MNT CI

.travis.yml

@@ -29,13 +29,13 @@ env:
   matrix:
     # This environment tests the using anaconda
     - DISTRIB="conda" PYTHON_VERSION="2.7" COVERAGE="true"
-      NUMPY_VERSION="1.9.3" SCIPY_VERSION="0.16.0"
+      NUMPY_VERSION="1.11.2" SCIPY_VERSION="0.18.1" SKLEARN_VERSION="0.18.1"
     - DISTRIB="conda" PYTHON_VERSION="3.5" COVERAGE="true"
-      NUMPY_VERSION="1.10.4" SCIPY_VERSION="0.17.0"
+      NUMPY_VERSION="1.11.2" SCIPY_VERSION="0.18.1" SKLEARN_VERSION="0.18.1"
       # flake8 linting on diff wrt common ancestor with upstream/master
     - RUN_FLAKE8="true" SKIP_TESTS="true"
       DISTRIB="conda" PYTHON_VERSION="3.5"
-      NUMPY_VERSION="1.10.4" SCIPY_VERSION="0.17.0"
+      NUMPY_VERSION="1.11.2" SCIPY_VERSION="0.18.1" SKLEARN_VERSION="0.18.1"
 
 install: source build_tools/travis/install.sh
 script: bash build_tools/travis/test_script.sh

appveyor.yml

@@ -24,7 +24,7 @@ install:
   - "python -c \"import struct; print(struct.calcsize('P') * 8)\""
 
   # Installed prebuilt dependencies from conda
-  - "conda install pip numpy scipy scikit-learn nose wheel matplotlib -y -q"
+  - "conda install pip numpy=1.11.2 scipy=0.18.1 scikit-learn=0.18.1 nose wheel matplotlib -y -q"
 
   # Install other nilearn dependencies
   - "pip install coverage nose-timer"

build_tools/travis/install.sh

@@ -24,20 +24,17 @@ popd
 
 # Configure the conda environment and put it in the path using the
 # provided versions
-conda create -n testenv --yes python=$PYTHON_VERSION pip nose \
+conda create -n testenv --yes python=$PYTHON_VERSION pip nose
     numpy=$NUMPY_VERSION scipy=$SCIPY_VERSION \
-    libgfortran nomkl flake8
-#source activate testenv
+    libgfortran nomkl
+source activate testenv
+
+# Install scikit-learn
+conda install --yes scikit-learn=$SKLEARN_VERSION
 
 # Install nose-timer via pip
 pip install nose-timer
 
-# Install libgfortran with conda
-conda install --yes libgfortran \
-    numpy=1.10.4 scipy=0.17.1 \
-    scikit-learn=0.17.1 \
-    six=1.10.0
-
 if [[ "$COVERAGE" == "true" ]]; then
     pip install coverage coveralls
 fi

imblearn/combine/smote_enn.py

@@ -128,7 +128,7 @@ class SMOTEENN(BaseBinarySampler):
     >>> sme = SMOTEENN(random_state=42)
     >>> X_res, y_res = sme.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({0: 900, 1: 865})
+    Resampled dataset shape Counter({0: 900, 1: 881})
 
     References
     ----------

imblearn/ensemble/balance_cascade.py

@@ -9,8 +9,7 @@
 from sklearn.neighbors import KNeighborsClassifier
 from sklearn.utils import check_random_state
 from sklearn.utils.validation import has_fit_parameter
-
-from six import string_types
+from sklearn.externals.six import string_types
 
 from ..base import BaseBinarySampler
 

imblearn/metrics/tests/test_score_objects.py

@@ -10,8 +10,8 @@
                               geometric_mean_score,
                               make_index_balanced_accuracy)
 # Get the version
-(major, minor, _) = sklearn.__version__.split('.')
-if int(minor) < 18:
+sk_version = sklearn.__version__
+if sk_version < '0.18':
     from sklearn.cross_validation import train_test_split
     from sklearn.grid_search import GridSearchCV
 else:

imblearn/over_sampling/adasyn.py

@@ -84,7 +84,7 @@ class ADASYN(BaseBinarySampler):
     >>> ada = ADASYN(random_state=42)
     >>> X_res, y_res = ada.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({0: 909, 1: 900})
+    Resampled dataset shape Counter({0: 904, 1: 900})
 
