修改测试文件

tensnsorflow/multi.py

@@ -38,7 +38,7 @@ def feature_engineer():
     validate_date = con_sql(db, sql)[0].values.tolist()[0]
     print("validate_date:" + validate_date)
     temp = datetime.datetime.strptime(validate_date, "%Y-%m-%d")
-    start = (temp - datetime.timedelta(days=3)).strftime("%Y-%m-%d")
+    start = (temp - datetime.timedelta(days=100)).strftime("%Y-%m-%d")
     print(start)
 
     sql = "select e.y,e.z,e.stat_date,e.ucity_id,feat.level2_ids,e.ccity_name,u.device_type,u.manufacturer," \
@@ -160,7 +160,7 @@ def get_predict(date,value_map,app_list_map,level2_map,level3_map):
     print("native")
     print(native_pre.count())
 
-    native_pre.write.format("avro").save(path="/recommend", mode="overwrite")
+    native_pre.write.format("avro").save(path="/recommend/pre_native", mode="overwrite")
 
     spark.createDataFrame(rdd.filter(lambda x: x[6] == 0)
                                    .map(lambda x: (x[0], x[1], x[2],x[9],x[10],x[11],x[12],x[13],x[14],x[15],
@@ -174,7 +174,7 @@ def get_predict(date,value_map,app_list_map,level2_map,level3_map):
         .toDF("city", "uid", "cid_id")
     print("nearby")
     print(nearby_pre.count())
-    nearby_pre.write.format("avro").save(path="/recommend", mode="overwrite")
+    nearby_pre.write.format("avro").save(path="/recommend/pre_nearby", mode="overwrite")
 
     spark.createDataFrame(rdd.filter(lambda x: x[6] == 1)
                           .map(lambda x: (x[0], x[1], x[2], x[9], x[10], x[11], x[12], x[13], x[14], x[15],