| clear query|facets|time |
Search criteria: .
Results from 21 to 30 from
1050 (0.436s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: machine learning algorithm giving wrong results - Mahout - [mail # user]
|
|
...To unsubscribe from this list, send an email to [EMAIL PROTECTED] On 1/10/13 11:14 AM, Walshe, Maurice (RBI-UK) wrote: ...
|
|
|
Author: Jeff Eastman,
2013-01-10, 17:04
|
|
|
Re: Seeding k-means with canopy clustering / Filter canopies - Mahout - [mail # user]
|
|
...Depending upon your data, 0.7 Canopy can be extremely sensitive to the value you specify for T2. Somewhere between the larger T2 value that yields 1 canopy and the smaller T2 val...
|
|
|
Author: Jeff Eastman,
2013-01-05, 19:27
|
|
|
Re: About Dirichlet clustering's threshold - Mahout - [mail # user]
|
|
...It could be a contradiction indeed. I wonder if you can help us to characterize it further, perhaps by reading the code or by running your data in sequential debug mode? Without ...
|
|
|
Author: Jeff Eastman,
2012-12-26, 15:33
|
|
|
Re: About Dirichlet clustering's threshold - Mahout - [mail # user]
|
|
...Here's a response to a similar question from a couple of months ago: The classification phase of Dirichlet uses a most-likely assignment of points to clusters by default. This me...
|
|
|
Author: Jeff Eastman,
2012-12-25, 20:44
|
|
|
Re: Is the implementation of CIMapper thread safe ? - Mahout - [mail # user]
|
|
...Ah, ok, that's interesting then. The train() method just ends up calling observe() on one or more of the cluster models and this just adds values to their counter vectors (usuall...
|
|
|
Author: Jeff Eastman,
2012-12-22, 20:01
|
|
|
Re: Is the implementation of CIMapper thread safe ? - Mahout - [mail # user]
|
|
...Hi Yunming, The problem I see with what you are proposing is that Hadoop only gives you a single input vector per call of CIMapper.map(). Using multiple threads to perform ...
|
|
|
Author: Jeff Eastman,
2012-12-21, 16:48
|
|
|
Re: Empty clusteredPoints after Dirichlet clustering - Mahout - [mail # user]
|
|
...The classification phase of Dirichlet uses a most-likely assignment of points to clusters by default. This means that, unlike the training phase where points are assigned statist...
|
|
|
Author: Jeff Eastman,
2012-11-28, 14:56
|
|
|
Re: Issue: Canopy is processing extremly slow, what goes wrong? - Mahout - [mail # user]
|
|
...Keep trying larger values until you get a tractable number of canopies, then run cluster dumper to see what they look like. You may also need to increase the heap memory availabl...
|
|
|
Author: Jeff Eastman,
2012-11-14, 13:45
|
|
|
Re: Issue: Canopy is processing extremly slow, what goes wrong? - Mahout - [mail # user]
|
|
...Canopy is very sensitive to the value of T2: Too small a value will cause the creation of very many canopies in each mapper and these will swamp the reducer. I suggest you ...
|
|
|
Author: Jeff Eastman,
2012-11-13, 14:01
|
|
|
Re: Introduction to Apache Mahout K-means clustering - Mahout - [mail # user]
|
|
...See the response "Re: Clustering without hadoop" by Johannes Schulte two postings earlier than yours on user@m.a.o. The driver functions can also be run in sequential mode from a...
|
|
|
Author: Jeff Eastman,
2012-11-13, 13:56
|
|
|
|
