Account Options

  1. Sign in
The old Google Groups will be going away soon, but your browser is incompatible with the new version.
Google Groups Home
« Groups Home
MLcomp
Home
+ new post
About this group
Join this group
Is there a way to see fast/memory efficient classification algorithms?
Options
There are currently too many topics in this group that display first. To make this topic appear first, remove this option from another topic.
There was an error processing your request. Please try again.
flag
   3 messages - Collapse all  -  Translate all to Translated (View all originals)
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
 
From:
To:
Cc:
Followup To:
Add Cc | Add Followup-to | Edit Subject
Subject:
Validation:
For verification purposes please type the characters you see in the picture below or the numbers you hear by clicking the accessibility icon. Listen and type the numbers you hear
 
Rob Renaud  
View profile  
 More options Jul 28 2011, 6:16 pm
From: Rob Renaud <rren...@gmail.com>
Date: Thu, 28 Jul 2011 18:16:09 -0400
Local: Thurs, Jul 28 2011 6:16 pm
Subject: Is there a way to see fast/memory efficient classification algorithms?
Print | Individual message | Show original | Report this message | Find messages by this author

I am mostly interested in seeing how different algorithms perform on this 170
MB dataset :

http://mlcomp.org/datasets/872

As you can see, half of the algorithms ran on the dataset have either timed
out or hit the memory limit.  Is it possible to sort classification
algorithms by speed/memory usage, and only run ones that are likely to
succeed?  I suspect anything that uses weka will hit the memory limit for
example.

I have a tiny sample of the same dataset at http://mlcomp.org/datasets/871.


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
Zeno Gantner  
View profile  
 More options Jul 29 2011, 3:17 am
From: Zeno Gantner <zeno.gant...@gmail.com>
Date: Fri, 29 Jul 2011 09:17:20 +0200
Local: Fri, Jul 29 2011 3:17 am
Subject: Re: Is there a way to see fast/memory efficient classification algorithms?
Print | Individual message | Show original | Report this message | Find messages by this author
Hi Rob,

I think your feature request makes sense.

If you are looking for another program that is likely to succeed on
your dataset,
try Vowpal Wabbit:
https://github.com/JohnLangford/vowpal_wabbit/wiki

Best regards,
  Zeno

--
MyMediaLite Recommender System Library: http://ismll.de/mymedialite


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
Percy Liang  
View profile  
 More options Aug 1 2011, 11:10 am
From: Percy Liang <percyli...@gmail.com>
Date: Mon, 1 Aug 2011 11:10:01 -0400
Local: Mon, Aug 1 2011 11:10 am
Subject: Re: Is there a way to see fast/memory efficient classification algorithms?
Print | Individual message | Show original | Report this message | Find messages by this author
Thanks for the feedback - I've added it to our issue tracker.  In the
meantime, this text file (which is periodically updated) has the list
of all the runs in our system:
http://www.mlcomp.org/download/runs.txt
You can use a simple script to see which programs have failed the
least number of times.

 -Percy


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
End of messages
« Back to Discussions « Newer topic     Older topic »

Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
©2013 Google