Oops, I'll only be able to commit the code updates in the morning. JIRAs have been created.
On Wed, Apr 13, 2011 at 10:17 PM, Jason Baldridge <[email protected]>wrote: > Sorry, I got into a very busy spell there and didn't even read the rest of > the thread because it was on my stack... > > I'll add the JIRAs and commit the code updates. > > Jason > > > On Tue, Apr 12, 2011 at 12:15 PM, Jörn Kottmann <[email protected]>wrote: > >> Jason, >> >> can you give me a short update about these changes, >> should we go ahead with the release, or is this something you >> really need in it? >> >> If not, I would build RC 6 and do the vote. >> >> Thanks, >> Jörn >> >> >> On 4/11/11 2:25 PM, Jörn Kottmann wrote: >> >>> On 4/11/11 2:11 PM, Jason Baldridge wrote: >>> >>>> As it turns out, I found some issues with the way perceptron output was >>>> normalized. It was sort of a strange way to handle negative numbers that >>>> didn't really work, so I changed it to exponentiation and then >>>> normalization. >>>> >>>> >>> Can you please open a jira for this issue, and maybe give us a reference >>> to the code? >>> >>> Also, the training accuracies reported during perceptron training were >>>> much higher than final training accuracy, which turned out to be an >>>> artifact >>>> of the way training examples were ordered. I changed this so that after >>>> each >>>> iteration, the training accuracy is scored without changing the parameters. >>>> This gives a coherent value reported on every iteration, and it also allows >>>> early stopping by checking whether the same accuracy has been obtained for >>>> some number of times (e.g. 4) in a row. (This could also be done by >>>> checking >>>> that parameter values haven't changed, which would be better, but which I'd >>>> only want to do after refactoring.) >>>> >>>> Please also make a jira for this one. >>> >>> I'm going to test the changes on a bunch of datasets this evening. If >>>> anyone else is using the perceptrons much, it would be good if they could >>>> do >>>> a before and after comparison. >>>> >>> >>> We only use the perceptron for the POSTagger currently, we can re-run the >>> accuracy we get on >>> some training/test sets. >>> >>> Jörn >>> >> >> > > > -- > Jason Baldridge > Assistant Professor, Department of Linguistics > The University of Texas at Austin > http://www.jasonbaldridge.com > http://twitter.com/jasonbaldridge > > > -- Jason Baldridge Assistant Professor, Department of Linguistics The University of Texas at Austin http://www.jasonbaldridge.com http://twitter.com/jasonbaldridge
