Build failing for Chang Model lectures

[duncanhobbs]

The execution test for Credible Government Policies in Chang Model and Competitive Equilibria in the Chang Model is failing.

There is a ValueError triggered when ch1.solve_sustainable() is called. It looks like this might be triggered by scipy/optimize.

~/anaconda3/lib/python3.7/site-packages/scipy/optimize/_linprog_util.py in _check_result(x, fun, status, slack, con, lb, ub, tol, message)
   1325         # nearly basic feasible solution. Postsolving can make the solution
   1326         # basic, however, this solution is NOT optimal
-> 1327         raise ValueError(message)
   1328 
   1329     return status, message

ValueError: The algorithm terminated successfully and determined that the problem is infeasible.

[jstac]

Thanks for the report @duncanhobbs. These are big, complex models that we have in mind to re-write. Would you mind to give us some information on your execution environment?

[duncanhobbs]

I didn't actually build the lectures, I just looked at the lectures on Google Chrome from the latest build of the website on October 13. I am running windows 7, and am working trying to get Make to work properly.

.

[jstac]

Roger that. Thanks for pointing it out. @mmcky , it looks like a problem in the linear optimization routine. Have we seen this one before?

[mmcky]

@jstac this is one of the two lectures that stochastically fails to converge. It seems that scipy routine sometimes fails to converge (due to infeasibility)? I'll be upgrading anaconda=2019.10 then will recheck coverage for this lecture.