MAINT maintenance updates

Author: mfeurer

ci_scripts/circle_install.sh

@@ -9,7 +9,7 @@ source activate testenv
 export CC=`which gcc`
 # install documentation building dependencies
 pip install --upgrade numpy
-pip install --upgrade matplotlib setuptools nose coverage sphinx==1.5.5 sphinx_bootstrap_theme numpydoc
+pip install --upgrade matplotlib setuptools nose coverage sphinx==1.5.5 sphinx_bootstrap_theme numpydoc sphinx_gallery
 # And finally, all other dependencies
 cat requirements.txt | xargs -n 1 -L 1 pip install
 

ci_scripts/test.sh

@@ -5,7 +5,7 @@ set -e
 mkdir -p $TEST_DIR
 
 cwd=`pwd`
-examples_dir=$cwd/example
+examples_dir=$cwd/examples
 test_dir=$cwd/test/
 
 cd $TEST_DIR

examples/example_eips.py

@@ -3,7 +3,8 @@
 EIPS
 ====
 
-Example description goes here.
+This example demonstrates the usage of a different acquisition function inside SMAC, namely
+`Expected Improvement per Second (EIPS) <https://papers.nips.cc/paper/4522-practical-bayesian-optimization-of-machine-learning-algorithms.html>_`.
 """
 
 import sklearn.model_selection

examples/example_random_search.py

@@ -7,6 +7,10 @@
 through SMAC, introduced `here <http://ml.informatik.uni-freiburg.de/papers/11-LION5-SMAC.pdf>`_.
 Additionally, it is possible to use `random search <http://www.jmlr.org/papers/v13/bergstra12a.html>`_
 instead of SMAC, as demonstrated in the example below.
+Furthermore, the example also demonstrates how to use
+`Random Online Aggressive Racing (ROAR) <http://ml.informatik.uni-freiburg.de/papers/11-LION5-SMAC.pdf>`_
+as yet another alternative optimizatino strategy.
+Both examples are intended to show how the optimization strategy in *auto-sklearn* can be adapted.
 """
 
 import sklearn.model_selection

test/test_automl/test_estimators.py

@@ -311,7 +311,7 @@ def test_multilabel(self):
         score = f1_macro(Y_test, predictions)
         self.assertGreaterEqual(score, 0.9)
         probs = automl.predict_proba(X_train)
-        self.assertAlmostEqual(np.mean(probs), 0.33, places=2)
+        self.assertAlmostEqual(np.mean(probs), 0.33, places=1)
 
     def test_binary(self):
         output = os.path.join(self.test_dir, '..', '.tmp_binary_fit')