     References
     ----------

imblearn/pipeline.py

@@ -103,10 +103,10 @@ class Pipeline(pipeline.Pipeline):
     >>> print(classification_report(y_test, y_hat))
                  precision    recall  f1-score   support
     <BLANKLINE>
-              0       0.71      1.00      0.83        24
-              1       1.00      0.96      0.98       226
+              0       0.87      1.00      0.93        26
+              1       1.00      0.98      0.99       224
     <BLANKLINE>
-    avg / total       0.97      0.96      0.96       250
+    avg / total       0.99      0.98      0.98       250
     <BLANKLINE>
 
     """

imblearn/under_sampling/edited_nearest_neighbours.py

@@ -91,7 +91,7 @@ class EditedNearestNeighbours(BaseMulticlassSampler):
     >>> enn = EditedNearestNeighbours(random_state=42)
     >>> X_res, y_res = enn.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({1: 883, 0: 100})
+    Resampled dataset shape Counter({1: 887, 0: 100})
 
     References
     ----------
@@ -331,7 +331,7 @@ class RepeatedEditedNearestNeighbours(BaseMulticlassSampler):
     >>> renn = RepeatedEditedNearestNeighbours(random_state=42)
     >>> X_res, y_res = renn.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({1: 883, 0: 100})
+    Resampled dataset shape Counter({1: 887, 0: 100})
 
     References
     ----------
@@ -576,7 +576,7 @@ class AllKNN(BaseMulticlassSampler):
     >>> allknn = AllKNN(random_state=42)
     >>> X_res, y_res = allknn.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({1: 883, 0: 100})
+    Resampled dataset shape Counter({1: 887, 0: 100})
 
     References
     ----------

imblearn/under_sampling/instance_hardness_threshold.py

@@ -103,17 +103,16 @@ class InstanceHardnessThreshold(BaseBinarySampler):
 
     >>> from collections import Counter
     >>> from sklearn.datasets import make_classification
-    >>> from imblearn.under_sampling import \
-    RepeatedEditedNearestNeighbours # doctest: +NORMALIZE_WHITESPACE
+    >>> from imblearn.under_sampling import InstanceHardnessThreshold
     >>> X, y = make_classification(n_classes=2, class_sep=2,
     ... weights=[0.1, 0.9], n_informative=3, n_redundant=1, flip_y=0,
     ... n_features=20, n_clusters_per_class=1, n_samples=1000, random_state=10)
     >>> print('Original dataset shape {}'.format(Counter(y)))
     Original dataset shape Counter({1: 900, 0: 100})
-    >>> renn = RepeatedEditedNearestNeighbours(random_state=42)
-    >>> X_res, y_res = renn.fit_sample(X, y)
+    >>> iht = InstanceHardnessThreshold(random_state=42)
+    >>> X_res, y_res = iht.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({1: 883, 0: 100})
+    Resampled dataset shape Counter({1: 840, 0: 100})
 
     References
     ----------

imblearn/under_sampling/neighbourhood_cleaning_rule.py

@@ -77,7 +77,7 @@ class NeighbourhoodCleaningRule(BaseMulticlassSampler):
     >>> ncr = NeighbourhoodCleaningRule(random_state=42)
     >>> X_res, y_res = ncr.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({1: 884, 0: 100})
+    Resampled dataset shape Counter({1: 891, 0: 100})
 
     References
     ----------

imblearn/under_sampling/one_sided_selection.py

@@ -82,7 +82,7 @@ class OneSidedSelection(BaseBinarySampler):
     >>> oss = OneSidedSelection(random_state=42)
     >>> X_res, y_res = oss.fit_sample(X, y)
     >>> print('Resampled dataset shape {}'.format(Counter(y_res)))
-    Resampled dataset shape Counter({1: 595, 0: 100})
+    Resampled dataset shape Counter({1: 496, 0: 100})
 
     References
     ----------