Merge pull request #91 from amueller/rename_private

make functions that should be private private, rename stuff

Author: amueller

doc/api.rst

@@ -29,18 +29,13 @@ Dataset Functions
    :toctree: generated/
    :template: function.rst
 
-    datasets_active
-    download_dataset_arff
-    download_dataset_description
-    download_dataset_features
-    download_dataset_qualities
-    download_dataset
-    download_datasets
-    get_cached_datasets
-    get_list_of_cached_datasets
-    get_dataset_list
-    get_cached_dataset
-    get_dataset_list
+    check_datasets_active
+    get_dataset_description
+    get_dataset_features
+    get_dataset_qualities
+    get_dataset
+    get_datasets
+    list_datasets
 
 Run Functions
 --------------
@@ -50,12 +45,8 @@ Run Functions
    :toctree: generated/
    :template: function.rst
 
-   construct_description_dictionary
-   create_setup_string
-   get_version_information
-   openml_run
-   download_run
-   get_cached_run
+   run_task
+   get_run
 
 Task Functions
 --------------
@@ -65,10 +56,8 @@ Task Functions
    :toctree: generated/
    :template: function.rst
 
-    download_task
-    get_task_list
-    get_cached_splits
-    get_cached_split
+    get_task
+    list_tasks
 
 Flow Functions
 --------------
@@ -78,4 +67,3 @@ Flow Functions
    :toctree: generated/
    :template: function.rst
 
-    check_flow_exists

doc/progress.rst

@@ -61,10 +61,10 @@ Convenience Functions
 =============================================== =========== ====== =============== ========== =====================
 Method                                          implemented tested properly tested loads json proper error handling
 =============================================== =========== ====== =============== ========== =====================
-get_cached_split                                yes
-get_cached_splits                               yes
-get_cached_dataset                              yes         yes
-get_cached_datasets                             yes         yes
+_get_cached_split                               yes
+_get_cached_splits                              yes
+_get_cached_dataset                             yes         yes
+_get_cached_datasets                            yes         yes
 get_cached_task                                 yes
 get_cached_tasks                                yes
 =============================================== =========== ====== =============== ========== =====================

doc/usage.rst

@@ -33,7 +33,7 @@ Working with datasets
 .. code:: python
 
     >>> dataset_id = 31
-    >>> dataset = connector.download_dataset(dataset_id)
+    >>> dataset = connector.get_dataset(dataset_id)
 
 Attributes of the dataset are stored as member variables:
 

examples/OpenMLDemo.ipynb

@@ -126,7 +126,7 @@
     }
    ],
    "source": [
-    "datasets = openml.datasets.get_dataset_list(connector)\n",
+    "datasets = openml.datasets.list_datasets(connector)\n",
     "\n",
     "data = pd.DataFrame(datasets)\n",
     "print(\"First 10 of %s datasets...\" % len(datasets))\n",
@@ -236,7 +236,7 @@
     }
    ],
    "source": [
-    "dataset = openml.datasets.download_dataset(connector, 61)\n",
+    "dataset = openml.datasets.get_dataset(connector, 61)\n",
     "\n",
     "print(\"This is dataset '%s', the target feature is called '%s'\" % (dataset.name, dataset.default_target_attribute))\n",
     "print(\"URL: %s\" % dataset.url)\n",
@@ -394,7 +394,7 @@
     }
    ],
    "source": [
-    "dataset = openml.datasets.download_dataset(connector, 61)\n",
+    "dataset = openml.datasets.get_dataset(connector, 61)\n",
     "X, y = dataset.get_dataset(target=dataset.default_target_attribute)\n",
     "clf = ensemble.RandomForestClassifier()\n",
     "clf.fit(X, y)"
@@ -532,7 +532,7 @@
     }
    ],
    "source": [
-    "task_list = openml.tasks.get_task_list(connector)\n",
+    "task_list = openml.tasks.list_tasks(connector)\n",
     "\n",
     "tasks = pd.DataFrame(task_list)\n",
     "print(\"First 5 of %s tasks:\" % len(tasks))\n",
@@ -563,7 +563,7 @@
     }
    ],
    "source": [
-    "task = openml.tasks.download_task(connector, 10)\n",
+    "task = openml.tasks.get_task(connector, 10)\n",
     "print(task)"
    ]
   },
@@ -598,10 +598,10 @@
     }
    ],
    "source": [
-    "from openml.runs import openml_run\n",
+    "from openml.runs import run_task\n",
     "\n",
     "clf = ensemble.RandomForestClassifier()\n",
-    "run = openml_run(connector, task, clf)\n",
+    "run = run_task(connector, task, clf)\n",
     "print(\"RandomForest has run on the task.\")"
    ]
   },

examples/sklearn/openml_run_example.py

@@ -1,10 +1,10 @@
 from openml.apiconnector import APIConnector
-from openml.autorun import openml_run
+from openml.autorun import run_task
 from sklearn import ensemble
 import xmltodict
 import os
 """
-An example of an automated machine learning experiment using openml_run
+An example of an automated machine learning experiment using run_task
 """
 
 key_file_path = "apikey.txt"
@@ -15,9 +15,9 @@
 
 clf = ensemble.RandomForestClassifier()
 connector = APIConnector(apikey = key)
-task = connector.download_task(task_id)
+task = connector.get_task(task_id)
 
