mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David LaBarbera <davidlabarb...@localresponse.com>
Subject Re: Question about latent dirichlet allocation
Date Thu, 07 Feb 2013 21:08:30 GMT
What version of mahout are you running and what was the command you used?
In Mahout 0.8, if you typed
mahout lda
you would get 
Try the new Collapsed Variation Bayes LDA, try bin/mahout cvb or bin/mahout cvb0_local

The exception you have looks like a classpath error

David

On Feb 7, 2013, at 4:01 PM, yunming zhang <zhangyunming1990@gmail.com> wrote:

> Hi Yutaka,
> 
> Thanks for the reply!
> 
> So I think lda ($MAHOUT lea) is still in the mahout package? Are you saying
> 1)  it is not doing latent dirichlet allocation now and is replaced by
> CVB,so I can only use cvb?
> or
> 2) it is based on CVB, I could use cvb or lea?
> 
> Thanks
> 
> Yunming
> 
> 
> On Thu, Feb 7, 2013 at 3:41 AM, Yutaka Mandai <20525entradero@gmail.com>wrote:
> 
>> Yunming
>> Hi there!
>> I believe Dirichlet clustering and LDA is a different thing.
>> Dirichlet clustering is a model based clustering algorithm where data is
>> assumed to fit into Gaussian distribution , while LDA is essentially based
>> on CVB (Collapsed variational  Bayes)
>> 
>> LDA used to be LDA in Mahout until 0.5 but beyond this, it's now
>> implemented as CVB.
>> 
>> You might have to do rowed first from your original term vector and then
>> feed that generated matrix to CVB.
>> 
>> Hope this will give you some idea.
>> Regards,,,
>> Y.Mandai
>> 
>> iPhoneから送信
>> 
>> On 2013/02/07, at 13:49, yunming zhang <yz17@rice.edu> wrote:
>> 
>>> Hi,
>>> 
>>> I am trying to get Latent Dirichlet Allocation to work,
>>> 
>>> I was following the instructions on this page
>>> https://cwiki.apache.org/MAHOUT/latent-dirichlet-allocation.html
>>> 
>>> I have two questions
>>> 1) I want to make sure LDA is a different algorithm than the dirichlet
>>> clustering algorithm? I am only asking because when I tried to run
>>> cluster-reuters.sh  under Mahout/trunk/examples/bin, it gave me choices
>>> between kmeans, fuzzy kmeans, meanshift and dirichlet, so it seems as if
>>> LDA and Dirichlet clustering algorithm is the same thing?
>>> 
>>> 2) I tried to run the script cluster-reuters.sh and I got the error
>>> 
>>> SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an
>>> explanation.
>>> SLF4J: slf4j-api 1.6.x (or later) is incompatible with this binding.
>>> SLF4J: Your binding is version 1.5.5 or earlier.
>>> SLF4J: Upgrade your binding to version 1.6.x.
>>> Exception in thread "main" java.lang.NoSuchMethodError:
>>> 
>> org.slf4j.impl.StaticLoggerBinder.getSingleton()Lorg/slf4j/impl/StaticLoggerBinder;
>>> 
>>> 
>>> Not sure what is going on there?
>>> 
>>> Thanks
>>> 
>>> Yunming
>> 


Mime
View raw message