search for: ranker

...m Jiarong Wei, a third year undergraduate student in Zhejiang University, China. In GSoC 2014, I will work on Letor module with Hanxiao Sun. Here are some questions I encountered these days, 1. In letor.cc, we have two parts of functions: the training part and the ranking part. I?ll use SVMRanker as an example. The training part basically uses the libsvm library and training data to train a model, then save the model file. The ranking part will calculate score for each document in searching results (MSet) by using the trained model file. My question is for each of our three rank...

...n the source code for deep learning, is it actually included in letor? For the source code https://github.com/rishabhmehrotra/xapian/tree/397034af42c9b1998730160176d219d6f8f38b25/xapian-letor, the last update is about 2 years ago, is that the latest version of the code? For several files such as ranker.cc, evalmetric.cc, there is no implementations of functions, I don't know if they have been implemented somewhere in the module(as far as I read through the source code, I didn't see any). For the tests, are there any benchmark tests on SVM based or listnet models on sample datasets and...

Hi I am on Debian verson 7.7, GCC 4.7.2 and make 3.8.1. I cloned the code : git clone https://github.com/v-hasu/xapian git checkout gsoc2014-evaluation rm xapian-letor/nobootstrap ./bootstrap ./configure --no-documentation make CXXFLAGS='-w' and I get a different set of errors --- ranker.cc: In member function 'std::string Xapian::Ranker::get_cwd()': ranker.cc:80:36: error: 'getcwd' was not declared in this scope make[3]: *** [ranker.lo] Error 1 make[3]: Leaving directory `/home/manugupt1/han/xapian/xapian-letor' make[2]: *** [all-recursive] Error 1 make[2]: Lea...

Hello Rishabh, As per our previous discussion, we will talk each week now about the project. Please update the timeline/todo/journal accordingly so that we can check the status on this saturday for this week. Please remember that this half of the programme is smaller and hence we would need to be very punctual about the goals. Regards, Parth. -------------- next part -------------- An HTML

Hi try this function ive written it should be self explantory but let me know if you have any problems. I've only been using R for a few eeeks so apologies if its not the most efficient! rankit2<-function(rankvar,cuts,data,factor) { ranker<-rankvar ranker<-0 range<-c(1:cuts) range2<-range/cuts range3<-quantile(factor,range2) over<-length(factor) for (i in 1:over){ for (j in 1:cuts) { if (data[[i,1]]<=range3[[j]]) {data[[i,3]]<-j ##test<-j ##print(j) } if (data[[i,3]]>0) break } } out2<-data re...

I use Ubuntu12.04, GNU Make 3.81 and I am not sure if I use some special configure options. The commands I use after clone the repo are as follows: cd xapian ./bootstarp ./configure make cd xapian-letor aclocal -I ../xapian-core/m4 libtoolize --force autoconf autoheader automake --add-missing ./configure make Is this normal? 2014-12-07 22:37 GMT+08:00 James Aylett <james-xapian at

As Olly has already pointed out the 2012 branch is not merged. I think there are some compilation errors in the branch. The code in branch is better refactored. The Ranker and FeatureManager classes are well defined and implemented. Parth. On Wed, Sep 25, 2013 at 9:02 AM, Olly Betts <olly at survex.com> wrote: > On Tue, Sep 24, 2013 at 08:34:10PM +0800, jiangwen jiang wrote: > > There's xapian-letor folder in current master trunk, is this code...

...anager as to what Features sub-classes to initialize. A vector<double> fval(*) function in FeatureManager can operate over vector<Features*> FeatureList to return fvals to the FeatureVector. Maybe your meaning of FeatureList is something different. Can you please explain? > > * Ranker should really be responsible for doing most of the work > currently done by Letor. (Preparing training files, training the > ranking algorithm &c.) > Preparing training file is limited to FeatureVector calculation only. Would there be a specific reason to include it in ranker?...

Hello, My name is Prakhar Pratyush. I am a 4th year Electronics & Communication engineering undergraduate student at Indian Institute of Technology (IIT) Roorkee. My primary areas of Interest are machine learning, and Information Security and I am proficient in C++/Python/JavaScript programming languages as well. I was looking into the past gsoc organizations, and I came across Xapian. I

...r (e.g. ranklist.h, letor_internal.h etc) in xapian-letor/include/letor 2. Rewriting letor.h and placing it in xapian-letor/include so that it does a job similar to xapian.h. The current letor.h contains some methods for which it is not the appropriate file. 3. Creating subdirectories like "ranker", "features" etc. and placing appropriate files in them. 4. Fixing minor bugs that I encounter in completing this process. (For instance, including svmranker in Makefile.am, since it does not do so at present). 5. Ask for a review and merge changes to master. Is this the right wa...

