hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Edward J. Yoon" <edwardy...@apache.org>
Subject Re: using hama in gsoc
Date Wed, 23 Jul 2008 14:26:25 GMT
> Actually, in my work, I'll use Newton's method which involves many
> matrix multiplication and subtraction. It seems that you have finished
> this part.

Yes, It done.

> BTW, although I plan to use Quasi-Newton method to avoid calculate the
> inverse of Hessian directly, I'm wonder do you plan to implement the
> inverse operation?

Of course, We also have a lot of needs and issues regarding inverse operation.

Regards, Edward J. Yoon

On Wed, Jul 23, 2008 at 11:02 PM, Yun Jiang <helenjyun@gmail.com> wrote:
> Hi,
> That's great,  since this can reduce much of my work:)
> Actually, in my work, I'll use Newton's method which involves many
> matrix multiplication and subtraction. It seems that you have finished
> this part.
>
> BTW, although I plan to use Quasi-Newton method to avoid calculate the
> inverse of Hessian directly, I'm wonder do you plan to implement the
> inverse operation?
>
> Best,
>
> On Wed, Jul 23, 2008 at 9:28 PM, Chanwit Kaewkasi <chanwit@gmail.com> wrote:
>> Hello Yun,
>>
>> Glad to see another gsoc-er around :-)
>> I'm doing gsoc for another project (Groovy) while helping Edward
>> incubate this one.
>>
>> Cheers,
>>
>> Chanwit
>>
>> 2008/7/23 Yun Jiang <helenjyun@gmail.com>:
>>> Hi, all
>>> I'm implementing Logistic Regression in project of mahout in gsoc.
>>> Is that ok if I use hama in my project?
>>>
>>> Thanks,
>>> Yun
>>>
>>
>



-- 
Best regards, Edward J. Yoon
edwardyoon@apache.org
http://blog.udanax.org

Mime
View raw message