Info

이 질문은 마감되었습니다. 편집하거나 답변을 올리려면 질문을 다시 여십시오.

[Anfis] Optimize FIS

조회 수: 2 (최근 30일)
Enrico
Enrico 2012년 12월 27일
마감: Sam Chak 2025년 4월 15일
Hi, I just created a file .fis .
Now, I have to optimise it through ANFIS .
My original fis is a Mamdani System, so I used the function:
mam2sug( . ).
Then, I had to change some things about my fis file, because Anfis doesnt accept all fis file .
After this, I loaded my fis file in Anfis, loaded my Data and clicked on train.
My problem is that the training doesn't work, the performance doesnt improve.
Is it normal? I dont think. Or Have i made some mistakes ?
This is the results: * ANFIS info: * Number of nodes: 40 * Number of linear parameters: 24 * Number of nonlinear parameters: 44 * Total number of parameters: 68 * Number of training data pairs: 1313 * Number of checking data pairs: 0 * Number of fuzzy rules: 4
Start training ANFIS ...
1 42.6173
2 1.#QNAN
3 1.#QNAN
4 1.#QNAN
5 1.#QNAN
6 1.#QNAN
7 1.#QNAN
8 1.#QNAN
9 1.#QNAN
10 1.#QNAN
Bye Thanks in advance
  댓글 수: 1
chaks
chaks 2013년 6월 7일
Please help me. I have similar kind of problem. Actually I need a mamdani system as I want to build the rule base myself. However, I want to train the system. But not getting any suitable command. Converting my mamdani system to sugeno type using mam2sug() does not help.
Please reply asap!!!

답변 (0개)

이 질문은 마감되었습니다.

제품

Community Treasure Hunt

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

Start Hunting!

Translated by