Merge remote-tracking branch 'origin/master' into is/9360

Author: glemaitre

README.rst

@@ -138,7 +138,7 @@ Below is a list of the methods currently implemented in this module.
     2. Extraction of majority-minority Tomek links [1]_
     3. Under-sampling with Cluster Centroids
     4. NearMiss-(1 & 2 & 3) [2]_
-    5. Condensend Nearest Neighbour [3]_
+    5. Condensed Nearest Neighbour [3]_
     6. One-Sided Selection [4]_
     7. Neighboorhood Cleaning Rule [5]_
     8. Edited Nearest Neighbours [6]_
@@ -163,7 +163,7 @@ Below is a list of the methods currently implemented in this module.
 
 The different algorithms are presented in the sphinx-gallery_.
 
-.. _sphinx-gallery: http://contrib.scikit-learn.org/imbalanced-learn/auto_examples
+.. _sphinx-gallery: http://contrib.scikit-learn.org/imbalanced-learn/stable/auto_examples/index.html
 
 
 References:

appveyor.yml

@@ -1,3 +1,5 @@
+build: false
+
 environment:
   # There is no need to run the build for all the Python version /
   # architectures combo as the generated nilearn wheel is the same on all
@@ -30,31 +32,12 @@ install:
   # Add Library/bin directory to fix issue
   # https://github.com/conda/conda/issues/1753
   - "SET PATH=%PYTHON%;%PYTHON%\\Scripts;%PYTHON%\\Library\\bin;%PATH%"
-
-  # Check that we have the expected version and architecture for Python
-  - "python --version"
-  - "python -c \"import struct; print(struct.calcsize('P') * 8)\""
-
-  # Installed prebuilt dependencies from conda
-  - "conda install pip numpy scipy scikit-learn=0.19.0 pandas nose wheel matplotlib -y -q"
-
-  # Install other nilearn dependencies
-  - "pip install coverage nose-timer pytest pytest-cov"
-  - "python setup.py bdist_wheel"
-  - ps: "ls dist"
-
-  # Install the generated wheel package to test it
-  - "pip install --pre --no-index --find-links dist/ imbalanced-learn"
-
-# Not a .NET project, we build in the install step instead
-build: false
+  - conda install pip scipy numpy scikit-learn=0.19 pandas -y -q
+  - conda install pytest pytest-cov -y -q
+  - conda install nose -y -q  # FIXME: remove this line when using sklearn > 0.19
+  - pip install .
 
 test_script:
-  - powershell ./build_tools/appveyor/test.ps1
-
-artifacts:
-  # Archive the generated packages in the ci.appveyor.com build report.
-  - path: dist\*
-
-#on_success:
-#  - TODO: upload the content of dist/*.whl to a public wheelhouse
+  - mkdir for_test
+  - cd for_test
+  - pytest --pyargs imblearn --cov-report term-missing --cov=imblearn

build_tools/appveyor/test.ps1

@@ -0,0 +1,14 @@
+# This file is here so that when running from the root folder
+# ./sklearn is added to sys.path by pytest.
+# See https://docs.pytest.org/en/latest/pythonpath.html for more details.
+# For example, this allows to build extensions in place and run pytest
+# doc/modules/clustering.rst and use sklearn from the local folder
+# rather than the one from site-packages.
+
+# Set numpy array str/repr to legacy behaviour on numpy > 1.13 to make
+# the doctests pass
+import numpy as np
+try:
+    np.set_printoptions(legacy='1.13')
+except TypeError:
+    pass

conftest.py

@@ -110,7 +110,7 @@ takes the same parameters than the scikit-learn
 
 It also possible to turn a balanced bagging classifier into a balanced random
 forest using a decision tree classifier and setting the parameter
-``max_features='auto'`. It allows to randomly select a subset of features for
+``max_features='auto'``. It allows to randomly select a subset of features for
 each tree::
 
   >>> brf = BalancedBaggingClassifier(

doc/ensemble.rst

@@ -60,9 +60,9 @@ for usage example.
 From random over-sampling to SMOTE and ADASYN
 ---------------------------------------------
 
-Apart from the random sampling with replacement, there is two popular methods
-to over-sample minority classes: (i) Synthetic Minority Oversampling Technique
-(SMOTE) and (ii) Adaptive Synthetic (ADASYN) sampling method. These algorithm
+Apart from the random sampling with replacement, there are two popular methods
+to over-sample minority classes: (i) the Synthetic Minority Oversampling Technique
+(SMOTE) and (ii) the Adaptive Synthetic (ADASYN) sampling method. These algorithms
 can be used in the same manner::
 
