How to tuning parameters for SVM to predict BEAM deformation
Dear community memeber
--- Question ---
What direction I have to do to get Nearest predict-Model for Beam deformation.
---BackGround---
It's 1st time for me to use Rapidminer ,so I started simple formulation model.
Beam (section is quad) deformation model looks good for it....
if Force:f,Length:l,quadWidth:h1,quad_height:h2 then formula for delta = (f * l^3)/(3 * 205 * h1 * h2^3 / 12)
I want to predict delta from 4 numberr(f,l,h1,h2)
I try SupportVectorMachine but cannot get better model...
attached file is csv file I used for training.
Please give me advice for direction what to do.
Thank you for reading.
Mai Moritaka
--- Question ---
What direction I have to do to get Nearest predict-Model for Beam deformation.
---BackGround---
It's 1st time for me to use Rapidminer ,so I started simple formulation model.
Beam (section is quad) deformation model looks good for it....
if Force:f,Length:l,quadWidth:h1,quad_height:h2 then formula for delta = (f * l^3)/(3 * 205 * h1 * h2^3 / 12)
I want to predict delta from 4 numberr(f,l,h1,h2)
I try SupportVectorMachine but cannot get better model...
attached file is csv file I used for training.
Please give me advice for direction what to do.
Thank you for reading.
Mai Moritaka
Find more posts tagged with
Sort by:
1 - 2 of
21
Altair EmployeeHi,
if you already know the function, why do you want to build a model?
BR,
Martin
>Martin
thanks for reply!
its only my practice (study?) to build prescribe model.
if function is known , create test-train-data is easy.
In my opinion, if I cannot build formula model, real world model is more difficult to build.
thanks for reply!
its only my practice (study?) to build prescribe model.
if function is known , create test-train-data is easy.
In my opinion, if I cannot build formula model, real world model is more difficult to build.