Clear Filters
Clear Filters

I need a code for sensitivity analyses and parameter optimization

2 views (last 30 days)
Hey guys,
I need to run a sensitivity analyses and parameter optimization on my modeled data, but the situation is little bit complicated!
The "pipe score.txt" contains 1083 pipes with 6 different scores between 1 to 4. These 6 scores should be multiplied by their weights and then sum up as a total score for each pipe.
Since the validation data ("validation class.txt") is categorized to 4 classes, we should divide the pipe's score by the total weight and then round it down to attach it to a integer class number between 1 to 4.
finally, the designated pipe class (between 1 to 4) should be subtracted (absolute value) from its validation class individually and then be averaged for entire model as the "calculated error" with one specific weight distribution.
the efficiency of the model will then be equal to: (1-"calculated error”) *100
for the weight optimization and sensitivity analyses, I would like to change the weight parameters from 0 to 10 in 1 step for all 6 weight parameters. in other words, I need to run 1 million (10^6) iterations and calculate the model efficiency for each iteration.
it would be great if you write the MATLAB code for me, i'm not good with it <3
Thank you and Cheers.

Answers (1)

Walter Roberson
Walter Roberson on 18 Dec 2022
Our answer is: We are not going to write your code for you.
If you ask questions about which function performs a specific task, or what an error message means, or how a particular part of MATLAB works, then we will attempt to assist you with that.
We help people understand MATLAB; we do not write their (non-trivial) program for them.
So start by coding as much as you already understand the steps for. Then when you get stuck, ask us for help on the concept or error that you get stuck on.

Categories

Find more on Get Started with Optimization Toolbox in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!