Maximization problem get stuck for hours

[vishal Kashyap]

Hi All,

I am running optimization problem in Pulp. It is Linear Integer maximization with 2 constraints. I have 2,000 cost curves for that many cusips. I am approximating the cost curve with Random Forest Regressor with upto 5 splits of the range for each cusip. But Pulp gets stuck for hours. 

On the other hand, if I split the problem with 500 cusips in different groups and run the optimization separately, It finishes within seconds for some groups. But for other group of cusips it gets stuck again.  

I have tried increasing CPU and RAM without success.  

Any clues what could be happening? 

What can I do to overcome this problem? 

Any help is highly appreciated.

Thanks,

Vishal

[Franco Peschiera]

Hello!

I think the best would be to share your pulp code in a minimal reproducible example: https://stackoverflow.com/help/minimal-reproducible-example

regards,

Franco