service model 优化

spark/featureEng2.py

@@ -83,7 +83,7 @@ def parseTags(tags,i):
 def numberToBucket(num):
     res = 0
     if not num:
-        return res
+        return str(res)
     if num >= 1000:
         res = 1000//10
     else:
@@ -93,7 +93,7 @@ def numberToBucket(num):
 def priceToBucket(num):
     res = 0
     if not num:
-        return res
+        return str(res)
     if num >= 100000:
         res = 100000//1000
     else:
@@ -627,7 +627,7 @@ def init_es_query():
         },
         "query": {
             "bool": {
-                "must": [],
+                "must": [{"term": {"is_online": True}}],
                 "must_not": [],
                 "should": []
             }
@@ -790,6 +790,7 @@ if __name__ == '__main__':
     for k, v in dataVocab.items():
         print(k, len(v))
 
+
     itemDF_spark = spark.createDataFrame(itemDF)
     itemDF_spark.printSchema()
     itemDF_spark.show(10, truncate=False)
@@ -797,8 +798,6 @@ if __name__ == '__main__':
     # 统计数据处理
     ratingSamplesWithLabel = addStaticsFeatures(ratingSamplesWithLabel,dataVocab)
 
-
-
     samples = ratingSamplesWithLabel.join(itemDF_spark, on=['item_id'], how='inner')
 
     print("处理user特征...")