ADD class_name and custom_name fields

Author: mfeurer

openml/flows/flow.py

@@ -52,6 +52,11 @@ class OpenMLFlow(object):
         A list of dependencies necessary to run the flow. This field should
         contain all libraries the flow depends on. To allow reproducibility
         it should also specify the exact version numbers.
+    class_name : str
+        The development language name of the class which is described by this
+        flow.
+    custom_name : str
+        Custom name of the flow given by the owner.
     binary_url : str, optional
         Url from which the binary can be downloaded. Added by the server.
         Ignored when uploaded manually. Will not be used by the python API
@@ -75,7 +80,8 @@ class OpenMLFlow(object):
 
     def __init__(self, name, description, model, components, parameters,
                  parameters_meta_info, external_version, tags, language,
-                 dependencies, binary_url=None, binary_format=None,
+                 dependencies, class_name=None, custom_name=None,
+                 binary_url=None, binary_format=None,
                  binary_md5=None, uploader=None, upload_date=None,
                  flow_id=None, version=None):
         self.name = name
@@ -93,6 +99,7 @@ def __init__(self, name, description, model, components, parameters,
         self.components = components
         self.parameters = parameters
         self.parameters_meta_info = parameters_meta_info
+        self.class_name = class_name
 
         keys_parameters = set(parameters.keys())
         keys_parameters_meta_info = set(parameters_meta_info.keys())
@@ -110,6 +117,7 @@ def __init__(self, name, description, model, components, parameters,
         self.external_version = external_version
         self.uploader = uploader
 
+        self.custom_name = custom_name
         self.tags = tags if tags is not None else []
         self.binary_url = binary_url
         self.binary_format = binary_format
@@ -164,9 +172,9 @@ def _to_dict(self):
             if getattr(self, required) is None:
                 raise ValueError("self.{} is required but None".format(
                     required))
-        for attribute in ["uploader", "name", "version", "external_version",
-                          "description", "upload_date", "language",
-                          "dependencies"]:
+        for attribute in ["uploader", "name", "custom_name", "class_name",
+                          "version", "external_version", "description",
+                          "upload_date", "language", "dependencies"]:
             _add_if_nonempty(flow_dict, 'oml:{}'.format(attribute),
                              getattr(self, attribute))
 
@@ -248,7 +256,7 @@ def _from_dict(cls, xml_dict):
         # non-mandatory parts in the xml file
         for key in ['uploader', 'description', 'upload_date', 'language',
                     'dependencies', 'version', 'binary_url', 'binary_format',
-                    'binary_md5']:
+                    'binary_md5', 'class_name', 'custom_name']:
             arguments[key] = dic.get("oml:" + key)
 
         # has to be converted to an int if present and cannot parsed in the

openml/flows/sklearn_converter.py

@@ -244,7 +244,7 @@ def _serialize_model(model):
     # Create a flow name, which contains all components in brackets, for
     # example RandomizedSearchCV(Pipeline(StandardScaler,AdaBoostClassifier(DecisionTreeClassifier)),StandardScaler,AdaBoostClassifier(DecisionTreeClassifier))
     # TODO the name above is apparently wrong, I need to test and check this
-    name = model.__module__ + "." + model.__class__.__name__
+    class_name = model.__module__ + "." + model.__class__.__name__
 
     # will be part of the name (in brackets)
     sub_components_names = ""
@@ -256,10 +256,13 @@ def _serialize_model(model):
 
     if sub_components_names:
         # slice operation on string in order to get rid of leading comma
-        name = '%s(%s)' % (name, sub_components_names[1:])
+        name = '%s(%s)' % (class_name, sub_components_names[1:])
+    else:
+        name = class_name
 
     external_version = _get_external_version_info()
     flow = OpenMLFlow(name=name,
+                      class_name=class_name,
                       description='Automatically created sub-component.',
                       model=model,
                       components=sub_components,
@@ -276,12 +279,7 @@ def _serialize_model(model):
 
 def _deserialize_model(flow, **kwargs):
 
-    model_name = flow.name
-    # Remove everything after the first bracket, it is not necessary for
-    # creating the current flow
-    pos = model_name.find('(')
-    if pos >= 0:
-        model_name = model_name[:pos]
+    model_name = flow.class_name
 
     parameters = flow.parameters
     components = flow.components