From joubert.leo at gmail.com Mon Sep 5 10:07:41 2016 From: joubert.leo at gmail.com (=?UTF-8?B?TMOpbyBKb3ViZXJ0?=) Date: Mon, 05 Sep 2016 08:07:41 +0000 Subject: [Traminer-users] R crash creating a big event sequence object Message-ID: Hi, I am experimenting some difficulties when attempting te create a (quite) big event sequence object. My datas contiains 451 sequences, with three events possible, and a sequence length from 10 to 30000. When I launch this code : library(TraMineRextras) seq <- seqecreate(id = data$id, timestamp = data$timestamp, event = data$event) My RStudio session wait some minutes and then crash after "encountering fatal error". I am sorry to not be able to give you more information. Here's my sessionInfo R version 3.3.1 (2016-06-21) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows >= 8 x64 (build 9200) locale: [1] LC_COLLATE=French_France.1252 LC_CTYPE=French_France.1252 LC_MONETARY=French_France.1252 [4] LC_NUMERIC=C LC_TIME=French_France.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached): [1] magrittr_1.5 assertthat_0.1 htmltools_0.3.5 tools_3.3.1 yaml_2.1.13 [6] tibble_1.1 Rcpp_0.12.6 stringi_1.1.1 rmarkdown_1.0.9005 knitr_1.14 [11] stringr_1.0.0 digest_0.6.10 evaluate_0.9 Thank you by advance for your help -- ------------------------------------------------------------------------------------ L?o Joubert Doctorant en sociologie Laboratoire d'Economie et de Sociologie du Travail - LEST, UMR 7317 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matthias.Studer at unige.ch Mon Sep 5 10:17:09 2016 From: Matthias.Studer at unige.ch (Matthias Studer) Date: Mon, 5 Sep 2016 08:17:09 +0000 Subject: [Traminer-users] R crash creating a big event sequence object In-Reply-To: References: Message-ID: <54052579d3c3423eb9acbeb3e6b3fad4@wuniex13-5.isis.unige.ch> Hi, I guess that this is related to a memory issue, because of the size of the data. Anyway, all the algorithms implemented in TraMineR cannot deal with long sequences (sequence with more than 3-4 events per sequences). I therefore suggest you look for another software. Sorry for this answer. Best, Matthias De : traminer-users-bounces at lists.r-forge.r-project.org [mailto:traminer-users-bounces at lists.r-forge.r-project.org] De la part de L?o Joubert Envoy? : lundi 5 septembre 2016 10:08 ? : traminer-users at lists.r-forge.r-project.org Objet : [Traminer-users] R crash creating a big event sequence object Hi, I am experimenting some difficulties when attempting te create a (quite) big event sequence object. My datas contiains 451 sequences, with three events possible, and a sequence length from 10 to 30000. When I launch this code : library(TraMineRextras) seq <- seqecreate(id = data$id, timestamp = data$timestamp, event = data$event) My RStudio session wait some minutes and then crash after "encountering fatal error". I am sorry to not be able to give you more information. Here's my sessionInfo R version 3.3.1 (2016-06-21) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows >= 8 x64 (build 9200) locale: [1] LC_COLLATE=French_France.1252 LC_CTYPE=French_France.1252 LC_MONETARY=French_France.1252 [4] LC_NUMERIC=C LC_TIME=French_France.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached): [1] magrittr_1.5 assertthat_0.1 htmltools_0.3.5 tools_3.3.1 yaml_2.1.13 [6] tibble_1.1 Rcpp_0.12.6 stringi_1.1.1 rmarkdown_1.0.9005 knitr_1.14 [11] stringr_1.0.0 digest_0.6.10 evaluate_0.9 Thank you by advance for your help -- ------------------------------------------------------------------------------------ L?o Joubert Doctorant en sociologie Laboratoire d'Economie et de Sociologie du Travail - LEST, UMR 7317 -------------- next part -------------- An HTML attachment was scrubbed... URL: From bert.carremans at bnpparibasfortis.com Wed Sep 14 14:32:08 2016 From: bert.carremans at bnpparibasfortis.com (bert.carremans at bnpparibasfortis.com) Date: Wed, 14 Sep 2016 12:32:08 +0000 Subject: [Traminer-users] Flagging for the most discriminating subsequences on another data set in Traminer Message-ID: <93429EEFDB63114D9C6088A2B74023D93AEEDFB3@SPMW0324.mail.shared.fortis> I am working with Traminer to find subsequences that are discriminating between two groups. My original data set CJ contained events of about 110K people. To limit the analysis time, I took a sample of 10K people named CJ_SAMPLE. On this sample I searched for the most frequent subsequences with the following code: fsubseq_sample <- seqefsub(cj_sample.seq, pMinSupport = 0.05, maxK= 4) This list of subsequences is then analyzed for the findding the most discriminating subsequences. discr = seqecmpgroup(fsubseq_sample, group=cj_tbl$groupID, method="bonferroni") Now I want to create dummy variables on my original data set CJ to flag whether these discriminating subsequences occurred for each person in CJ. So I created the list of sequences with cj.seq <- seqecreate(id=cj$person_id, time=cj$date_in, event=cj$event) I was able to do this on CJ_SAMPLE with t = seqeapplysub(discr[0:51], method = 'presence') But how can I do this on the orignal data set CJ? Thanks! Best regards, Bert ====================================================== BNP Paribas Fortis disclaimer: http://www.bnpparibasfortis.com/e-mail-disclaimer.html BNP Paribas Fortis privacy policy: http://www.bnpparibasfortis.com/privacy-policy.html ====================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: