Sparksql query result is not same as presto on same sql

i use carbondata version is 1.2.0 and spark version is 1.6.0.
in my test case
1.Creating a Table
cc.sql("create table IF NOT EXISTS test.table5(id string,name
String,city String,age int) stored by 'carbondata' *
tblproperties('DICTIONARY_INCLUDE' = 'age')* ")

2.load csv data into table,data like this:
id,name,city,age
1,david,shenzhen,31
88,eason,shenzhen,27
3,jarry,wuhan,35

3.select from sparksql,result is :
------------------------

------------------------

------------------------
this result is correct

4.select from presto,result is:
id | name | city | age
----------------+----
1 | david | shenzhen | 3
3 | jarry | wuhan | 4
88 | eason | shenzhen | 2
(3 rows)
look at the age filed,is wrong

I know why this happens because I used dictionary encoding in the age field。