biologyguy / RD-MCL / 453

Builds Branch Commit Type Ran Committer Via Coverage
453 develop Fix bug affecting Worker jobs with new limited graph calc. push biologyguy travis-ci pending completion   set done
452 develop Fix recent issue with test_rdmcl.py::test_full_run There was disagreement between TravisCI and local. push biologyguy travis-ci pending completion   set done
451 develop Fix recent issue with test_rdmcl.py::test_full_run There was disagreement between TravisCI and local. push biologyguy travis-ci pending completion   set done
450 develop Stop calculating ALL all-by-all graphs... There was so much extra, unnecessary, work being done. There is no need to calculate a graph for every single cluster identified by MCL. So stop, and only calculate those that are accepted just prior to... push biologyguy travis-ci pending completion   set done
449 develop Minor push biologyguy travis-ci pending completion   set done
447 develop Some tidying up of test_merge_orthogroups push biologyguy travis-ci pending completion   set done
446 develop Refactor some helper unit tests for better type consistency push biologyguy travis-ci pending completion   set done
445 develop Replace .get_value() calls with .at[] and account for float32 issues push biologyguy travis-ci pending completion   set done
443 develop Fix a bunch of helper tests Switch to proper array comparisons instead of comparing strings. This is more consistent across systems. Pull #12 web-flow travis-ci pending completion   set done
442 develop Set max threads for OpenBLAS push biologyguy travis-ci pending completion   set done
440 develop Strip out MCL trans matrix hashing and some refactoring Set dtype for MCL arrays to float32 to reduce memory and increase the speed of convergence Only call finalize_transition_matrix() when there is a value in the array >0 and <1. Update tes... push biologyguy travis-ci pending completion   set done
438 develop Refactor and simplify finalize_clusters into finalize_transition_matrix It's now a static method and the logic slow is much simpler Also added unit test to cover it push biologyguy travis-ci pending completion   set done
437 develop Catch MCL runs that do not stabilize, selecting the 'best' cluster Hash the np.array and keep track to stop if a transition matrix is observed a second time, and stop if so. Also, stop saving all substates, which was taking up gobs of RAM for l... push biologyguy travis-ci pending completion   set done
436 develop Catch MCL runs that do not stabilize, selecting the 'best' cluster Hash the np.array and keep track to stop if a transition matrix is observed a second time, and stop if so. Also, stop saving all substates, which was taking up gobs of RAM for l... push biologyguy travis-ci pending completion   set done
435 develop Catch MCL runs that do not stabilize, selecting the 'best' cluster Hash the np.array and keep track to stop if a transition matrix is observed a second time, and stop if so. Also, stop saving all substates, which was taking up gobs of RAM for l... push biologyguy travis-ci pending completion   set done