Package scikit-learn

.circleci/config.yml

@@ -24,6 +24,12 @@ jobs:
           name: build
           no_output_timeout: 1200
           command: |
+
+            # download scipy package from https://github.com/iodide-project/pyodide/pull/211
+            mkdir -p build
+            wget -q -O build/scipy.js https://1463-122663163-gh.circle-artifacts.com/0/home/circleci/repo/build/scipy.js
+            wget -q -O build/scipy.data https://1463-122663163-gh.circle-artifacts.com/0/home/circleci/repo/build/scipy.data
+
             ccache -z
             make
             ccache -s

packages/scikit-learn/meta.yaml

+package:
+  name: scikit-learn
+  version: 0.20.0
+
+source:
+  url: https://pypi.io/packages/source/s/scikit-learn/scikit-learn-0.20.0.tar.gz
+  sha256: 97d1d971f8ec257011e64b7d655df68081dd3097322690afa1a71a1d755f8c18
+
+  patches:
+   #- patches/unvendor-joblib.patch
+   - patches/use-site-joblib.patch
+
+build:
+  cflags: -Wno-implicit-function-declaration
+
+requirements:
+  run:
+    - numpy   # TODO: add scipy, joblib once the corresponding PRs are merged
+
+test:
+  imports:
+    - sklearn
+    - sklearn.cluster
+    - sklearn.compose
+    - sklearn.covariance
+    - sklearn.cross_decomposition
+    - sklearn.datasets
+    - sklearn.decomposition

packages/scikit-learn/patches/use-site-joblib.patch

+commit 16cf9dc5f79533a121a421b095b6e7ef9ee76e9c
+Author: Roman Yurchak <rth.yurchak@pm.me>
+Date:   Thu Oct 25 16:56:54 2018 +0200
+
+    Use site joblib
+
+diff --git a/sklearn/datasets/species_distributions.py b/sklearn/datasets/species_distributions.py
+index 6d8acddcc..8191048d7 100644
+--- a/sklearn/datasets/species_distributions.py
++++ b/sklearn/datasets/species_distributions.py
+@@ -51,7 +51,7 @@ from .base import _fetch_remote
+ from .base import RemoteFileMetadata
+ from ..utils import Bunch
+ from sklearn.datasets.base import _pkl_filepath
+-from sklearn.externals import joblib
++import joblib
+ 
+ PY3_OR_LATER = sys.version_info[0] >= 3
+ 
+diff --git a/sklearn/ensemble/tests/test_forest.py b/sklearn/ensemble/tests/test_forest.py
+index d7586c286..d1168cdcf 100644
+--- a/sklearn/ensemble/tests/test_forest.py
++++ b/sklearn/ensemble/tests/test_forest.py
+@@ -23,7 +23,11 @@ import pytest
+ 
+ from sklearn.utils import parallel_backend
+ from sklearn.utils import register_parallel_backend
+-from sklearn.externals.joblib.parallel import LokyBackend
++try:
++    from sklearn.externals.joblib.parallel import LokyBackend
++except ImportError:
++    LokyBackend = object
++
+ 
+ from sklearn.utils.testing import assert_almost_equal
+ from sklearn.utils.testing import assert_array_almost_equal
+diff --git a/sklearn/metrics/tests/test_score_objects.py b/sklearn/metrics/tests/test_score_objects.py
+index da04b4215..fc3f6a6b1 100644
+--- a/sklearn/metrics/tests/test_score_objects.py
++++ b/sklearn/metrics/tests/test_score_objects.py
+@@ -40,7 +40,7 @@ from sklearn.datasets import load_diabetes
+ from sklearn.model_selection import train_test_split, cross_val_score
+ from sklearn.model_selection import GridSearchCV
+ from sklearn.multiclass import OneVsRestClassifier
+-from sklearn.externals import joblib
++import joblib
+ 
+ 
+ REGRESSION_SCORERS = ['explained_variance', 'r2',
+diff --git a/sklearn/neighbors/tests/test_kde.py b/sklearn/neighbors/tests/test_kde.py
+index 990942c9e..e9a6c31bd 100644
+--- a/sklearn/neighbors/tests/test_kde.py
++++ b/sklearn/neighbors/tests/test_kde.py
+@@ -10,7 +10,7 @@ from sklearn.pipeline import make_pipeline
+ from sklearn.datasets import make_blobs
+ from sklearn.model_selection import GridSearchCV
+ from sklearn.preprocessing import StandardScaler
+-from sklearn.externals import joblib
++import joblib
+ 
+ 
+ def compute_kernel_slow(Y, X, kernel, h):
+diff --git a/sklearn/tests/test_site_joblib.py b/sklearn/tests/test_site_joblib.py
+index bffd43cc1..df4c96893 100644
+--- a/sklearn/tests/test_site_joblib.py
++++ b/sklearn/tests/test_site_joblib.py
+@@ -1,7 +1,10 @@
+ import os
+ import pytest
+ from sklearn import externals
+-from sklearn.externals import joblib as joblib_vendored
++try:
++    from sklearn.externals import joblib as joblib_vendored
++except ImportError:
++    joblib_vendored = None
+ from sklearn.utils import Parallel, delayed, Memory, parallel_backend
+ 
+ if os.environ.get('SKLEARN_SITE_JOBLIB', False):
+diff --git a/sklearn/utils/testing.py b/sklearn/utils/testing.py
+index 75b378961..b81b9ab58 100644
+--- a/sklearn/utils/testing.py
++++ b/sklearn/utils/testing.py
+@@ -44,7 +44,7 @@ except NameError:
+ 
+ import sklearn
+ from sklearn.base import BaseEstimator
+-from sklearn.externals import joblib
++import joblib
+ from sklearn.utils.fixes import signature
+ from sklearn.utils import deprecated, IS_PYPY, _IS_32BIT
+ 
+diff --git a/sklearn/utils/tests/test_estimator_checks.py b/sklearn/utils/tests/test_estimator_checks.py
+index bf8412b3e..2eebb36b0 100644
+--- a/sklearn/utils/tests/test_estimator_checks.py
++++ b/sklearn/utils/tests/test_estimator_checks.py
+@@ -5,7 +5,7 @@ import numpy as np
+ import scipy.sparse as sp
+ 
+ from sklearn.externals.six.moves import cStringIO as StringIO
+-from sklearn.externals import joblib
++import joblib
+ 
+ from sklearn.base import BaseEstimator, ClassifierMixin
+ from sklearn.utils import deprecated

