Merge pull request uwdata#4 from uwdata/master

minor fix: 1. allennlp can be run with gpu, 2. check the dict entries…

errudite/builts/attribute.py

@@ -492,7 +492,10 @@ def create_from_json(cls, raw: Dict[str, str]) -> 'Attribute':
         Attribute
             The re-created attribute.
         """
-        return Attribute(raw['name'], raw['description'], raw['cmd'])
+        return Attribute(
+            raw['name'] if "name" in raw else None, 
+            raw['description'] if "description" in raw else None, 
+            raw['cmd'] if "cmd" in raw else None)
 
     @staticmethod 
     def create(

errudite/builts/group.py

@@ -459,7 +459,10 @@ def create_from_json(cls, raw: Dict[str, str]) -> 'Group':
         Group
             The re-created group.
         """
-        return Group(raw['name'], raw['description'], raw['cmd'])
+        return Group(
+            raw['name'] if "name" in raw else None, 
+            raw['description'] if "description" in raw else None, 
+            raw['cmd'] if "cmd" in raw else None)
 
     @staticmethod 
     def create(

errudite/predictors/predictor.py

@@ -118,10 +118,10 @@ def create_from_json(cls, raw: Dict[str, str]) -> 'Predictor':
         """
         try:
             return Predictor.by_name(raw["model_class"])(
-                name=raw["name"], 
-                description=raw["description"],
-                model_path=raw["model_path"],
-                model_online_path=raw["model_online_path"])
+                name=raw["name"] if "name" in raw else None, 
+                description=raw["description"] if "description" in raw else None,
+                model_path=raw["model_path"] if "model_path" in raw else None,
+                model_online_path=raw["model_online_path"] if "model_online_path" in raw else None)
         except:
             raise
 

errudite/predictors/predictor_allennlp.py

@@ -4,6 +4,7 @@
 from allennlp.models.archival import load_archive
 from allennlp.predictors.predictor import Predictor as AllenPredictor
 from .predictor import Predictor
+import torch
 
 # bidaf-model-2017.08.31.tar.gz
 # bidaf-model-2017.09.15-charpad.tar.gz
@@ -37,10 +38,16 @@ def __init__(self, name: str,
         """
         model = None
         if model_path:
-            archive = load_archive(model_path)
+            if torch.cuda.is_available():
+                archive = load_archive(model_path, cuda_device=0)
+            else:    
+                archive = load_archive(model_path)
             model = AllenPredictor.from_archive(archive, model_type)
         elif model_online_path:
-            model = AllenPredictor.from_path(model_online_path, model_type)
+            if torch.cuda.is_available():
+	            model = AllenPredictor.from_path(model_online_path, model_type, cuda_device=0)
+	        else:
+	            model = AllenPredictor.from_path(model_online_path, model_type)
         self.predictor = model
         Predictor.__init__(self, name, description, model, ['accuracy'])