Skip to main content

how to solve numerical issues

Answered

Comments

4 comments

  • Riley Clement
    • Gurobi Staff

    Hi 建城 张,

    Maliheh has given a very detailed answer to this question when it was asked previously: https://support.gurobi.com/hc/en-us/community/posts/4676253771025-Large-matrix-coefficients-and-large-RHS-warning

    Also, the section of our reference manual Does my model have numerical issues? and the pages that follow it will be useful reading.

    - Riley

    -1
  • 建城 张
    • Gurobi-versary
    • First Comment
    • First Question

    hi Riley:

    So happy that u reply me soon. i had read the article u shown me above.but unfortnately,i had try the solving ways the article given and it didnt works. could u please give me some detailed solving answer.i will appreciate it .the detailed error are shown below

    Gurobi Optimizer version 11.0.0 build v11.0.0rc2 (win64 - Windows 10.0 (19045.2))

    CPU model: AMD Ryzen 5 4500U with Radeon Graphics, instruction set [SSE2|AVX|AVX2]
    Thread count: 6 physical cores, 6 logical processors, using up to 6 threads

    Optimize a model with 724 rows, 724 columns and 1445 nonzeros
    Model fingerprint: 0x5ee04fa0
    Variable types: 0 continuous, 724 integer (724 binary)
    Coefficient statistics:
      Matrix range     [1e+00, 1e+00]
      Objective range  [6e+02, 1e+10]
      Bounds range     [1e+00, 1e+00]
      RHS range        [1e+00, 1e+00]
    Warning: Model contains large objective coefficients
             Consider reformulating model or setting NumericFocus parameter
             to avoid numerical issues.
    Found heuristic solution: objective 2.187660e+08

    Explored 0 nodes (0 simplex iterations) in 0.00 seconds (0.00 work units)
    Thread count was 1 (of 6 available processors)

    Solution count 1: 2.18766e+08

    Optimal solution found (tolerance 1.00e-04)
    Best objective 2.187660000000e+08, best bound 2.187660000000e+08, gap 0.0000%

    your sincerely

    -zhang

    0
  • Riley Clement
    • Gurobi Staff

    Hi Zhang,

    The log indicates that the solver is finding an optimal solution very quickly - almost no time at all. I would not worry about the warning for this model.

    - Riley

    0
  • 建城 张
    • Gurobi-versary
    • First Comment
    • First Question

    hi Riley,

    words can not express how  grateful i am to u。i try to optimize my model now and i hope the work will

    pay off. if u have any suggestions or improved guidence to my codes or my model.please tell me asap.i will appreciate it.hope u have a good day. best wish for u. my bro. 

    yours sincerely

    -zhang

    1

Please sign in to leave a comment.