Gurobi Error 10005: Unable to retrieve attribute 'Pi'

回答済み

ANIL KAYA

2021年12月20日 07:48

• 
• 
• 

I am solving my model with Julia program (Gurobi. solver). I am working on SDDP. My random variables are multiplied by decision variables. Due to this, the model seems as Quadratic. In order to set Gurobi solver as Quadratic and get dual variables, I put code into my model as follow: (optimizer = with_optimizer(Gurobi.Optimizer, QCPDual=1,OutputFlag=0, gurobi_env),)

My algorithm has 10 iterations. When I run, the algorithm is stopped after 3,4 or 5 iterations, due to this error (Gurobi Error 10005: Unable to retrieve attribute 'Pi').

I know that this problem indicates that my model is infeasible. Why I am getting after severaal iterations ? This is still related to infeasibility in my model ? For example, when I set 3 iteration numbers, I received optimal solutions.

0

• 正式なコメント

  

  Simranjit Kaur

  • Gurobi Staff

  2025年06月23日 02:42 編集済み

  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?.
• 

  Jaromił Najman

  • Gurobi Staff

  2021年12月21日 08:40

  If your model is infeasible, then it is not possible to compute Pi values. Thus, accessing them in this case raises an error.

  You should always check the optimization status and then proceed accordingly

  if model.status == GRB.INFEASIBLE:
    # handle infeasibility
  elif model.status == GRB.OPTIMAL:
    # handle optimality

  Best regards,
  Jaromił

  0

投稿コメントは受け付けていません。