Update module api to support gpu and batch_size config

demo/lac/lac_demo.py

@@ -16,7 +16,7 @@ if __name__ == "__main__":
     inputs = {"text": test_text}
 
     # execute predict and print the result
-    results = lac.lexical_analysis(data=inputs)
+    results = lac.lexical_analysis(data=inputs, use_gpu=True, batch_size=10)
     for result in results:
         if six.PY2:
             print(

paddlehub/commands/run.py

@@ -21,6 +21,7 @@ import argparse
 import json
 import os
 import sys
+import ast
 
 import six
 import pandas
@@ -80,6 +81,18 @@ class RunCommand(BaseCommand):
                 default=config['default'],
                 help=config['help'])
 
+        self.arg_config_group.add_argument(
+            '--use_gpu',
+            type=ast.literal_eval,
+            default=False,
+            help="whether use GPU for prediction")
+
+        self.arg_config_group.add_argument(
+            '--batch_size',
+            type=int,
+            default=1,
+            help="batch size for prediction")
+
         self.arg_config_group.add_argument(
             '--config',
             type=str,
@@ -224,7 +237,11 @@ class RunCommand(BaseCommand):
             return False
 
         results = self.module(
-            sign_name=self.module.default_signature.name, data=data, **config)
+            sign_name=self.module.default_signature.name,
+            data=data,
+            use_gpu=self.args.use_gpu,
+            batch_size=self.args.batch_size,
+            **config)
 
         if six.PY2:
             try:

paddlehub/module/module.py

@@ -434,7 +434,7 @@ class Module(object):
         for key, value in self.extra_info.items():
             utils.from_pyobj_to_module_attr(value, extra_info.map.data[key])
 
-    def __call__(self, sign_name, data, **kwargs):
+    def __call__(self, sign_name, data, use_gpu=False, batch_size=1, **kwargs):
         self.check_processor()
 
         def _get_reader_and_feeder(data_format, data, place):
@@ -463,15 +463,13 @@ class Module(object):
         with fluid.program_guard(program):
             result = []
             index = 0
-            if "PADDLEHUB_CUDA_ENABLE" in os.environ:
-                place = fluid.CUDAPlace(0)
-            else:
-                place = fluid.CPUPlace()
+            try:
+                _places = os.environ["CUDA_VISIBLE_DEVICES"]
+                int(_places[0])
+            except:
+                use_gpu = False
 
-            if "PADDLEHUB_BATCH_SIZE" in os.environ:
-                batch_size = os.environ["PADDLEHUB_BATCH_SIZE"]
-            else:
-                batch_size = 1
+            place = fluid.CUDAPlace(0) if use_gpu else fluid.CPUPlace()
 
             exe = fluid.Executor(place=place)
             data = self.processor.preprocess(