瀏覽代碼

generate label for mysql

罗俊辉 1 年之前
父節點
當前提交
895bcd22ed
共有 1 個文件被更改,包括 4 次插入3 次删除
  1. 4 3
      process_data.py

+ 4 - 3
process_data.py

@@ -210,9 +210,9 @@ class UserProcess(object):
         user_path = 'data/train_data/daily-user-info-{}-{}.json'.format(start_date, end_date)
         with open(user_path) as f:
             data = json.loads(f.read())
-        sql = "select video_id, hour_dt_str from lighgbm_data where type = 'userupload' and address is NULL;"
+        sql = "select video_id, hour_dt_str from lightgbm_data where type = 'userupload' and address is NULL;"
         dt_list = self.client_spider.select(sql)
-        for item in dt_list:
+        for item in tqdm(dt_list):
             video_id, dt = item
             user_info_obj = data.get(dt, {}).get(video_id)
             if user_info_obj:
@@ -226,7 +226,8 @@ class UserProcess(object):
                     return_videos_30 = user_info_obj['30day_return_2000_videos']
                     update_sql = f"""UPDATE lighgbm_data set address='{address}', user_return_3={return_3}, user_view_3={view_3}, user_share_3={share_3}, user_return_videos_3={return_videos_3}, user_return_videos_30={return_videos_30} where video_id = '{video_id}';"""
                     self.client_spider.update(update_sql)
-                except:
+                except Exception as e:
+                    print(e)
                     pass
 
     def generate_user_data(self, flag, dt_time=None):