   >>> from imblearn.over_sampling import SMOTE, ADASYN

doc/over_sampling.rst

@@ -1,4 +1,4 @@
-"""
+"""
 ``imbalanced-learn`` is a set of python methods to deal with imbalanced
 datset in machine learning and pattern recognition.
 """
@@ -21,4 +21,5 @@
 # Dev branch marker is: 'X.Y.dev' or 'X.Y.devN' where N is an integer.
 # 'X.Y.dev0' is the canonical version of 'X.Y.dev'
 #
+
 __version__ = '0.4.0.dev0'

imblearn/_version.py

@@ -119,7 +119,7 @@ def __getstate__(self):
 
     def __setstate__(self, dict):
         """Re-open the logger."""
-        logger = logging.getLogger(__name__)
+        logger = logging.getLogger(self.__module__)
         self.__dict__.update(dict)
         self.logger = logger
 
@@ -133,7 +133,7 @@ class BaseSampler(SamplerMixin):
 
     def __init__(self, ratio='auto'):
         self.ratio = ratio
-        self.logger = logging.getLogger(__name__)
+        self.logger = logging.getLogger(self.__module__)
 
     def fit(self, X, y):
         """Find the classes statistics before to perform sampling.

imblearn/base.py

@@ -16,8 +16,6 @@
 from ..pipeline import Pipeline
 from ..under_sampling import RandomUnderSampler
 
-old_generate = _generate_bagging_indices
-
 
 class BalancedBaggingClassifier(BaggingClassifier):
     """A Bagging classifier with additional balancing.

imblearn/ensemble/classifier.py

@@ -16,10 +16,8 @@
 
 
 class ADASYN(BaseOverSampler):
-    """Perform over-sampling using ADASYN.
-
-    Perform over-sampling using Adaptive Synthetic Sampling Approach for
-    Imbalanced Learning.
+    """Perform over-sampling using Adaptive Synthetic (ADASYN) sampling
+    approach for imbalanced datasets.
 
     Read more in the :ref:`User Guide <smote_adasyn>`.
 

imblearn/over_sampling/adasyn.py

@@ -2,7 +2,6 @@
 # Authors: Guillaume Lemaitre <[email protected]>
 #          Christos Aridas
 # License: MIT
-
 from sklearn.utils.testing import _named_check
 
 from imblearn.utils.estimator_checks import check_estimator, _yield_all_checks
@@ -34,3 +33,7 @@ def test_non_meta_estimators():
             continue
         for check in _yield_all_checks(name, Estimator):
             yield _named_check(check, name), name, Estimator
+
+            logger_name = Estimator().logger.name
+            class_module_name = Estimator.__module__
+            assert logger_name == class_module_name

imblearn/tests/test_common.py

@@ -24,7 +24,6 @@
 from sklearn.exceptions import NotFittedError
 from sklearn.utils.testing import assert_allclose
 from sklearn.utils.testing import set_random_state
-from sklearn.externals.funcsigs import signature
 
 from imblearn.base import SamplerMixin
 from imblearn.over_sampling.base import BaseOverSampler
@@ -214,12 +213,10 @@ def check_samplers_sparse(name, Sampler):
                             estimator=KMeans(random_state=1,
                                              algorithm='full'))]
     else:
-        sampler_attr = signature(Sampler.__init__).parameters.keys()
-        if 'random_state' in sampler_attr:
-            samplers = [Sampler(random_state=0)]
-        else:
-            samplers = [Sampler()]
+        samplers = [Sampler()]
+
     for sampler in samplers:
+        set_random_state(sampler)
         X_res_sparse, y_res_sparse = sampler.fit_sample(X_sparse, y)
         X_res, y_res = sampler.fit_sample(X, y)
         if not isinstance(sampler, BaseEnsembleSampler):
@@ -246,16 +243,16 @@ def check_samplers_pandas(name, Sampler):
         samplers = [Sampler(random_state=0, kind=kind)
                     for kind in ('regular', 'borderline1',
                                  'borderline2', 'svm')]
+
     elif isinstance(Sampler(), NearMiss):
-            samplers = [Sampler(version=version)
-                        for version in (1, 2, 3)]
+        samplers = [Sampler(version=version)
+                    for version in (1, 2, 3)]
+
     else:
-        sampler_attr = signature(Sampler.__init__).parameters.keys()
-        if 'random_state' in sampler_attr:
-            samplers = [Sampler(random_state=0)]
-        else:
-            samplers = [Sampler()]
+        samplers = [Sampler()]
+
     for sampler in samplers:
+        set_random_state(sampler)
         X_res_pd, y_res_pd = sampler.fit_sample(X_pd, y_pd)
         X_res, y_res = sampler.fit_sample(X, y)
         assert_allclose(X_res_pd, X_res)

imblearn/utils/estimator_checks.py

@@ -3,11 +3,13 @@
 
 import codecs
 import os
-import sys
 
 from setuptools import find_packages, setup
-from imblearn import __version__
 
+# get __version__ from _version.py
+ver_file = os.path.join('imblearn', '_version.py')
+with open(ver_file) as f:
+    exec(f.read())
 
 DISTNAME = 'imbalanced-learn'
 DESCRIPTION = 'Toolbox for imbalanced dataset in machine learning.'