Thank you!

On Mon, Feb 29, 2016 at 4:17 PM, vincent gromakowski <
vincent.gromakow...@gmail.com> wrote:

> if you want to auto load, go in zeppelin UI to settings, edit spark
> interpreter config and ad a line in dependencies artifact...
>
> 2016-02-29 14:16 GMT+01:00 Aleksandr Modestov <aleksandrmodes...@gmail.com
> >:
>
>> Thank you!
>> It does work!
>> "%dep
>> z.load("ai.h2o:sparkling-water-core_2.10:1.3.7")"
>>
>> On Mon, Feb 29, 2016 at 3:43 PM, vincent gromakowski <
>> vincent.gromakow...@gmail.com> wrote:
>>
>>> Try to use the dependency loader in Spark interpreter configuration
>>> page. I have encountered strange behaviors with spark.jars options...
>>>
>>> 2016-02-29 13:35 GMT+01:00 Aleksandr Modestov <
>>> aleksandrmodes...@gmail.com>:
>>>
>>>> Hello!
>>>> Excuse me, but it doesn't work...
>>>> I open an interpreter window and create several additional lines.
>>>> spark.jars .../sparkling-water-assembly-1.5.10-all.jar
>>>> spark.jars.packages ai.h2o:sparkling-water-core_2.10
>>>> Inside the notebook I try to add a h2o-lib: import
>>>> org.apache.spark.h2o._
>>>> But I have a problem: <console>:39: error: object h2o is not a member
>>>> of package org.apache.spark
>>>> import org.apache.spark.h2o._
>>>> Thank you
>>>>
>>>>
>>>>
>>>> On Sun, Feb 21, 2016 at 9:21 AM, moon soo Lee <m...@apache.org> wrote:
>>>>
>>>>> As Felix mentioned,
>>>>>
>>>>> Loading ai.h2o:sparkling-water-core_2.10 package [1] in
>>>>> SparkInterpreter [2] would let H2O work in Zeppelin.
>>>>>
>>>>> Let me know if it does not work for you.
>>>>>
>>>>> Thanks,
>>>>> moon
>>>>>
>>>>> [1]
>>>>> https://github.com/h2oai/sparkling-water#sparkling-water-as-spark-package
>>>>> [2]
>>>>> http://zeppelin.incubator.apache.org/docs/latest/interpreter/spark.html 
>>>>> Dependency
>>>>> Management section
>>>>>
>>>>> On Sat, Feb 20, 2016 at 9:01 AM Felix Cheung <
>>>>> felixcheun...@hotmail.com> wrote:
>>>>>
>>>>>> According to this
>>>>>>
>>>>>> https://github.com/h2oai/sparkling-water
>>>>>>
>>>>>> It can be loaded as a spark package into a spark shell - the same way
>>>>>> should work with Zeppelin Spark interpreter (which is running the spark
>>>>>> shell).
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Sat, Feb 20, 2016 at 12:58 AM -0800, "Aleksandr Modestov" <
>>>>>> aleksandrmodes...@gmail.com> wrote:
>>>>>>
>>>>>> "H2o works in Python, Java, Scala or with Spark (Sparkling Water) as
>>>>>> well."
>>>>>> Thank you:)
>>>>>> I know I work with H2O from Jupyter or from shells...
>>>>>> But I hope that I can use Scala (for instance) from zeppelin notebook
>>>>>> it's better that use shell...
>>>>>> I can not find where I can point out zeppelin how to work with h2o
>>>>>> algoriths.
>>>>>> It sounds very good that I can work from Zeppelin notebook with Spark
>>>>>> and H2O algorithms inside one workplace.
>>>>>>
>>>>>> On Sat, Feb 20, 2016 at 8:44 AM, Felix Cheung <
>>>>>> felixcheun...@hotmail.com> wrote:
>>>>>>
>>>>>> H2o works in Python, Java, Scala or with Spark (Sparkling Water) as
>>>>>> well.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Fri, Feb 19, 2016 at 10:11 AM -0800, "Girish Reddy" <
>>>>>> gir...@springml.com> wrote:
>>>>>>
>>>>>> You'll need an R interpreter -
>>>>>> https://github.com/elbamos/Zeppelin-With-R
>>>>>>
>>>>>> You can then load the H2O libraries just as you would from RStudio.
>>>>>>
>>>>>>
>>>>>> On Fri, Feb 19, 2016 at 8:41 AM, Aleksandr Modestov <
>>>>>> aleksandrmodes...@gmail.com> wrote:
>>>>>>
>>>>>> If I want to use h2o libraries from noteboke what shoul I do?
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>
>>>
>>
>

Reply via email to