| clear query|facets|time |
Search criteria: .
Results from 51 to 60 from
872 (0.103s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: performance study - Mahout - [mail # user]
|
|
...You do not use map-reduce algorithms to do compute jobs faster- they are usually much slower than the same computation in one program. You use map-reduce to do things that are otherwise impo...
|
|
|
Author: Lance Norskog,
2012-07-30, 21:10
|
|
|
Re: ERROR: OutOfMemoryError: Java heap space - Mahout - [mail # user]
|
|
...Increase the memory size or split the file! On Thu, Jul 26, 2012 at 5:37 AM, pricila rr wrote: Lance Norskog [EMAIL PROTECTED]...
|
|
|
Author: Lance Norskog,
2012-07-27, 02:12
|
|
|
Re: EMC Israel Data Science Challenge- classify open source to the project - Mahout - [mail # user]
|
|
...Do you mean CNB instead of plain NB for this task? If it is a short answer, why? On Wed, Jul 25, 2012 at 8:10 PM, Lance Norskog wrote: Lance Norskog [EMAIL PROTECTED...
|
|
|
Author: Lance Norskog,
2012-07-26, 03:25
|
|
|
Re: EMC Israel Data Science Challenge- classify open source to the project - Mahout - [mail # user]
|
|
...Or track down cyberwarfare. C.f. Stuxnet and Flame. http://www.wired.com/threatlevel/2012/06/flame-tied-to-stuxnet/ On Wed, Jul 25, 2012 at 9:29 AM, Robin Anil wrote: &nbs...
|
|
|
Author: Lance Norskog,
2012-07-26, 03:10
|
|
|
Re: .txt to vector - Mahout - [mail # user]
|
|
...It is a jar file, so just java -jar luke.....jar But, there's a problem. Luke releases are keyed to different Lucene releases. You need the right Luke download for your version of Luce...
|
|
|
Author: Lance Norskog,
2012-07-25, 07:57
|
|
|
EMC Israel Data Science Challenge- classify open source to the project - Mahout - [mail # user]
|
|
...http://www.kaggle.com/c/emc-data-science Match source code files to the open source code project Lance Norskog [EMAIL PROTECTED] p.s. or Stuxnet :)...
|
|
|
Author: Lance Norskog,
2012-07-25, 07:55
|
|
|
Re: .txt to vector - Mahout - [mail # user]
|
|
...The Luke program lets you examine a Lucene index. Try that and check for your term vectors. http://code.google.com/p/luke/ It uses Swing, so you need the index on your local PC. ...
|
|
|
Author: Lance Norskog,
2012-07-25, 07:23
|
|
|
Re: .txt to vector - Mahout - [mail # user]
|
|
...You're making progress! Run "bin/mahout lucene.vector" and look at the help message: --maxPercentErrorDocs (-err) maxPercentErrorDocs The max percentage of ...
|
|
|
Author: Lance Norskog,
2012-07-25, 06:59
|
|
|
Re: Visualize clusters - Mahout - [mail # user]
|
|
...Here is the only tool I know: use 'bin/mahout clusterdump' to export clusters with the graphml option. The the 'Giraph' program (available for free somewhere on the internet) can read these ...
|
|
|
Author: Lance Norskog,
2012-07-24, 02:30
|
|
|
Re: .txt to vector - Mahout - [mail # user]
|
|
...You have to add termvectors to the field type you want to use. Then, you have to reindex all of the data. You will now have another file in the index with the suffix .tvf. This has the data ...
|
|
|
Author: Lance Norskog,
2012-07-24, 02:27
|
|
|
|
