Merge branch 'develop' into joaquinvanschoren-patch-1

Author: janvanrijn

openml/__version__.py

@@ -1,4 +1,4 @@
 """Version information."""
 
 # The following line *must* be the last in the module, exactly as formatted:
-__version__ = "0.5.0dev"
+__version__ = "0.6.0dev"

openml/runs/functions.py

@@ -221,6 +221,7 @@ def initialize_model_from_trace(run_id, repeat, fold, iteration=None):
     base_estimator.set_params(**current.get_parameters())
     return base_estimator
 
+
 def _run_exists(task_id, setup_id):
     '''
     Checks whether a task/setup combination is already present on the server.
@@ -244,6 +245,7 @@ def _run_exists(task_id, setup_id):
         assert(exception.code == 512)
         return False
 
+
 def _get_seeded_model(model, seed=None):
     '''Sets all the non-seeded components of a model with a seed.
        Models that are already seeded will maintain the seed. In
@@ -356,6 +358,7 @@ def _prediction_to_row(rep_no, fold_no, sample_no, row_id, correct_label,
     arff_line.append(correct_label)
     return arff_line
 
+
 # JvR: why is class labels a parameter? could be removed and taken from task object, right?
 def _run_task_get_arffcontent(model, task, class_labels):
     X, Y = task.get_X_and_y()
@@ -467,7 +470,6 @@ def _calculate_local_measure(sklearn_fn, openml_name):
            user_defined_measures_sample
 
 
-
 def _extract_arfftrace(model, rep_no, fold_no):
     if not isinstance(model, sklearn.model_selection._search.BaseSearchCV):
         raise ValueError('model should be instance of'\
@@ -490,6 +492,7 @@ def _extract_arfftrace(model, rep_no, fold_no):
         arff_tracecontent.append(arff_line)
     return arff_tracecontent
 
+
 def _extract_arfftrace_attributes(model):
     if not isinstance(model, sklearn.model_selection._search.BaseSearchCV):
         raise ValueError('model should be instance of'\
@@ -682,6 +685,7 @@ def _create_run_from_xml(xml):
                      sample_evaluations=sample_evaluations,
                      tags=tags)
 
+
 def _create_trace_from_description(xml):
     result_dict = xmltodict.parse(xml)['oml:trace']
 
@@ -714,6 +718,52 @@ def _create_trace_from_description(xml):
 
     return OpenMLRunTrace(run_id, trace)
 
+
+def _create_trace_from_arff(arff_obj):
+    """
+    Creates a trace file from arff obj (for example, generated by a local run)
+
+    Parameters
+    ----------
+    arff_obj : dict
+        LIAC arff obj, dict containing attributes, relation, data and description
+
+    Returns
+    -------
+    run : OpenMLRunTrace
+        Object containing None for run id and a dict containing the trace iterations
+    """
+    trace = dict()
+    attribute_idx = {att[0]: idx for idx, att in enumerate(arff_obj['attributes'])}
+    for required_attribute in ['repeat', 'fold', 'iteration', 'evaluation', 'selected']:
+        if required_attribute not in attribute_idx:
+            raise ValueError('arff misses required attribute: %s' %required_attribute)
+
+    for itt in arff_obj['data']:
+        repeat = int(itt[attribute_idx['repeat']])
+        fold = int(itt[attribute_idx['fold']])
+        iteration = int(itt[attribute_idx['iteration']])
+        evaluation = float(itt[attribute_idx['evaluation']])
+        selectedValue = itt[attribute_idx['selected']]
+        if selectedValue == 'true':
+            selected = True
+        elif selectedValue == 'false':
+            selected = False
+        else:
+            raise ValueError('expected {"true", "false"} value for selected field, received: %s' % selectedValue)
+
+        # TODO: if someone needs it, he can use the parameter
+        # fields to revive the setup_string as well
+        # However, this is usually done by the OpenML server
+        # and if we are going to duplicate this functionality
+        # it needs proper testing
+
+        current = OpenMLTraceIteration(repeat, fold, iteration, None, evaluation, selected)
+        trace[(repeat, fold, iteration)] = current
+
+    return OpenMLRunTrace(None, trace)
+
+
 def _get_cached_run(run_id):
     """Load a run from the cache."""
     cache_dir = config.get_cache_directory()