Closed wbryant closed 7 years ago
Hi @wbryant,
Thanks for reporting these issues.
Did you run initCobraToolbox
before trying to run the tutorial?
initCobraToolbox
adjusts your paths and update all submodules (where the cplexFVA1271
function is).
The version of the uniformSampling tutorial that you are trying to run has been updated yesterday and does not rely anymore on the ecoli_core model. Please update your COBRAToolbox to get the latest version of the tutorial.
If you still have issues with FastFVA this might be due to your version of MATLAB and/or your macOS version. You may find instructions to generate new mex file here.
If you have any question or issue, please ping me!
cc: @Gpreciat
Thanks @syarra I'll try the steps you suggest and let you know how it goes!
Despite banging my head against this for a day I can't get MATLAB mex to play nice with Xcode, which is to say I have been unable to produce the mex file. Does COBRA Toolbox create cplexFVA1271
? I still cannot find it when I search the Toolbox after running initCobraToolbox
.
Anyway, this could be a system-specific bug, so it would be a waste of time for me to continue fiddling when I could be evaluating another tutorial. I will remove my name from the spreadsheet and hopefully someone else with a working CPLEX can follow it!
@Gpreciat please edit the sampling code to call fluxVariability.m using the default matlab+solver implementation, then only if fastFVA is selected as the implementation option, and only when the installation is verified to be working, then fastFVA is run (try catch), otherwise it should switch back to the default implementation.
I am having great difficulty getting this tutorial to run for me. I am using MATLAB 2015b on Mac OSX El Capitan. I cloned the master repo to install it fresh on my system. Since fastFVA requires CPLEX I also installed and linked the latest CPLEX 12.71. I wonder if all of these bugs are related to my version of MATLAB, but hacking fastFVA can only get me so far: I have no idea where the function 'cplexFVA1271' is defined.
Bugs so far:
I hereby confirm that I have:
(Note: You may replace [] with [X] to check the box)
EDIT: expanded on yyaxis workaround