初始化

applications/functions.py

@@ -1,7 +1,8 @@
 """
 @author: luojunhui
 """
-import asyncio
+import jieba.analyse
+from pandas as pd
 
 from .model_init import models
 
@@ -10,10 +11,35 @@ class ParamProcess(object):
     """
     处理 params， 继承 models
     """
+
     def __init__(self):
         self.model_v1 = models.model_v1
         self.model_v2 = models.model_v2
-        self.layer_encoder = models.label_encoder
+        self.label_encoder = models.label_encoder
+
+    async def title_to_tags(self, features):
+        """
+        process video title to tags and transform features_json_to_dataFrame
+        :param features:
+        :return:
+        """
+        title = features['title']
+        if title:
+            title = title.strip()
+            title_tags = list(jieba.analyse.textrank(title, topK=3))
+            if title_tags:
+                for i in range(3):
+                    try:
+                        features['tag_{}'.format(i + 1)] = title_tags[i]
+                    except:
+                        features['tag_{}'.format(i + 1)] = None
+            else:
+                features['tag_1'] = None
+                features['tag_2'] = None
+                features['tag_3'] = None
+        df = pd.DataFrame(features)
+        df.drop('title', axis=1)
+        return df
 
     async def predict_score(self, version, features):
         """
@@ -24,9 +50,12 @@ class ParamProcess(object):
         """
         match version:
             case "v1":
-                return await self.model_v1.predict(features)
+                result = await self.model_v1(features)
+                print(result)
+                return result
             case "v2":
-                return await self.model_v2.predict(features)
+                result = await self.model_v2.predict(features)
+                return result
 
     async def process_label(self, params):
         """
@@ -36,15 +65,27 @@ class ParamProcess(object):
         """
         version = params['version']
         features = params['features']
+        features = await self.title_to_tags(features)
         match version:
             case "v1":
                 # 全部转化为类别
-                print("all to string cate")
-                # features = []
+                str_column = [
+                    "channel",
+                    "type",
+                    "tag1",
+                    "tag2",
+                    "tag3"
+                ]
+                for key in str_column:
+                    features[key] = self.label_encoder.fit_transform(features[key])
                 return version, features
             case "v2":
-                print("all to float cate")
-                # features = []
+                float_column = ["out_play_cnt", "out_like_cnt", "out_share_cnt", "lop", "duration"]
+                str_column = ["channel", "mode", "out_user_id", "tag1", "tag2", "tag3"]
+                for key in float_column:
+                    features[key] = pd.to_numeric(features[key], errors="coerce")
+                for key in str_column:
+                    features[key] = self.label_encoder.fit_transform(features[key])
                 return version, features
 
     async def process(self, params):
@@ -55,6 +96,4 @@ class ParamProcess(object):
         """
         version, features = await self.process_label(params)
         print(version, features)
-        # return await self.predict_score(version, features)
-
-
+        return await self.predict_score(version, features)

applications/routes.py

@@ -26,4 +26,7 @@ async def post_data():
     p = ParamProcess()
     data = await request.get_json()
     processed_data = await p.process(data)
-    return jsonify(processed_data)
+    r = {
+        'data': processed_data
+    }
+    return jsonify(r)

test.py

@@ -0,0 +1,3 @@
+"""
+@author: luojunhui
+"""