc"], dtype=[np.float64, np.float32]
         )
+        self.n_features_in_ = X.shape[1]
         if self.n_clusters > X.shape[0]:
             raise ValueError(
                 "The number of medoids (%d) must be less "
@@ -650,6 +651,8 @@ def fit(self, X, y=None):
         self
         """
         X = check_array(X, dtype=[np.float64, np.float32])
+        self.n_features_in_ = X.shape[1]
+
         n = len(X)
 
         random_state_ = check_random_state(self.random_state)
diff --git a/sklearn_extra/cluster/tests/test_k_medoids.py b/sklearn_extra/cluster/tests/test_k_medoids.py
@@ -224,8 +224,11 @@ def test_precomputed():
     kmedoids.fit(D_1)
 
     assert_allclose(kmedoids.inertia_, 0.2)
-    assert_array_equal(kmedoids.medoid_indices_, [2, 0])
-    assert_array_equal(kmedoids.labels_, [1, 1, 0, 0])
+    assert_array_equal(
+        np.sort(kmedoids.medoid_indices_), np.sort([2, 0])
+    )  # the order may change
+    assert kmedoids.labels_[0] == kmedoids.labels_[1]
+    assert kmedoids.labels_[2] == kmedoids.labels_[3]
     assert kmedoids.cluster_centers_ is None
 
     med_1, med_2 = tuple(kmedoids.medoid_indices_)
diff --git a/sklearn_extra/kernel_approximation/_fastfood.py b/sklearn_extra/kernel_approximation/_fastfood.py
@@ -168,6 +168,7 @@ def fit(self, X, y=None):
             Returns the transformer.
         """
         X = check_array(X, order="C", dtype=np.float64)
+        self.n_features_in_ = X.shape[1]
 
         d_orig = X.shape[1]
         rng = check_random_state(self.random_state)
diff --git a/sklearn_extra/kernel_methods/_eigenpro.py b/sklearn_extra/kernel_methods/_eigenpro.py
@@ -322,6 +322,7 @@ def _raw_fit(self, X, Y):
             ensure_min_samples=3,
             y_numeric=True,
         )
+        self.n_features_in_ = X.shape[1]
         Y = Y.astype(np.float32)
         random_state = check_random_state(self.random_state)
 
diff --git a/sklearn_extra/robust/robust_weighted_estimator.py b/sklearn_extra/robust/robust_weighted_estimator.py
diff --git a/sklearn_extra/robust/tests/test_mean_estimators.py b/sklearn_extra/robust/tests/test_mean_estimators.py
diff --git a/sklearn_extra/robust/tests/test_robust_weighted_estimator.py b/sklearn_extra/robust/tests/test_robust_weighted_estimator.py

Original file line number	Diff line number	Diff line change
`@@ -28,4 +28,3 @@ scikit-learn-extra is a Python module for machine learning that extends scikit-l`
`28`	`28`
`29`	`29`	`contributing`
`30`	`30`	`changelog`
`31`		`-`