Can't get relaxed_constants to work...

[1ozturkbe]

Since last night, haven't been able to get anything to run. relaxed_constants keeps throwing the following error:

Traceback (most recent call last):
  File "C:\Users\Berk\Anaconda2\lib\site-packages\IPython\core\interactiveshell.py", line 2666, in safe_execfile
    self.compile if kw['shell_futures'] else None)
  File "C:\Users\Berk\Anaconda2\lib\site-packages\IPython\utils\py3compat.py", line 276, in execfile
    exec(compiler(scripttext, filename, 'exec'), glob, loc)
  File "C:\Users\Berk\Dropbox (MIT)\MIT Graduate School\D8\d8\SP_aircraft.py", line 543, in <module>
    m_relax = relaxed_constants(m, None, ['ReqRng'])
  File "C:\Users\Berk\Dropbox (MIT)\MIT Graduate School\D8\d8\relaxed_constants.py", line 23, in relaxed_constants
    constsrelaxed = ConstantsRelaxed(model, include_only, exclude)
  File "c:\users\berk\dropbox (mit)\mit graduate school\code\gpkit\gpkit\constraints\relax.py", line 134, in __init__
    if value == 0:
  File "c:\users\berk\dropbox (mit)\mit graduate school\code\gpkit\gpkit\nomials\nomial_math.py", line 397, in __eq__
    return MonomialEquality(self, "=", other)
  File "c:\users\berk\dropbox (mit)\mit graduate school\code\gpkit\gpkit\nomials\nomial_math.py", line 581, in __init__
    ScalarSingleEquationConstraint.__init__(self, left, oper, right)
  File "c:\users\berk\dropbox (mit)\mit graduate school\code\gpkit\gpkit\nomials\nomial_math.py", line 424, in __init__
    self).__init__(Signomial(left), oper, Signomial(right))
  File "c:\users\berk\dropbox (mit)\mit graduate school\code\gpkit\gpkit\nomials\nomial_math.py", line 120, in __init__
    raise ValueError("each c must be positive.")
ValueError: each c must be positive.

so confused... I checked the subs, and booleans, and they seem ok. Any ideas @mayork? I'm surprised you aren't having the same issue.

[mayork]

no idea everything is working great for me right now...i would suggest pulling gpkit master and running straight from command line (i..e not in an interactive shell) and see what happens

[mayork]

every file gives this? how are you running them, importing run_XXX into an interactive enviornment or running main on SP_aircraft?

[1ozturkbe]

I pulled gpkit, that's not the issue. And neither main nor the functions work. And I tried different terminals, no difference.

[mayork]

what commit of D8 are you on

[mayork]

is this like your mosek again or something lol?

[1ozturkbe]

I am so confused. I am in your latest commit. Mosek seems ok, the issue is in nomial_math. wtf?

[mayork]

i don't know what to tell you, i'm having zero issues this morning

[bqpd]

Huh, might be a recent PR; just to check, you're both on master?

On May 10, 2017 10:18, "Berk Ozturk" notifications@github.com wrote:

I am so confused. I am in your latest commit. Mosek seems ok, the issue is in nomial_math. wtf?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/hoburg/d8/issues/65#issuecomment-300496587, or mute the thread https://github.com/notifications/unsubscribe-auth/ABagGOSFNlH2Kdgk74DyzIXHaS414nauks5r4cdCgaJpZM4NWtld .

[1ozturkbe]

Appreciate the help here @bqpd. Let me know if you have any ideas, because it sucks not being able to write when there is so much work to do :P.

[1ozturkbe]

Actually, since the issue is definitely gpkit related, I reverted all of the commits you made two days ago @bqpd, and it works. I am currently on 4cce21bc. Let me know if that helps @bqpd. @mayork just as a heads up, don't pull gpkit :P

[mayork]

@bqpd the same error is showing up in our test output on jenkins

[mayork]

@bqpd is there a path forward on this? I really want to avoid not being able to run on master as much as possible (got ourselves into a bit of a bind with that last term)

[bqpd]

@mayork, @1ozturkbe, how can I reproduce this