AdaBoost

[Wojciech Indyk <wo...@gmail.com>]

Hi!
I want implement AdaBoost in Mahout. Could it be useful in Mahout? I
think so, because it's strong algorithm and very powerful, but Mahout
is specific, so who knows :)
I thought about training data and I know, that I must parallelize by
data, rather than by algorithms, so it will be not so easy - I must
run all mapers of chosed algorithms in my training maper, but I have
no idea how could i choose algorithms to adaboost (in architecture
way) like a parameter.

Regards

Re: AdaBoost

[Hector Yee <he...@gmail.com>]

It is if you use the grad boost variant. I'll work on it next week on vacation...

Sent from my iPad

On May 24, 2011, at 4:48 PM, Ted Dunning <te...@gmail.com> wrote:

> Is AdaBoost a scalable algorithm?
> 
> It seems to me that it is inherently very sequential.
> 
> On Tue, May 24, 2011 at 1:57 PM, Wojciech Indyk <wo...@gmail.com>wrote:
> 
>> Hi!
>> I want implement AdaBoost in Mahout. Could it be useful in Mahout? I
>> think so, because it's strong algorithm and very powerful, but Mahout
>> is specific, so who knows :)
>> I thought about training data and I know, that I must parallelize by
>> data, rather than by algorithms, so it will be not so easy - I must
>> run all mapers of chosed algorithms in my training maper, but I have
>> no idea how could i choose algorithms to adaboost (in architecture
>> way) like a parameter.
>> 
>> Regards
>> 

Re: AdaBoost

[Ted Dunning <te...@gmail.com>]

Is AdaBoost a scalable algorithm?

It seems to me that it is inherently very sequential.

On Tue, May 24, 2011 at 1:57 PM, Wojciech Indyk <wo...@gmail.com>wrote:

> Hi!
> I want implement AdaBoost in Mahout. Could it be useful in Mahout? I
> think so, because it's strong algorithm and very powerful, but Mahout
> is specific, so who knows :)
> I thought about training data and I know, that I must parallelize by
> data, rather than by algorithms, so it will be not so easy - I must
> run all mapers of chosed algorithms in my training maper, but I have
> no idea how could i choose algorithms to adaboost (in architecture
> way) like a parameter.
>
> Regards
>

Re: AdaBoost

[Hector Yee <he...@gmail.com>]

Wojciech, I've opened a ticked you can watch

 https://issues.apache.org/jira/browse/MAHOUT-716

I should have the in core code ready in ~3 days. The gradient portion is
easily parallelizable if you want to implement it as mapreduce.

On Tue, May 24, 2011 at 1:57 PM, Wojciech Indyk <wo...@gmail.com>wrote:

> Hi!
> I want implement AdaBoost in Mahout. Could it be useful in Mahout? I
> think so, because it's strong algorithm and very powerful, but Mahout
> is specific, so who knows :)
> I thought about training data and I know, that I must parallelize by
> data, rather than by algorithms, so it will be not so easy - I must
> run all mapers of chosed algorithms in my training maper, but I have
> no idea how could i choose algorithms to adaboost (in architecture
> way) like a parameter.
>
> Regards
>



-- 
Yee Yang Li Hector
http://hectorgon.blogspot.com/ (tech + travel)
http://hectorgon.com (book reviews)

Re: AdaBoost

[Hector Yee <he...@gmail.com>]

Patch uploaded

https://issues.apache.org/jira/browse/MAHOUT-716

On Tue, May 24, 2011 at 1:57 PM, Wojciech Indyk <wo...@gmail.com>wrote:

> Hi!
> I want implement AdaBoost in Mahout. Could it be useful in Mahout? I
> think so, because it's strong algorithm and very powerful, but Mahout
> is specific, so who knows :)
> I thought about training data and I know, that I must parallelize by
> data, rather than by algorithms, so it will be not so easy - I must
> run all mapers of chosed algorithms in my training maper, but I have
> no idea how could i choose algorithms to adaboost (in architecture
> way) like a parameter.
>
> Regards
>



-- 
Yee Yang Li Hector
http://hectorgon.blogspot.com/ (tech + travel)
http://hectorgon.com (book reviews)