test/packages/test_scikit-learn.py

+from textwrap import dedent
+
+def test_scikit_learn(selenium_standalone):
+    selenium = selenium_standalone
+    # no automatic depedency resolution for now
+    selenium.load_package(["numpy", "joblib"])
+    selenium.load_package("scipy")
+    selenium.load_package("scikit-learn")
+    assert selenium.run("""
+        import numpy as np
+        import sklearn
+        from sklearn.linear_model import LogisticRegression
+
+        rng = np.random.RandomState(42)
+        X = rng.rand(100, 20)
+        y = rng.randint(5, size=100)
+
+        estimator = LogisticRegression(solver='liblinear')
+        estimator.fit(X, y)
+        print(estimator.predict(X))
+        estimator.score(X, y)
+        """) > 0
+    print(selenium.logs)
+
+def test_import(selenium_standalone):
+    selenium = selenium_standalone
+    # no automatic depedency resolution for now
+    selenium.load_package(["numpy", "joblib"])
+    selenium.load_package("scipy")
+    selenium.load_package("scikit-learn")
+    cmd = dedent("""
+            import sklearn
+            import sklearn.calibration
+            import sklearn.calibration
+            import sklearn.cluster
+            import sklearn.compose
+            import sklearn.covariance
+            import sklearn.cross_decomposition
+            import sklearn.datasets
+            import sklearn.decomposition
+            import sklearn.discriminant_analysis
+            import sklearn.dummy
+            import sklearn.ensemble
+            import sklearn.exceptions
+            import sklearn.externals
+            import sklearn.feature_extraction
+            import sklearn.feature_selection
+            import sklearn.gaussian_process
+            import sklearn.impute
+            import sklearn.isotonic
+            import sklearn.kernel_approximation
+            import sklearn.kernel_ridge
+            import sklearn.linear_model
+            import sklearn.manifold
+            import sklearn.metrics
+            import sklearn.mixture
+            import sklearn.model_selection
+            import sklearn.multiclass
+            import sklearn.multioutput
+            import sklearn.naive_bayes
+            import sklearn.neighbors
+            import sklearn.neural_network
+            import sklearn.pipeline
+            import sklearn.preprocessing
+            import sklearn.random_projection
+            import sklearn.semi_supervised
+            import sklearn.svm
+            import sklearn.tree
+            import sklearn.utils
+            """).splitlines()
+
+    for line in cmd:
+        try:
+            selenium.run(line)
+            print(f'{line} -- OK')
+        except:
+            print(f'Error: {line} failed')
+            print(selenium.logs)

test/test_common.py

@@ -23,7 +23,7 @@ def registered_packages_meta():
             for name in packages}
 
 
-UNSUPPORTED_PACKAGES = {'chrome': ['pandas', 'scipy'],
+UNSUPPORTED_PACKAGES = {'chrome': ['pandas', 'scipy', 'scikit-learn'],
                         'firefox': []}