similar to: Added feature tests to the PL2 pull request

Hello guys. I just ran the updated tests for PL2 and they are not giving the mset order I expect.Now,the thing is, dfr's behavior is a bit hard to predict and so even if I expect a particular order ,it may give another order and still be correct.So,the only way to write correct tests for PL2 is to manually calculate the weight of the documents to decide the expected order.For that,I need to

Hello guys.I just realized that Ive not set the weighting scheme to PL2 in the tests for PL2 and so a default weighting scheme of BM25 is used. I am extremely sorry for this and am updating the tests by setting the weighting scheme to PL2. -Regards -Aarsh -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hello guys.I am working on implementing the PL2 weighting scheme of the DFR framework by Gianni Amati. It uses the Poisson approximation of the Binomial as the probabilistic model (P), the Laplace law of succession to calculate the after effect of sampling or the risk gain (L) and within document frequency normalization H2(2) (as proposed by Amati in his PHD thesis). The formula for w(t,d) in

Hi olly, I rewrote the test for PL2 after adding code to deal with negative weights. It passes on all backends other than inmemory . I see a different value of weight for inmemory backend. The code to calculate the lower bound is implemented in init().Please can I get some help with this ? -Regards -Aarsh -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hey guys,hi.:) I've finished implementing the PL2 scheme . The bounds I have implemented for it are as good as I could, given the nature of the scheme and my mathematical skills.However,tight bounds for other named DFR schemes will be easier to implement because their forumlas are quite simpler compared to PL2 . Will send in a pull request in a couple of days once I'm done with the tests

Hello guys.I have sent a pull request for the code and tests of the Tf-Idf weighting scheme. Please do let me know if any changes are required.Meanwhile,Ill begin working on implementing normalizations which require additional statistics and on the DFR schemes. https://github.com/xapian/xapian/pull/6 On Tue, Feb 26, 2013 at 5:30 PM, <xapian-devel-request at lists.xapian.org>wrote: >

Hello guys.As discussed on IRC,I have sent a pull request for a test for testing TradWeight with an Rset. On Fri, Mar 1, 2013 at 5:30 PM, <xapian-devel-request at lists.xapian.org>wrote: > Send Xapian-devel mailing list submissions to > xapian-devel at lists.xapian.org > > To subscribe or unsubscribe via the World Wide Web, visit >

Hey guys,I have added a python indexer example to the SampleCode page of our wiki.Please do have a look.The code can also be found here :- https://github.com/aarshkshah1992/xapian/blob/efcf443527b74326119bbc0935fc41a002ce60db/xapian-bindings/python/docs/examples/simpleindexgrep.py/ Thanks :) -Regards -Aarsh -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hello Guys. I have updated the code,tests,documentation,makefile entries and the registry entry of the* *TfIdf patch as per the feedback.Please do let me know if any additional changes are required before the patch can be merged, -Regards -Aarsh On Sun, Mar 3, 2013 at 2:50 PM, aarsh shah <aarshkshah1992 at gmail.com> wrote: > Hello guys.I have sent a pull request for the code and

Hello guys,I have added code for tfidf to the weight.cc file in omega/ . Here is the patch : - https://github.com/aarshkshah1992/xapian/commit/5ff41a15f574e6780cc61e67e7f3da3d97ff4ec8 It compiles well and I think it'll work well. Here's the link to the documentation file omegascript.rst where I've added tfidf.

Hi everyone, I am thinking of working on the following ideas for my GSOC proposal based on my discussions with Olly and my own understanding. Rather than focusing on an entire perftest module, I have decided to focus on implementing performance tests for weighting schemes based on a wikipedia dump and in addition to that, build a framework to measure the

Hello, I am beginning work on the perf test module. The initial steps that I aim to accomplish are :- -> Download the wikipedia dumps for multiple languages . -> Write python scripts to tokenize the dump (will probably use something like nltk which has powerful inbuilt tokenizers) -> Discuss and finalize the design of the search and query expansion perf tests as I want to complete them

Hi Parth, I implemented DFR algorithms in Xapian as a part of GSOC last year under the mentorship of Olly. This year, I want to work on analyzing and optimizing the performance of the DFR algorithms and comparing them with BM25.I also want to work on profiling the query expansion schemes and test the relevance(precision and recall) / speed(time taken) of the

Hey guys,I am a 3rd year Computer Science undergrad student.I a extremely interested in contributing code to the XAPIAN project. The work you people do sounds extremely fascinating and interesting.Can someone just give me a brief overview of how to proceed ?. I Can code in C,C++ and Python and have experience in Natural Lanuage Processing.Am also quite comfortable with NLTK and using Wordnet.Am

Hello Dan and Olly, this is the code for the base class for query expansion that I have written. The code will not compile as I have written only the base class until now. Have yet to use it. Please do tell me what you think of the base class and what changes you suggest I should make before I move forward with the project. https://github.com/xapian/xapian/pull/23 -Regards -Aarsh --------------

Hello there, I'm trying to do lots of plots in one for-loop. But somehow ggplot does not evaluate arguments as expected. Here is an example: library(lattice) library(ggplot2) pl <- list() pl2 <- list() cDat <- as.data.frame(cbind(x1=0:100,x2=0:10,x3=1:20)) for(obs in c("x1", "x2")) { pl[[obs]] <- xyplot(cDat[,obs] ~ cDat[,"x3"], main=obs)

Hello, I'm trying to generate a log-log plot with ggplot. Within this plot, I would like to draw a straight line with the help of ggabline which does not work. The code is pl1 <- qplot(correlFunc, correlFunc.ref, data=all, log=c("xy")) pl1 <- ggabline(pl1, slope=1, intercept=0) print(pl1) ## no line in plot :( I can produce plots as I want, if I transform all values before

Hi, Here's a problem in openssh, some logs, and a very minor patch that cures this: Issue: (open)ssh client WILL NOT talk to F secure SSH-2.0-2.1.0pl2 client S/W version: openssh-2.3.0p1 client O/S version: SunOS 5.7 Generic_106541-11 sun4u sparc server S/W version: SSH-2.0-2.1.0pl2 server O/S version: SunOS 5.7 Generic_106541-11 sun4u sparc Log/Details: : % telnet <mymachine> 22

Dear all, let consider the following function: Fun1 <- function() { library(lattice) plot1 <- 1:10~1:10 pl1 <- xyplot(plot1) return(pl1$call$x) } In R 2.5.0 (or older version) we have > Fun1() plot1 but starting from R 2.5.1 until the latest R 2.8.0 we obtain instead > Fun1() NULL because pl1$call seems to be equal to xyplot() without arguments. Something like

Hello guys :) I have sent a pull request for the Tf-Idf Weighting scheme incorporating as many normalizations as I could with the help of statistics currently available from Xapian::Weight . Please let me know what you'll think about it. I used the weighting scheme in a simple searcher and it did a fine job with it. I have no experience with writing tests for features like this.Please give me