update demo

demo/image-classification/resources/module_info.yml

 name: Resnet50
 type: CV/classification
 author: paddlepaddle
-author-email: paddle-dev@baidu.com
+author_email: paddle-dev@baidu.com
+summary: "Resnet50 is a model used to image classfication, we trained this model on ImageNet-2012 dataset."
 version: 1.0.0

demo/image-classification/retrain.py

@@ -15,7 +15,7 @@ def train():
         img = input_dict[0]
         feature_map = output_dict[0]
 
-        config = hub.FinetuneConfig(
+        config = hub.RunConfig(
             use_cuda=True,
             num_epoch=10,
             batch_size=32,

demo/lac/resources/module_info.yml

@@ -2,4 +2,5 @@ name: lac
 type: nlp/lexical_analysis
 author: paddlepaddle
 author_email: paddle-dev@baidu.com
+summary: "'Lexical Analysis of Chinese', which abbreviated as LAC, is a model used to process lexical analysis. People can use LAC to process Chinese text segmentation, part-of-speech tagging and named entity recognition"
 version: 1.0.0

demo/senta/processor.py

@@ -52,14 +52,14 @@ class Processor(hub.BaseProcessor):
 
     def preprocess(self, sign_name, data_dict):
         result = {'text': []}
-        processed = self.lac.segment(data=data_dict)
+        processed = self.lac.lexical_analysis(data=data_dict)
         unk_id = len(self.word_dict)
         for index, data in enumerate(processed):
             result_i = {'processed': []}
             result_i['origin'] = data_dict['text'][index]
-            for result_dict in data:
-                if result_dict['word'] in self.word_dict:
-                    _index = self.word_dict[result_dict['word']]
+            for word in data['word']:
+                if word in self.word_dict:
+                    _index = self.word_dict[word]
                 else:
                     _index = unk_id
                 result_i['processed'].append(_index)

demo/senta/resources/module_info.yml

 name: senta
 type: nlp/sentiment_analysis
 author: paddlepaddle
-author-email: paddle-dev@baidu.com
+author_email: paddle-dev@baidu.com
+summary: "Senta is a model used to analyse sentiment tendency of Chinese sentences. We divide sentiment tendencies into three levels, score 2 means positive, score 1 means neuter, and score 0 means negative"
 version: 1.0.0

demo/ssd/resources/module_info.yml

@@ -2,4 +2,5 @@ name: ssd_mobilenet_v1_pascalvoc
 type: CV/object-detection
 author: paddlepaddle
 author_email: paddle-dev@baidu.com
+summary: "SSD(Single Shot MultiBox Detector) is a object detection model use to detect target category and location in a image picture. This model is trained with PASCAL VOC dataset, and therefore provides 20 categories of recognition capability, which mentioned below : aeroplane,bicycle,bird,boat,bottle,bus,car,cat,chair,cow,diningtable,dog,horse,motorbike,person,pottedplant,sheep,sofa,train,tvmonitor"
 version: 1.0.0