Why identical outputs despite different inputs to a machine learning models

3 次查看（过去 30 天）
RICHARD BIGEGA
RICHARD BIGEGA 2019-3-4
评论： RICHARD BIGEGA 2019-3-6
why do I get the same predicted values despite having different inputs in a SVM model. For example, suppose the training data is matrixA, and the Two different Prediction data are MatrixC and MatrixD. Why is the predicted values identical?
A=trainedModel.predictFcn(matrixA);
B=trainedModel.predictFcn(matrixB);
MatrixB is a concatination of MatrixA with another Matrix---
I appreciate any help I can get?

请先登录，再回答此问题。

回答（1 个）

Bernhard Suhm
Bernhard Suhm 2019-3-6
Of course it could predict the same category for different kinds of inputs, especially if there aren't a lot of categories... Or for some reason, only the features represented by matrixA determined the final model, and the training of B ignored the additional features provided in the concatenated matrix. Or what am I missing?
  1 个评论
RICHARD BIGEGA
RICHARD BIGEGA 2019-3-6
My exercise is not classification. I am working on forecasting future values given a set of know past values X(x). I ampuzzled because when I train a model I get the same outputs in some instances dispite that the sets on which I am doing predictions are different. Below is an example of the matrix whicth the outputs in redCapture.JPG

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Gaussian Process Regression 的更多信息

标签

产品


版本

R2018a

Community Treasure Hunt

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

Start Hunting!

Translated by