SIP_SENSAI code for measure-theoretic inverse sensitivity problems by Simon Tavener
-
Using matlab, open the file
sip-sensai/SENSAI_MuPad/Examples/ODE_examples/Logistic/Logistic.mnAnd set the correct paths at the top (set the last one to wherever you'd like the SIP stuff to go)
From the menu, select
Notebook > Evaluate allto generate "input" .m files
-
In matlab, browse to
sip-sensai/SENSAI_MuPadAnd generate further m files using:
run_MuPAD('Examples/ODE_examples/Logistic')Ignore the warning
-
After generating the files, run
run_gtype('Examples/ODE_examples/Logistic', 0)Figures 15200, 15300 and 15301 are the ones in the paper
-
Do the same as for Logistic, but with HodgkinHuxley/HodgkinHuxley.mn
Figures 15200, 15300 and 15301 are the ones in the paper, but with different scalings for the x-axis (0-12). The y-axis of the black dotted graph should also be set to approx. -84, 34
-
Generate input m files as before using MuPAD and Evaluate all
-
Now MANUALLY MODIFY the generated file
sip-sensai/SENSAI_MuPad/Examples/ODE_examples/HodgkinHuxley/user_inputs.mand change
qtype = 1; stype = 3;to
qtype = 0; stype = 1; -
Generate m-files using
run_MuPAD('Examples/ODE_examples/Logistic')Ignore the warning
-
Check
user_SIP.mand see if the grid size (ngrid) is set to something smallngrid = [10 10 10] -
Now run an analysis using
run_gtype('Examples/ODE_examples/HodgkinHuxley', 15)This should generate a bunch of files in the location specified way back in the MuPAD file.
-
Analyse the results by changing matlab's working directory to
sip-sensai/SIPand running
SIP_SENSAI('/home/michael/sensai', '/home/michael/sensai', 'HodgkinHuxley','HH-SpringBreak', 2, 2)where
home/michael/sensaishould be whatever you specified asOUTPUT_DIRECTORYway back in the MuPAD file.