MATLAB Answers

What to do when training doesn't fit training data well?

4 views (last 30 days)
Jordan Pauls
Jordan Pauls on 28 May 2020
Answered: Nagasai Bharat on 29 Sep 2020
Hello,
I've been working on a Deep Learning system to learn some simple communication system properties and I'm having trouble with training/predicting. First, the training process quickly goes to zero, which would indicate that it has fit the data well, or even overfit the training data.
However, when using the predict function on the training data to double check, the plot indicates that the network does not predict the data well:
And cross validation prediction is even worse:
Does anyone have a guess as to why the training process shows an error close to zero, but both training set and cv set prediction is poor?
Thanks!
  3 Comments
vaibhav mishra
vaibhav mishra on 30 Jun 2020
maybe your model is getting overfit.
try to adopt some dropout and regularization in your model.

Sign in to comment.

Answers (2)

vaibhav mishra
vaibhav mishra on 30 Jun 2020
maybe your model is getting overfit.
try to adopt some dropout and regularization in your model.

Nagasai Bharat
Nagasai Bharat on 29 Sep 2020
Hi,
This issue may be mainly due to the overfitting of the data with respect to your model. As dropout is already applied while training you could use regularization methods (E.g. Batch Normalization, L2 Norm) to the model while training. Also, you could try altering the learning rate so that the model does not overfit.
You can refer to the following documentation and other similar training functions.

Products


Release

R2020a

Community Treasure Hunt

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

Start Hunting!