-prediction_path, description_path = openml_run(task, clf)
+prediction_path, description_path = run_task(task, clf)
 
 prediction_abspath = os.path.abspath(prediction_path)
 description_abspath = os.path.abspath(description_path)

openml/datasets/__init__.py

@@ -1,13 +1,10 @@
-from .functions import (get_list_of_cached_datasets, get_cached_datasets,
-                        get_cached_dataset, get_dataset_list, datasets_active,
-                        download_datasets, download_dataset,
-                        download_dataset_description, download_dataset_arff,
-                        download_dataset_features, download_dataset_qualities)
+from .functions import (list_datasets, check_datasets_active,
+                        get_datasets, get_dataset,
+                        get_dataset_description,
+                        get_dataset_features, get_dataset_qualities)
 from .dataset import OpenMLDataset
 
-__all__ = ['datasets_active', 'download_dataset', 'download_datasets',
-           'download_datasets_arf', 'download_dataset_features',
-           'download_dataset_qualities', 'get_cached_datasets',
-           'OpenMLDataset', 'get_list_of_cached_datasets', 'get_dataset_list',
-           'get_cached_dataset', 'download_dataset_description',
-           'download_dataset_arff', 'get_dataset_list']
+__all__ = ['check_datasets_active', 'get_dataset', 'get_datasets',
+           'get_datasets_arf', 'get_dataset_features',
+           'get_dataset_qualities', 'OpenMLDataset', 'list_datasets',
+           'get_dataset_description', 'list_datasets']

openml/datasets/functions.py

@@ -15,7 +15,7 @@
 ############################################################################
 # Local getters/accessors to the cache directory
 
-def get_list_of_cached_datasets(api_connector):
+def _list_cached_datasets(api_connector):
     """Return list with ids of all cached datasets"""
     datasets = []
 
@@ -46,20 +46,20 @@ def get_list_of_cached_datasets(api_connector):
     return datasets
 
 
-def get_cached_datasets(api_connector):
+def _get_cached_datasets(api_connector):
     """Searches for all OpenML datasets in the OpenML cache dir.
 
     Return a dictionary which maps dataset ids to dataset objects"""
-    dataset_list = get_list_of_cached_datasets(api_connector)
+    dataset_list = _list_cached_datasets(api_connector)
     datasets = OrderedDict()
 
     for did in dataset_list:
-        datasets[did] = get_cached_dataset(api_connector, did)
+        datasets[did] = _get_cached_dataset(api_connector, did)
 
     return datasets
 
 
-def get_cached_dataset(api_connector, did):
+def _get_cached_dataset(api_connector, did):
     # This code is slow...replace it with new API calls
     description = _get_cached_dataset_description(api_connector, did)
     arff_file = _get_cached_dataset_arff(api_connector, did)
@@ -103,7 +103,7 @@ def _get_cached_dataset_arff(api_connector, did):
                                "cached" % did)
 
 
-def get_dataset_list(api_connector):
+def list_datasets(api_connector):
     """Return a list of all dataset which are on OpenML.
 
     Returns
@@ -144,7 +144,7 @@ def get_dataset_list(api_connector):
     return datasets
 
 
-def datasets_active(api_connector, dids):
+def check_datasets_active(api_connector, dids):
     """Check if the dataset ids provided are active.
 
     Parameters
@@ -158,7 +158,7 @@ def datasets_active(api_connector, dids):
         A dictionary with items {did: active}, where active is a boolean. It
         is set to True if the dataset is active.
     """
-    dataset_list = get_dataset_list(api_connector)
+    dataset_list = list_datasets(api_connector)
     dids = sorted(dids)
     active = {}
 
@@ -171,7 +171,7 @@ def datasets_active(api_connector, dids):
         dataset_list_idx = idx
 
 
-def download_datasets(api_connector, dids):
+def get_datasets(api_connector, dids):
     """Download datasets.
 
