Merge pull request #576 from shankari/create_tree_pipeline

Minor fixes to inference before redoing it completely

emission/analysis/classification/inference/mode/seed/pipeline.py

@@ -313,10 +313,12 @@ def selectFeatureIndicesStep(self):
     logging.debug("location features = %s" % LocationFeatureIndices)
     logging.debug("time features = %s" % TimeFeatureIndices)
     logging.debug("bus train features = %s" % BusTrainFeatureIndices)
+    retIndices = genericFeatureIndices
     if eac.get_config()["classification.inference.mode.useAdvancedFeatureIndices"]:
-        return genericFeatureIndices + AdvancedFeatureIndices + BusTrainFeatureIndices
-    else:
-        return genericFeatureIndices + BusTrainFeatureIndices
+        retIndices = retIndices + AdvancedFeatureIndices
+    if eac.get_config()["classification.inference.mode.useBusTrainFeatureIndices"]:
+        retIndices = retIndices + BusTrainFeatureIndices
+    return retIndices
 
   def buildModelStep(self):
     from sklearn import ensemble

emission/analysis/section_features.py

@@ -102,9 +102,9 @@ def calHCR(section_entry):
         return 0
 
 def calSR(section):
-    if 'speeds' not in section:
+    if 'speeds' not in section.data:
         return 0
-    speeds = section.speeds
+    speeds = section.data["speeds"]
     if len(speeds) < 2:
         return 0
     else:
@@ -114,7 +114,7 @@ def calSR(section):
             if currVelocity != None and currVelocity <= 0.75:
                 stopNum += 1
 
-        sectionDist = section.distance
+        sectionDist = section.data.distance
         if sectionDist != None and sectionDist != 0:
             return stopNum/sectionDist
         else:

emission/storage/decorations/analysis_timeseries_queries.py

@@ -83,7 +83,7 @@ def get_timeseries_for_user(user_id):
 def get_time_query_for_trip_like(key, trip_like_id):
     """
     Returns the query that returns all the points associated with this
-    trip-like (raw trip, cleaned trip, raw section) 
+    trip-like (examples of trip-like objects are: raw trip, cleaned trip, raw section) 
     """
     trip = get_object(key, trip_like_id)
     return get_time_query_for_trip_like_object(trip)