| clear query|facets|time |
Search criteria: position.
Results from 1 to 10 from
68 (0.633s).
|
|
|
Loading phrases to help you
refine your search...
|
|
Re: Logistic Regression: number of positives and negatives - Mahout - [mail # user]
|
|
...Downsampling negatives should make little difference to accuracy. It can substantially affect training time however. Sent from my iPhone On Jul 11, 2011, at 6:56, Svetlomir Kasa...
|
|
|
Author: Ted Dunning,
2011-07-11, 14:29
|
|
|
Re: Ratio between positive and negative data in a classification model - Mahout - [mail # user]
|
|
...Having lots of negative samples won't improve performance that much (shouldn't hurt much either). The negative examples that you really want are the ones that are close to your positive...
|
|
|
Author: Ted Dunning,
2012-10-02, 11:09
|
|
|
Re: Call to action – Mahout needs your help - Mahout - [mail # dev]
|
|
... contributors positively b) won't break things too often (breaking the build sometimes is probably a good sign) c) will make significant contributions without leading the project into a situation...
|
|
|
Author: Ted Dunning,
2013-03-26, 08:29
|
|
|
Re: Cross recommendation - Mahout - [mail # user]
|
|
... dithering set up) it will quickly tune away the false-positive mistakes it is making by showing the events in question and seeing that they don't lead to success. This closed loop nature makes...
|
|
|
Author: Ted Dunning,
2013-02-22, 05:41
|
|
|
Re: Decision Forest - Partial implementation - Mahout - [mail # user]
|
|
... be position independent. Not quite true. See http://docs.oracle.com/javase/1.4.2/docs/api/java/nio/channels/FileChannel.html#map(java.nio.channels.FileChannel.MapMode,%20long,%20long) You...
|
|
|
Author: Ted Dunning,
2012-12-09, 09:29
|
|
|
Re: Should I be using OnlineLogisticRegression? - Mahout - [mail # user]
|
|
...Try transforming them as well, likely with a log if they are positive and have heavily skewed values. Can you suck the data into R and paste in the results of summary(x)? (assuming you put...
|
|
|
Author: Ted Dunning,
2012-09-07, 00:14
|
|
|
Re: Recommendation scores from LogLikelihood Similarity recommender - Mahout - [mail # user]
|
|
... recommend that you test a boolean approach where in *any* action is considered positive and another where you consider only your positive actions and ignore your negative actions. If necessary...
|
|
|
Author: Ted Dunning,
2012-05-06, 19:53
|
|
|
Re: Mixing simiarity measures - Mahout - [mail # user]
|
|
... for a lot, but just writing it for the sake of writing it isn't as likely to find a positive reception. On Wed, May 23, 2012 at 12:11 PM, Ted Dunning wrote: ...
|
|
|
Author: Ted Dunning,
2012-05-29, 22:24
|
|
|
Re: AUC=1.00 when confusion matrix shows prediction not perfect - Mahout - [mail # user]
|
|
...AUC is independent of threshold. The confusion matrix is not. If all scores for the positive class are greater than all scores for the negative class you will have AUC = 1...
|
|
....00. On the other hand, that doesn't say that all of the positive scores are > 0.5 and all the negative ones wrote: ...
|
|
|
Author: Ted Dunning,
2011-05-23, 00:14
|
|
|
Re: Classification beginner questions - Mahout - [mail # user]
|
|
...Joscha, There is no implicit training. you need to give negative examples as well as positive. On Sat, Jun 11, 2011 at 9:08 AM, Joscha Feth wrote: ...
|
|
|
Author: Ted Dunning,
2011-06-12, 08:07
|
|
|
|
