Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

[fix1048] Not computing sensitivities to relaxed constants

See original GitHub issue

[For this, I am running commit d39b91a9 of hoburg/d8, in the fix1048 branch, which tries to resolve issues with NEAR_OPTIMAL solutions] Trying to solve the model with relaxed constants, and calling for the sensitivities, returns this:

sol['sensitivities']
Out[7]: 
{'constants': {\theta_{db}_Mission, Aircraft, Fuselage: -0.00020274757236380924,
  M_{takeoff}_Mission, Aircraft, Engine, EngineConstants: -0.017451664358190976},
 'la': array([  1.00000000e+00,   4.11813259e-06,   3.42431630e-06, ...,
          1.00000729e+01,   9.99992711e+00,   0.00000000e+00]),
 'nu': array([  9.93715754e-01,   6.28424712e-03,   1.13610955e-06, ...,
          1.00000729e+01,   9.99992711e+00,   0.00000000e+00])}

The only sensitivities shown are for the constants that are in the ‘exclude’ category of the relaxed_constants. It seems like the sensitivities for the other variables are not even being computed. Please help.

Issue Analytics

State:
Created 6 years ago
Comments:14 (14 by maintainers)

Top GitHub Comments

1reaction

mayorkcommented, Apr 13, 2017

i don’t see any merge of relaxed constants to master…can we merge the PR and make an issue to add a unit test after merging relaxed constants into master?

0reactions

mayorkcommented, Apr 13, 2017

yeah can we merge the fix1048 branch now…can’t really write a unit test for it until relaxedconsts is added to gpkit