Merge branch 'develop' into dataupload

Author: janvanrijn

doc/usage.rst

@@ -143,7 +143,7 @@ We can filter the list of tasks to only contain datasets with more than
     >>> filtered_tasks = tasks.query('NumberOfInstances > 500 and NumberOfInstances < 1000')
     >>> print(list(filtered_tasks.index))                               # doctest: +SKIP
     [2, 11, 15, 29, 37, 41, 49, 53, ..., 146597, 146600, 146605]
-    >>> print(len(filtered_tasks))
+    >>> print(len(filtered_tasks))                                      # doctest: +SKIP
     210
 
 Then, we can further restrict the tasks to all have the same resampling

openml/runs/functions.py

@@ -638,8 +638,11 @@ def get_run(run_id):
     run : OpenMLRun
         Run corresponding to ID, fetched from the server.
     """
-    run_file = os.path.join(config.get_cache_directory(), "runs",
-                            "run_%d.xml" % run_id)
+    run_dir = os.path.join(config.get_cache_directory(), "runs", str(run_id))
+    run_file = os.path.join(run_dir, "description.xml")
+
+    if not os.path.exists(run_dir):
+        os.makedirs(run_dir)
 
     try:
         return _get_cached_run(run_id)
@@ -667,7 +670,7 @@ def _create_run_from_xml(xml, from_server=True):
     run : OpenMLRun
         New run object representing run_xml.
     """
-
+    
     def obtain_field(xml_obj, fieldname, from_server, cast=None):
         # this function can be used to check whether a field is present in an object.
         # if it is not present, either returns None or throws an error (this is
@@ -694,7 +697,6 @@ def obtain_field(xml_obj, fieldname, from_server, cast=None):
     else:
         task_evaluation_measure = None
 
-
     flow_id = int(run['oml:flow_id'])
     flow_name = obtain_field(run, 'oml:flow_name', from_server)
     setup_id = obtain_field(run, 'oml:setup_id', from_server, cast=int)
@@ -872,10 +874,9 @@ def _create_trace_from_arff(arff_obj):
 def _get_cached_run(run_id):
     """Load a run from the cache."""
     cache_dir = config.get_cache_directory()
-    run_cache_dir = os.path.join(cache_dir, "runs")
+    run_cache_dir = os.path.join(cache_dir, "runs", str(run_id))
     try:
-        run_file = os.path.join(run_cache_dir,
-                                "run_%d.xml" % int(run_id))
+        run_file = os.path.join(run_cache_dir, "description.xml")
         with io.open(run_file, encoding='utf8') as fh:
             run = _create_run_from_xml(xml=fh.read())
         return run

openml/setups/functions.py

@@ -1,8 +1,11 @@
 from collections import OrderedDict
 
+import io
 import openml
+import os
 import xmltodict
 
+from .. import config
 from .setup import OpenMLSetup, OpenMLParameter
 from openml.flows import flow_exists
 
@@ -54,8 +57,23 @@ def setup_exists(flow, model=None):
         return False
 
 
+def _get_cached_setup(setup_id):
+    """Load a run from the cache."""
+    cache_dir = config.get_cache_directory()
+    setup_cache_dir = os.path.join(cache_dir, "setups", str(setup_id))
+    try:
+        setup_file = os.path.join(setup_cache_dir, "description.xml")
+        with io.open(setup_file, encoding='utf8') as fh:
+            setup_xml = xmltodict.parse(fh.read())
+            setup = _create_setup_from_xml(setup_xml)
+        return setup
+
+    except (OSError, IOError):
+        raise openml.exceptions.OpenMLCacheException("Setup file for setup id %d not cached" % setup_id)
+
+
 def get_setup(setup_id):
-    '''
+    """
      Downloads the setup (configuration) description from OpenML
      and returns a structured object
 
@@ -68,9 +86,22 @@ def get_setup(setup_id):
         -------
         OpenMLSetup
             an initialized openml setup object
-    '''
-    result = openml._api_calls._perform_api_call('/setup/%d' %setup_id)
-    result_dict = xmltodict.parse(result)
+    """
+    setup_dir = os.path.join(config.get_cache_directory(), "setups", str(setup_id))
+    setup_file = os.path.join(setup_dir, "description.xml")
+
+    if not os.path.exists(setup_dir):
+        os.makedirs(setup_dir)
+
+    try:
+        return _get_cached_setup(setup_id)
+
+    except (openml.exceptions.OpenMLCacheException):
+        setup_xml = openml._api_calls._perform_api_call('/setup/%d' % setup_id)
+        with io.open(setup_file, "w", encoding='utf8') as fh:
+            fh.write(setup_xml)
+
+    result_dict = xmltodict.parse(setup_xml)
     return _create_setup_from_xml(result_dict)
 
 
@@ -212,6 +243,7 @@ def _to_dict(flow_id, openml_parameter_settings):
 
     return xml
 
+
 def _create_setup_from_xml(result_dict):
     '''
      Turns an API xml result into a OpenMLSetup object

openml/utils.py

@@ -1,4 +1,6 @@
+import xmltodict
 import six
+from ._api_calls import _perform_api_call
 
 from openml.exceptions import OpenMLServerException
 
@@ -40,6 +42,55 @@ def extract_xml_tags(xml_tag_name, node, allow_none=True):
         else:
             raise ValueError("Could not find tag '%s' in node '%s'" %
                              (xml_tag_name, str(node)))
+
+
+def _tag_entity(entity_type, entity_id, tag, untag=False):
+    """Function that tags or untags a given entity on OpenML. As the OpenML
+       API tag functions all consist of the same format, this function covers
+       all entity types (currently: dataset, task, flow, setup, run). Could
+       be used in a partial to provide dataset_tag, dataset_untag, etc.
+
+        Parameters
+        ----------
+        entity_type : str
+            Name of the entity to tag (e.g., run, flow, data)
+
+        entity_id : int
+            OpenML id of the entity
+
+        tag : str
+            The tag
+
+        untag : bool
+            Set to true if needed to untag, rather than tag
+
+        Returns
+        -------
+        tags : list
+            List of tags that the entity is (still) tagged with
+        """
+    legal_entities = {'data', 'task', 'flow', 'setup', 'run'}
+    if entity_type not in legal_entities:
+        raise ValueError('Can\'t tag a %s' %entity_type)
+
+    uri = '%s/tag' %entity_type
+    main_tag = 'oml:%s_tag' %entity_type
+    if untag:
+        uri = '%s/untag' %entity_type
+        main_tag = 'oml:%s_untag' %entity_type
+
+
+    post_variables = {'%s_id'%entity_type: entity_id, 'tag': tag}
+    result_xml = _perform_api_call(uri, post_variables)
+
+    result = xmltodict.parse(result_xml, force_list={'oml:tag'})[main_tag]
+
+    if 'oml:tag' in result:
+        return result['oml:tag']
+    else:
+        # no tags, return empty list
+        return []
+
 
 def list_all(listing_call, batch_size=10000, *args, **filters):
     """Helper to handle paged listing requests.