Purpose of %dep is loading external libraries to Zeppelin and distribute
them to workers.
For example, if your pre-loaded scala script needs to import spark-csv
library, you will need to run dep interpreter first as below prior to
running other scala, pyspark code:
%dep
z.load("com.databricks.spark-csv_2.10:1.3.0")

For your information, %dep interpreter load libraries only to %spark,
%pyspark but not to %sql. If you have *table* made with spark-csv library,

%spark
z.show(sqlc.sql("select * from table")
will work,

%sql
select * from table
won't work. I am currently working on this bug so patch will be made soon.

On Wed, Jan 20, 2016 at 6:17 PM, MichaelYoung <452611...@qq.com> wrote:

> Thanks for the tip, but my analytic queries work fine after my Scala
> script is pre-loaded. In my notebook, I just type in "%sql
> SQL-QUERY-STATEMENT", and it works fine. What exactly is %dep used for ? I
> never used this interpreter.
>
>
> ------------------ 原始邮件 ------------------
> *发件人:* "bear gan";<bear...@hotmail.com>;
> *发送时间:* 2016年1月21日(星期四) 上午10:14
> *收件人:* "Mina Lee"<mina...@apache.org>; "
> users@zeppelin.incubator.apache.org"<users@zeppelin.incubator.apache.org>;
>
> *主题:* 答复: error: object jblas is not a member of package org
>
> Thanks so much! It works fine as I expected.
>
>
>
>  Note that%dep interpreter should be used before %spark, %pyspark, %sql.
>
>
>
>
>
>
> *发件人: *bear gan <bear...@hotmail.com>
> *发送时间: *2016年1月20日 17:16
> *收件人: *Mina Lee <mina...@apache.org>; users@zeppelin.incubator.apache.org
> *主题: *re: error: object jblas is not a member of package org
>
>
>
> Run
>
> //////////////////////////////////////////////////////////////////
>
> %dep
>
> z.load("org.scalanlp:jblas:1.2.1")
>
> //////////////////////////////////////////////////////////////////
>
> Out:
>
> dep interpreter not found
>
>
>
>
>
>
>
>
>
> Hi,
>
>
>
> have you tried to load jblas library using %dep interpreter?
>
>
>
> For example, you can try to run below:
>
> ```
>
> %dep
>
> z.load("org.scalanlp:jblas:1.2.1")
>
> ```
>
>
>
> And then run
>
> `import org.jblas.DoubleMatrix` in next paragraph.
>
>
>
> If you want to know another way to load external library, please check
> "Dependency Management" section in
> https://zeppelin.incubator.apache.org/docs/0.6.0-incubating-SNAPSHOT/interpreter/spark.html
>
>
>
>
> Hope this helps.
>
>
>
> On Tue, Jan 19, 2016 at 11:46 PM, bear gan <bear...@hotmail.com> wrote:
>
>
>
> How to solve this problem?
>
>
> ////////////////////////////////////////////////////////////////////////////////
>
> import org.jblas.DoubleMatrix
>
> <console>:21: error: object jblas is not a member of package org
>
>        import org.jblas.DoubleMatrix
>
>
> ////////////////////////////////////////////////////////////////////////////////
>
>
>
>
>
>
>
>
>
>
>
>
>

Reply via email to