Merge pull request #280 from openml/setuplist

Setuplist

Author: janvanrijn

openml/flows/sklearn_converter.py

@@ -533,7 +533,7 @@ def _serialize_cross_validator(o):
     for key in args:
         # We need deprecation warnings to always be on in order to
         # catch deprecated param values.
-        # This is set in utils/__init__.py.py.py but it gets overwritten
+        # This is set in utils/__init__.py but it gets overwritten
         # when running under python3 somehow.
         warnings.simplefilter("always", DeprecationWarning)
         try:

openml/setups/__init__.py

@@ -1,4 +1,4 @@
 from .setup import OpenMLSetup
-from .functions import get_setup, setup_list, setup_exists, initialize_model
+from .functions import get_setup, list_setups, setup_exists, initialize_model
 
-__all__ = ['get_setup', 'setup_list', 'setup_exists', 'initialize_model']
+__all__ = ['get_setup', 'list_setups', 'setup_exists', 'initialize_model']

openml/setups/functions.py

@@ -4,7 +4,7 @@
 import xmltodict
 
 from .setup import OpenMLSetup, OpenMLParameter
-from openml.flows import sklearn_to_flow, flow_exists
+from openml.flows import flow_exists
 
 
 def setup_exists(flow, model=None):
@@ -74,7 +74,7 @@ def get_setup(setup_id):
     return _create_setup_from_xml(result_dict)
 
 
-def setup_list(flow=None, tag=None, offset=None, size=None):
+def list_setups(flow=None, tag=None, setup=None, offset=None, size=None):
     """List all setups matching all of the given filters.
 
         Perform API call `/setup/list/{filters}
@@ -85,6 +85,8 @@ def setup_list(flow=None, tag=None, offset=None, size=None):
 
         tag : str, optional
 
+        setup : list(int), optional
+
         offset : int, optional
 
         size : int, optional
@@ -100,6 +102,8 @@ def setup_list(flow=None, tag=None, offset=None, size=None):
         api_call += "/offset/%d" % int(offset)
     if size is not None:
         api_call += "/limit/%d" % int(size)
+    if size is not None:
+        api_call += "/setup/%s" % ','.join([str(int(i)) for i in setup])
     if flow is not None:
         api_call += "/flow/%s" % flow
     if tag is not None:
@@ -120,11 +124,11 @@ def _list_setups(api_call):
                          % str(setups_dict))
     elif '@xmlns:oml' not in setups_dict['oml:setups']:
         raise ValueError('Error in return XML, does not contain '
-                         '"oml:runs"/@xmlns:oml: %s'
+                         '"oml:setups"/@xmlns:oml: %s'
                          % str(setups_dict))
     elif setups_dict['oml:setups']['@xmlns:oml'] != 'http://openml.org/openml':
         raise ValueError('Error in return XML, value of  '
-                         '"oml:runs"/@xmlns:oml is not '
+                         '"oml:seyups"/@xmlns:oml is not '
                          '"http://openml.org/openml": %s'
                          % str(setups_dict))
 

tests/test_flows/test_flow.py

@@ -239,17 +239,26 @@ def get_sentinel():
     def test_existing_flow_exists(self):
         # create a flow
         nb = sklearn.naive_bayes.GaussianNB()
-        flow = openml.flows.sklearn_to_flow(nb)
-        flow, _ = self._add_sentinel_to_flow_name(flow, None)
-        #publish the flow
-        flow = flow.publish()
-        #redownload the flow
-        flow = openml.flows.get_flow(flow.flow_id)
-
-        # check if flow exists can find it
-        flow = openml.flows.get_flow(flow.flow_id)
-        downloaded_flow_id = openml.flows.flow_exists(flow.name, flow.external_version)
-        self.assertEquals(downloaded_flow_id, flow.flow_id)
+
+        steps = [('imputation', sklearn.preprocessing.Imputer(strategy='median')),
+                 ('hotencoding', sklearn.preprocessing.OneHotEncoder(sparse=False,
+                                                                     handle_unknown='ignore')),
+                 ('variencethreshold', sklearn.feature_selection.VarianceThreshold()),
+                 ('classifier', sklearn.tree.DecisionTreeClassifier())]
+        complicated = sklearn.pipeline.Pipeline(steps=steps)
+
+        for classifier in [nb, complicated]:
+            flow = openml.flows.sklearn_to_flow(classifier)
+            flow, _ = self._add_sentinel_to_flow_name(flow, None)
+            #publish the flow
+            flow = flow.publish()
+            #redownload the flow
+            flow = openml.flows.get_flow(flow.flow_id)
+
+            # check if flow exists can find it
+            flow = openml.flows.get_flow(flow.flow_id)
+            downloaded_flow_id = openml.flows.flow_exists(flow.name, flow.external_version)
+            self.assertEquals(downloaded_flow_id, flow.flow_id)
 
     def test_sklearn_to_upload_to_flow(self):
         iris = sklearn.datasets.load_iris()

tests/test_setups/test_setup_functions.py

@@ -47,7 +47,6 @@ def get_params(self, deep=True):
         return {}
 
 
-
 class TestRun(TestBase):
 
     def test_nonexisting_setup_exists(self):
@@ -118,15 +117,27 @@ def test_get_setup(self):
             else:
                 self.assertEquals(len(current.parameters), num_params[idx])
 
-
     def test_setup_list_filter_flow(self):
-        # TODO: please remove for better test
-        # openml.config.server = self.production_server
+        openml.config.server = self.production_server
 
-        flow_id = 31 # TODO please change
+        flow_id = 5873
 
-        setups = openml.setups.setup_list(flow=31)
+        setups = openml.setups.list_setups(flow=flow_id)
 
         self.assertGreater(len(setups), 0) # TODO: please adjust 0
         for setup_id in setups.keys():
             self.assertEquals(setups[setup_id].flow_id, flow_id)
+
+    def test_setuplist_offset(self):
+        # TODO: remove after pull on live for better testing
+        # openml.config.server = self.production_server
+
+        size = 100
+        setups = openml.setups.list_setups(offset=0, size=size)
+        self.assertEquals(len(setups), size)
+        setups2 = openml.setups.list_setups(offset=size, size=size)
+        self.assertEquals(len(setups), size)
+
+        all = set(setups.keys()).union(setups2.keys())
+
+        self.assertEqual(len(all), size * 2)