change train.py

eda/esmm/Model_pipline/train.py

@@ -449,14 +449,14 @@ if __name__ == "__main__":
     path = "hdfs://172.16.32.4:8020/strategy/esmm/"
     tf.logging.set_verbosity(tf.logging.INFO)
 
-    host='172.16.40.158'
-    port=4000
-    user='root'
-    password='3SYz54LS9#^9sBvC'
-    db='jerry_test'
-    engine = create_engine(str(r"mysql+pymysql://%s:" + '%s' + "@%s:%s/%s") % (user, password, host, port, db))
-    con = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test', charset = 'utf8')
-    cur = con.cursor()
+    # host='172.16.40.158'
+    # port=4000
+    # user='root'
+    # password='3SYz54LS9#^9sBvC'
+    # db='jerry_test'
+    # engine = create_engine(str(r"mysql+pymysql://%s:" + '%s' + "@%s:%s/%s") % (user, password, host, port, db))
+    # con = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test', charset = 'utf8')
+    # cur = con.cursor()
 
     if FLAGS.task_type == 'train':
         print("train task")
@@ -470,14 +470,16 @@ if __name__ == "__main__":
             result = main(te_files)
             df = df_sort(result,queue_name)
             # df2 = df.drop("time",1)
-            df.to_sql('esmm_device_diary_queue_nearby', con=engine, if_exists='replace', index=False, chunksize=8000)
+            df.to_csv(FLAGS.local_dir + "nearby_queue.csv", header=True, index=False)
+            # df.to_sql('esmm_device_diary_queue_nearby', con=engine, if_exists='replace', index=False, chunksize=8000)
         if queue_name == "native_queue":
             print(queue_name + " task")
             result = main(te_files)
             df_native = df_sort(result,queue_name)
-            nearby_sql = "select device_id,city_id,nearby_queue from esmm_device_diary_queue_nearby"
-            cur.execute(nearby_sql)
-            df_nearby = cur.fetchall()
+            # nearby_sql = "select device_id,city_id,nearby_queue from esmm_device_diary_queue_nearby"
+            # cur.execute(nearby_sql)
+            # df_nearby = cur.fetchall()
+            df_nearby = pd.read_csv(FLAGS.local_dir + "nearby_queue.csv")
             #merge
             df_merge = pd.merge(df_native,df_nearby, on=['device_id', 'city_id'], how='outer').fillna("")
             df_merge['device_id'] = df_merge['device_id'].astype(str)