Re: [SOLVED]Re: best way to add custom job to mahout?

2012-02-24 Thread Ioan Eugen Stan
Pe 24.02.2012 05:52, Lance Norskog a scris: bin/mahout can take a classname instead of a name from src/conf/driver.classes.props. So, you don't have to package up your job as a mahout job. Thank you Lance, I found it's more easy to just send everything to the hadoop script. I was going to

Re: Naive-Bayes work flow

2012-02-24 Thread Naveenchandra
The python code which used is : import random f = open(/home/hadoop/yellow.tsv, w) for i in range(0,1000): print f, yellow\t,random.randint(0,5),random.randint(0,5), random.randint(0,5),random.randint(0,5) same for purple.tsv also, i copied 1st 100 records from tsv files to use as test

recommender data live refresh

2012-02-24 Thread Aleksei Udatšnõi
Hi, I am using an item-based recommender inside a servlet and would like to reload its data periodically (from db). Each reload takes up to 30 sec. Recommender object is Refreshable and thus can do periodic reloads by using refresh() function. My only concern is the impact of reload on the live

Re: Goals for Mahout 0.7

2012-02-24 Thread Grant Ingersoll
One of our top goals, in my mind, has to be speeding up our tests! I only wish I knew how given basic attempts at parallelism and Maven have failed miserably. On Feb 14, 2012, at 3:29 PM, Jeff Eastman wrote: +users@ Just to be clear, I'm not advocating replacing the JIRA process with a

Re: Naive-Bayes work flow

2012-02-24 Thread Ted Dunning
If your synthetic data comes from the se distribution for yellow and purple then clearly no classifier will help. Also naive bayes wants words not numbers. Sent from my iPhone On Feb 24, 2012, at 5:08 AM, Naveenchandra naveenchandr...@gmail.com wrote: The python code which used is :

Lucene Revolution in Boston in May (with a side of Mahout)

2012-02-24 Thread Grant Ingersoll
Hi Mahout's, Thought some here might be interested as search and machine learning often go together. -- Lucene Revolution will be here May 9-10 in Boston. Reserve your spot today with Early Bird pricing of $575. Committers and accepted speakers are entitled to free admission. Our CFP is

Re: seqdirectory doesn't seem to be generating seqfiles...?

2012-02-24 Thread Temese Szalai
Hi Lance - Thanks for the follow up. I appreciate it. No, unfortunately, I don't think that is what I want. I am really just trying to get the basics down and one of those basics includes generating vectors from text files, which starts with generating sequence files from a directory of text

Re: seqdirectory doesn't seem to be generating seqfiles...?

2012-02-24 Thread Lance Norskog
Look at examples/bin/classify-reuters.sh. It runs seq2sparse on the reuters corpus. I'm an incrementalist- I find an example that works and then change things one at a time. There is so much behind the scenes in the text jobs. On Fri, Feb 24, 2012 at 4:01 PM, Temese Szalai temesesza...@gmail.com

Re: recommender data live refresh

2012-02-24 Thread Lance Norskog
You have to use two recommender objects if you want to serve recommendations while refreshing. I don't think any of the recommenders have that as part of the api contract. On Fri, Feb 24, 2012 at 8:31 AM, Aleksei Udatšnõi a.udac...@gmail.com wrote: Hi, I am using an item-based recommender

Re: problem:when no recommendation ?

2012-02-24 Thread Sebastian Schelter
A very simple solution is to present a list of very popular/recent items. --sebastian 2012/2/24 manish dunani manishd...@gmail.com: hi, what to do when no recommendation comes ? If i wanted  to give recommendation instead of  no recommendation,then is it possible to use content based

Re: recommender data live refresh

2012-02-24 Thread Sebastian Schelter
This is not true. If you are in a servlet environment, the servlet container will use the recommender in multiple threads which means the app does not block while the recommender is refreshed. During the refresh process a new DataModel is built and the reference to that is simply switched once it