This shows you the differences between two versions of the page.

Link to this comparison view

nlp-private:josh [2015/04/23 13:40] (current)
ryancha created
Line 1: Line 1:
 +Projects and other links:
 +*[[Spoken Language ID]]
 +*[[Feature Engineering Console]]
 +*[[Target Platforms]]
 +*[[NLPfest 2010 Notes]]
 +==Completed Tasks==
 +** <​s>​Get cmake to take care of compiling Language-ID?</​s>​ No, cmake'​s support for java building is not developed enough. We'll stick with [[ant]] for now.
 +** <​s>​Wait for stat-nlp and sphinx to be depended on as jars so we don't have to build those sources?</​s>​ We're already using the statnlp jar. Don't wait for the sphinx one, just let that be taken care of in HEAD.
 +** <​s>​Merge CMake into head</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​292/​ r292]
 +*** <​s>​Use -norm 1 in RBLDROPTS (verify that there aren't any other options -- check regression test)</​s> ​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​277/​ r277]. The correct parameter is --norm=1
 +**** <​s>​Double check change of '​norm'​ to integer in [[resultbuilder.pl]]</​s>​ - DONE, it seems okay
 +*** <​s>​Remove built-in ant</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​280 r280]
 +*** <​s>​Fix regression (norm?​)</​s>​ Same issue as "Use -norm 1 in RBLDROPTS"​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​277/​ r277]
 +*** <​s>​Verify that Makefile-original will work as expected - make detcurve, make metric, regressiontest.sh,​ etc.</​s>​ - DONE. I talked with Dr. Ringger and we decided that this is not a critical issue. The previous makefile will no longer work once cmake is checked in to head.
 +*** <​s>​Have CMake keep experiments directory where it was previously. This should be implemented when we finally check in.</​s>​ - DONE, with merge to head in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​292/​ r292]
 +*** <​s>​CODE REVIEW! w/ Dr. Ringger</​s>​ - DONE, with merge to head in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​292/​ r292]. Robbie actually did the review.
 +* [[Feature Engineering Console]]
 +** <​s>​Review ruby code completely</​s>​ - OBSELETED by port to Java
 +** <​s>​plot (detware) : ntrue = 15, nfalse = 99, npts = 113</​s>​ - DONE - rather than gathering these variables from console output, we in essence store the same information in the [http://​nlp.cs.byu.edu/​trac/​NIST/​browser/​branches/​feature-eng-console/​Language-ID/​src/​edu/​byu/​langid/​app/​datastructures/​results/​Result.java Result class]
 +** <​s>​tab for each experiment result</​s>​
 +** <​s>​Cost matrix: see board</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​324 r324]
 +** <​s>​GUI mockup - full-ish functionality showing the flow of information,​ the overall process</​s>​ - this sort of morphed into full development mode; design review was done, however, at the SpokenLID meeting 20 September 2007.
 +** <​s>​Create a Normal Deviate Form axis for JFreeChart</​s>​ - DONE using the DetcurveDataset structure instead in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​341 r341]
 +** <​s>​Feature weight viewer: maxent: p(y|x) = e^SUM(lambda_i * f_i(x,y)) / Z(x)</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​343 r343]
 +** <​s>​Check whether Plot class is really necessary / whether an array is needed</​s>​ - OBSELETED by port to Java?
 +** <s>?? Java GUI (is this a blocker?​)</​s>​ - DONE on creation of feature-eng-console branch in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​294 r294]
 +** <​s>​Generate unique models for each language pair.</​s>​
 +** <​s>​Also make outcome-* filenames specific to the languages involved</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​278 r278]
 +** <​s>​Ensure that different colors are used for the points corresponding to the two different experiments on a chart; otherwise, there is potential confusion.</​s>​ - DONE. This task was written with the old, console-based FEC in mind. Now that we're using JFreeChart rather than gnuplot, there'​s no problem.
 +** <​s>​View Features present in a file (be able to restrict based on what the model cares about)</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​343 r343]
 +** <​s>​Play wav files</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​343 r343]
 +** <​s>???​ == duration (seconds)</​s>​ - DONE
 +** <​s>​View Feature Weights (what the model cares about)</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​333 r333]
 +** <​s>​Per-language det curves</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​270 r270]
 +** <​s>​Support for superposed graphs to enable a feature engineer to see progress at a glance</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​270 r270]
 +** <​s>​Before/​after language-pair DET curve matrix ([[gnuplot]]?​):​ Aggregate curve on top with upper-triangular-style diagram showing language pairs below</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​270 r270]
 +** <​s>​Follow up with Bruce to see if he got anything, had any thoughts/​plans</​s>​ - DONE, I'm not sure if this is relevant anymore
 +** <​s>​Generate thumbnails</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​272 r272]. Depends on [[ImageMagick]]'​s '''​convert'''​ command.
 +** <​s>​Verify correctness:​ per-language models are real per-language;​ one-vs-rest is really one-vs-rest.</​s>​
 +*** <​s>​Change filenames to be more descriptive:​ en-ge.model or en-rest.model</​s>​ - DONE in [http://​nlp.cs.byu.edu/​trac/​NIST/​changeset/​278 r278]. The model names are ${RESULT_NAME}_${language}.model
 +*** <​s>​Restrict training and test sets at frontend of pipeline so we only deal with langs of interest. In other words, generate outcome* files on the restricted set of languages before passing to [[resultbuilder.pl]]</​s>​
nlp-private/josh.txt ยท Last modified: 2015/04/23 13:40 by ryancha
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0