Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-04-14

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:19 kcranstn joined #opentreeoflife
02:10 jar286 joined #opentreeoflife
02:18 kcranstn joined #opentreeoflife
05:44 kcranstn joined #opentreeoflife
06:05 kcranstn joined #opentreeoflife
07:58 scrollback4 joined #opentreeoflife
11:17 kcranstn joined #opentreeoflife
11:46 kcranstn joined #opentreeoflife
12:32 kcranstn joined #opentreeoflife
12:48 mtholder joined #opentreeoflife
13:10 josephwb joined #opentreeoflife
13:10 josephwb hey mtholder. thanks for fixing that so fast
13:37 mtholder joined #opentreeoflife
13:39 josephwb you there mtholder?
13:39 mtholder yes. just returned.
13:40 josephwb i am going to run the decomposition myself. will work, ya?
13:40 mtholder it should.
13:40 josephwb we just have slightly different newicks than i think you are using (alternate pruning of dups)
13:40 josephwb ok, i will give it a shot
13:40 mtholder I used the ones from your bitbucket reop.
13:40 mtholder repo.
13:41 josephwb oh, then great. saves me some time!
13:41 mtholder I put some notes in https://github.com/OpenTreeOfLife/otcetera/tree/master/supertree/step_1
13:41 mtholder and https://github.com/OpenTreeOfLife/otcetera/tree/master/supertree/step_4
13:42 josephwb looks good. thanks for being so thorough.
13:42 mtholder if you have those 2 directories in otcetera/supertree populated with links to your trees, and you have the otcetera tools on your path, then you should be able to "make" in the supertree subdir
13:43 josephwb haven't tried that yet; only use a small portion of otcetera
13:43 josephwb btw, did you see my email about the unsupported test taking dayas (7 now) to run with the subproblems as input?
13:43 josephwb full newicks is pretty quick
13:44 mtholder that is odd.
13:44 josephwb yeah, i don't know what is going on.  stuck on "reading trees"
13:45 mtholder can you send me the invocation (and files) that it is failing on?
13:45 mtholder presuming that your running it many times, that is...
13:46 josephwb will do
13:47 josephwb having individual files allows monitoring of progress, but i have been catting all into a single file
13:47 josephwb seems like it should drop "trees" like:
13:47 josephwb (ottt12345);
13:49 josephwb (the extra 't' is for 'tenacity')
13:49 mtholder which tool do you mean by "it"
13:49 josephwb find-unsupported-nodes
13:50 josephwb that "tree" is a taxonomic node. don't need to test.
13:50 josephwb it is in the taxonomy tree, and named in the synth (if present)
13:51 josephwb what is your preferred way to get these files? synth tree is big
13:51 mtholder I have the synth from the other day.
13:51 josephwb ok
13:52 josephwb i'll send you the one where the inputs have not been catted
13:52 josephwb [not clear that 'catting' or 'catted' are real words...]
13:53 mtholder they work for me
13:54 josephwb i am a good catter
14:03 josephwb just cleaning up things a bit
14:14 josephwb ok, email sent
14:14 josephwb both 1) concatenated file, and 2) individual files
14:14 josephwb old ones, mine you
14:16 mtholder got it thanks. what is the invocation that you use with the otc-find-unsupported-... ?
14:16 josephwb just a sec
14:17 josephwb otc-find-unsupported-nodes $TAX_TREE $SYNTH_TREE $OTC_TREES &> $OTC_LOG
14:17 josephwb from a bash script
14:17 mtholder where OTC_TREES is all of the subproblems?
14:17 josephwb $OTC_TREES is either the single file, or dir/*.tre
14:18 josephwb make sense?
14:19 mtholder probably a memory exhaustion situation or something requiring paging out to virtual memory
14:19 josephwb yeah, i don't know
14:19 josephwb got tons of memory
14:19 mtholder do you need to do it this way other than running with the original trees?
14:19 josephwb don't know how to allocate more to C++ (like with java)
14:19 josephwb no, it is just part of the pipeline
14:20 josephwb original files is fine
14:20 josephwb (can only get it to work that way)
14:20 mtholder It should take what the OS will give it (a C++ app) - that tool was never designed for a couple of huge trees and many thousands of small ones.
14:20 josephwb alright; i will stick with original 500 newicks
14:21 mtholder If you prune the synth and the taxonomy down to the leafset of a subproblem, then you should be able to check just that subproblem
14:21 mtholder (if you need finer grained tests).
14:21 josephwb yeah, i got weird problems trying to do that
14:22 josephwb pruning both taxonomy tree and synth tree to input leaf set, taxonomy tree and synth tree have different leaf sets
14:22 josephwb so, program stops
14:22 josephwb but the 500 newicks runs ok
14:56 kcranstn joined #opentreeoflife
15:48 jar286 joined #opentreeoflife
21:05 jar286 joined #opentreeoflife
21:29 josephwb joined #opentreeoflife
22:56 kcranstn joined #opentreeoflife

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary