computeIIS and default lower bound of 0

回答済み

Dimitar Dimitrov

• 2020年04月01日 15:52

When computeIIS is called and the infeasable constraints are written to ilp file on a model that is infesible due to default lower bound of a variable being 0, the default lower bound doesn't appear in the ilp file. That is super misleading to the user. 
Example:

m = Model('sth')
x0 = m.addVar(name='x0')

m.addConstr( x0, GRB.LESS_EQUAL, -1 )

m.computeIIS()

m.write('sth.ilp' )

 sth.ilp:
\ Model sth_copy
\ LP format - for model browsing. Use MPS format to capture full model detail.
Minimize

Subject To
R0: x0 <= -1
Bounds
End

Please fix this !!!

0

• 正式なコメント

  

  Simranjit Kaur

  Gurobi Staff

  • 2024年12月06日 00:48

  This post is more than three years old. Some information may not be up to date. For current information, please check the Gurobi Documentation or Knowledge Base. If you need more help, please create a new post in the community forum. Or why not try our AI Gurobot?.
• 

  Matthias Miltenberger

  Gurobi Staff

  • 2020年04月01日 17:58

  Hi Dimitar,

  The file generated by the computeIIS method follows the common conventions of LP files. Variables in LP file format are nonnegative by default and the lower bound of zero is not specified. The same applies for creating the variable and not passing a lower bound.

  There is nothing wrong and this is most likely not going to be changed anytime soon.

  Cheers,
  Matthias

  0

投稿コメントは受け付けていません。