Merge pull request #321 from openml/enable_doctests

enable doctests on travis-ci; fix doctests

Author: mfeurer

.travis.yml

@@ -20,6 +20,7 @@ env:
   - DISTRIB="conda" PYTHON_VERSION="3.5" SKLEARN_VERSION="0.18.2"
   - DISTRIB="conda" PYTHON_VERSION="3.6" COVERAGE="true" SKLEARN_VERSION="0.18.2"
   - DISTRIB="conda" PYTHON_VERSION="3.6" EXAMPLES="true" SKLEARN_VERSION="0.18.2"
+  - DISTRIB="conda" PYTHON_VERSION="3.6" DOCTEST="true" SKLEARN_VERSION="0.18.2"
 
 install: source ci_scripts/install.sh
 script: bash ci_scripts/test.sh

ci_scripts/install.sh

@@ -26,13 +26,16 @@ popd
 # provided versions
 conda create -n testenv --yes python=$PYTHON_VERSION pip
 source activate testenv
-pip install nose numpy scipy cython scikit-learn==$SKLEARN_VERSION oslo.concurrency
+pip install nose numpy scipy cython scikit-learn==$SKLEARN_VERSION \
+    oslo.concurrency
 
 if [[ "$EXAMPLES" == "true" ]]; then
     pip install matplotlib jupyter notebook nbconvert nbformat jupyter_client \
         ipython ipykernel pandas seaborn
 fi
-
+if [[ "$DOCTEST" == "true" ]]; then
+    pip install pandas sphinx_bootstrap_theme
+fi
 if [[ "$COVERAGE" == "true" ]]; then
     pip install codecov
 fi

ci_scripts/test.sh

@@ -6,11 +6,14 @@ mkdir -p $TEST_DIR
 
 cwd=`pwd`
 test_dir=$cwd/tests
+doctest_dir=$cwd/doc
 
 cd $TEST_DIR
 
 if [[ "$EXAMPLES" == "true" ]]; then
     nosetests -sv $test_dir/test_examples/
+elif [[ "$DOCTEST" == "true" ]]; then
+    python -m doctest $doctest_dir/usage.rst
 elif [[ "$COVERAGE" == "true" ]]; then
     nosetests --processes=4 --process-timeout=600 -sv --ignore-files="test_OpenMLDemo\.py" --with-coverage --cover-package=$MODULE $test_dir
 else

doc/usage.rst

@@ -83,7 +83,7 @@ def __init__(self, dataset_id=None, name=None, version=None, description=None,
                                             xmlfeature['oml:name'],
                                             xmlfeature['oml:data_type'],
                                             None, #todo add nominal values (currently not in database)
-                                            int(xmlfeature['oml:number_of_missing_values']))
+                                            int(xmlfeature.get('oml:number_of_missing_values', 0)))
                 if idx != feature.index:
                     raise ValueError('Data features not provided in right order')
                 self.features[feature.index] = feature

openml/datasets/dataset.py

@@ -150,12 +150,14 @@ def list_datasets(offset=None, size=None, tag=None):
 
     Returns
     -------
-    datasets : list of dicts
-        A list of datasets having the given tag (if applicable).
+    datasets : dict of dicts
+        A mapping from dataset ID to dict.
 
         Every dataset is represented by a dictionary containing
         the following information:
         - dataset id
+        - name
+        - format
         - status
 
         If qualities are calculated for the dataset, some of

openml/datasets/functions.py

@@ -1,5 +1,5 @@
 from .task import OpenMLTask
 from .split import OpenMLSplit
-from .functions import (get_task, list_tasks)
+from .functions import (get_task, get_tasks, list_tasks)
 
-__all__ = ['OpenMLTask', 'get_task', 'list_tasks', 'OpenMLSplit']
+__all__ = ['OpenMLTask', 'get_task', 'get_tasks', 'list_tasks', 'OpenMLSplit']

openml/tasks/__init__.py

@@ -192,6 +192,26 @@ def _list_tasks(api_call):
     return tasks
 
 
+def get_tasks(task_ids):
+    """Download tasks.
+
+    This function iterates :meth:`openml.tasks.get_task`.
+
+    Parameters
+    ----------
+    task_ids : iterable
+        Integers representing task ids.
+
+    Returns
+    -------
+    list
+    """
+    tasks = []
+    for task_id in task_ids:
+        tasks.append(get_task(task_id))
+    return tasks
+
+
 def get_task(task_id):
     """Download the OpenML task for a given task ID.