Hello Rishabh, The attached diff file should help you to refactor the existing letor code and plug in your code easily. I have defined the header files with the necessary flow. I have also specified the ranker class, which should be implemented by a new LTR models, ListNet and ListMLE in your case. The evaluation file should be used for validating the performance of the algorithm in the training so which you plan to implement. You should take the corresponding definitions from the letor_internal.cc and c...

...taset and leverage it to predict the rankings of other dataset) or is it for supervised learning? For instance - Xapian currently powers the Gmane search which is by default based on BM25 weighting scheme and now suppose we want to use LETOR to rank the top k retrieved search results, lets take SVMRanker for an example, will it rank the Gmane's search results based on the weights learned from INEX dataset because the client won't be providing any training file. And also I don't think it'll perform good for two datasets of different distributions. So how are we going to use it? PROP...

Hello James, Parth, Following our discussion on IRC and on code review, the way FeatureVector class works needs some discussion. Presently, the FeatureVector class is defined as follows, with a fixed number of feature count (19): class FeatureVector::Internal : public Xapian::Internal::intrusive_base{ friend class FeatureVector; double label; double score;

Dear all, My name is Hanxiao Sun and I am a master student in the University of Chinese Academy of Sciences. I am so glad to be selected by Xapian as the GSoC student. This is my first time to work with open source community. I hope I can do a good job with you in this summer ! In my proposal, I followed the scope of the LTR project on the ideas page. But, after the interview, it seems to need

...oning library we open-sourced at Harvest: http://hrv.st/f03yfi And github: https://github.com/harvesthq/ranked-model A couple reasons this is a nice option for positioning: * ARel! From the ground up this library uses ARel so sorts can be chained like any order() statement. * Multiple rankers for a single model. * Group sorted items by a common column or any scope. * Fewer database writes with a new position storage strategy. * Rspec! This lib is well tested. I think it''s a great solution for a common problem Rails folk often need to address. Any feedback or questions...

HI, Parth: Thanks for your reply. There's xapian-letor folder in current master trunk, is this code merged in 2011? the letor code of 2012 is not merged to the trunk? Regards 2013/9/23 Parth Gupta <pargup8 at gmail.com> > Dear Jiangwen Jiang, > > The current state of the project can be reached from the repo: > http://trac.xapian.org/wiki/GSoC2012/LTR > > I would

...rm, I am still having a 6% CPU in a very simple query (using PHP binding)... e.g.Xapian::Query(movie:(pos=1,wqf=12)) in a 10M docs db only indexed little terms (8.6G size) Matches Estimated 421,057 Time: 0.1850 This one uses 6.3% CPU I wonder, what is the cause of this usage of the CPU? is it the ranker? I already did all I can to minimize costs, what else can I do to prevent / load balance the situtation? Will i better off in using other binding? e.g. python? Will i better off in using distributed search? My goal is to optimize the search, while the doc size will grow to very big,e.g. 100M+ M...

...Letor::Internal intentionally named same as this Internal<https://github.com/xapian/xapian/blob/99eed23ae87bb96f5f9ee7e14a62e9f4af04af0c/xapian-core/include/xapian/intrusive_ptr.h#L44>namespace or is it just a coincidence? 3. If I am not wrong we want to dissolve letor_internal to create SVMRanker. So accordingly letor,h will also change. Right? Thanks -Mayank -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.xapian.org/pipermail/xapian-devel/attachments/20140310/89ac3f89/attachment-0002.html>

...hat I've put in Feature::Internal can actually be part of FeatureList class (maybe as FeatureList::Internal). For the time being, I'll at least make the internal member in Feature as protected, and we can take this up later when we are getting rid of Letor class and putting everything under Ranker. > > It's possible that we could also do the thing Weight does, and have each > Feature report which statistics it needs, so FeatureList can calculate them > each once in advance before calling get_values() on all its Feature > objects. (Weight objects I think get the statistics...

...of many people. See the file AUTHORS in the distribution for the complete list. ---------------------------------------------------------------- Bugs fixed in 5.7 (total 38): 10600 Multiple applications using skinning have distorted/unusable GUI (ABC Amber LIT Converter 2.0, GSA Search Engine Ranker v7.25) 18104 Multiple TTS applications using Microsoft SAPI runtime report 'Speech engine not installed' (CLSID_SpVoice '{96749377-3391-11D2-9EE3-00C04F797396}')(Final Draft, TactileView, Vocaboly 3.0) 19239 Multiple applications crash with division by zero on edit control wit...