     Parameters
@@ -186,16 +186,16 @@ def download_datasets(api_connector, dids):
 
     Notes
     -----
-    Uses :func:`download_dataset` internally. Please read
+    Uses :func:`get_dataset` internally. Please read
     the documentation of this.
     """
     datasets = []
     for did in dids:
-        datasets.append(download_dataset(api_connector, did))
+        datasets.append(get_dataset(api_connector, did))
     return datasets
 
 
-def download_dataset(api_connector, did):
+def get_dataset(api_connector, did):
     """Download a dataset.
 
     TODO: explain caching!
@@ -215,14 +215,14 @@ def download_dataset(api_connector, did):
         raise ValueError("Dataset ID is neither an Integer nor can be "
                          "cast to an Integer.")
 
-    description = download_dataset_description(api_connector, did)
-    arff_file = download_dataset_arff(api_connector, did, description=description)
+    description = get_dataset_description(api_connector, did)
+    arff_file = _get_dataset_arff(api_connector, did, description=description)
 
     dataset = _create_dataset_from_description(description, arff_file)
     return dataset
 
 
-def download_dataset_description(api_connector, did):
+def get_dataset_description(api_connector, did):
     # TODO implement a cache for this that invalidates itself after some
     # time
     # This can be saved on disk, but cannot be cached properly, because
@@ -260,7 +260,7 @@ def download_dataset_description(api_connector, did):
     return description
 
 
-def download_dataset_arff(api_connector, did, description=None):
+def _get_dataset_arff(api_connector, did, description=None):
     did_cache_dir = _create_dataset_cache_dir(api_connector, did)
     output_file = os.path.join(did_cache_dir, "dataset.arff")
 
@@ -274,7 +274,7 @@ def download_dataset_arff(api_connector, did, description=None):
         pass
 
     if description is None:
-        description = download_dataset_description(api_connector, did)
+        description = get_dataset_description(api_connector, did)
     url = description['oml:url']
     return_code, arff_string = api_connector._read_url(url)
     # TODO: it is inefficient to load the dataset in memory prior to
@@ -286,7 +286,7 @@ def download_dataset_arff(api_connector, did, description=None):
     return output_file
 
 
-def download_dataset_features(api_connector, did):
+def get_dataset_features(api_connector, did):
     did_cache_dir = _create_dataset_cache_dir(api_connector, did)
     features_file = os.path.join(did_cache_dir, "features.xml")
 
@@ -316,7 +316,7 @@ def download_dataset_features(api_connector, did):
     return features
 
 
-def download_dataset_qualities(api_connector, did):
+def get_dataset_qualities(api_connector, did):
     # Dataset qualities are subject to change and must be fetched every time
     did_cache_dir = _create_dataset_cache_dir(api_connector, did)
     qualities_file = os.path.join(did_cache_dir, "qualities.xml")

openml/flows/__init__.py

@@ -1,3 +1,3 @@
-from .flow import OpenMLFlow, check_flow_exists
+from .flow import OpenMLFlow
 
-__all__ = ['OpenMLFlow', 'check_flow_exists']
+__all__ = ['OpenMLFlow']

openml/flows/flow.py

@@ -5,7 +5,7 @@
 
 class OpenMLFlow(object):
     def __init__(self, model, id=None, uploader=None,
-                 description='Flow generated by openml_run', creator=None,
+                 description='Flow generated by run_task', creator=None,
                  contributor=None, tag=None):
         self.id = id
         self.upoader = uploader
@@ -66,7 +66,7 @@ def ensure_flow_exists(self, connector):
         """
         import sklearn
         flow_version = 'Tsklearn_' + sklearn.__version__
-        _, _, flow_id = check_flow_exists(connector, self.name, flow_version)
+        _, _, flow_id = _check_flow_exists(connector, self.name, flow_version)
 
         if int(flow_id) == -1:
             return_code, response_xml = self.publish(connector)
@@ -78,7 +78,7 @@ def ensure_flow_exists(self, connector):
         return int(flow_id)
 
 
-def check_flow_exists(api_connector, name, version):
+def _check_flow_exists(api_connector, name, version):
     """Retrieves the flow id of the flow uniquely identified by name+version.
 
     Returns flow id if such a flow exists,

openml/runs/__init__.py

@@ -1,8 +1,4 @@
 from .run import OpenMLRun
-from .run import (construct_description_dictionary, create_setup_string,
-                  get_version_information, openml_run, download_run,
-                  get_cached_run)
+from .run import run_task, get_run
 
-__all__ = ['OpenMLRun', 'construct_description_dictionary',
-           'create_setup_string', 'get_version_information', 'openml_run',
-           'download_run', 'get_cached_run']
+__all__ = ['OpenMLRun', 'run_task', 'get_run']