Hyper-parameters optimization using Bayesian optimization for LSTM regression program specifically for No. of network layers, No. of hidden units, and learning rate.

Dear Experts,

I need to perform Hyperparameters optimization using Bayesian optimization for my deep learning LSTM regression program. On Matlab, a solved example is only given for deep learning CNN classification program in which section depth, momentum etc are optimized. I have read all answers on MATLAB Answers for my LSTM program but no any clear guideline. I need to optimize No. of network layers, No. of hidden units, and learning rate. Please help me for this. Thank You

No benefit found from this MATLAB CNN example, for LSTM regression problem...Deep Learning Using Bayesian Optimization

Answer 1

[Jorge Calvo]

If you have R2020b or later, you can use the Experiment Manager app to run Bayesian optimization to determine the best combination of hyperparameters. For more information, see https://www.mathworks.com/help/deeplearning/ug/experiment-using-bayesian-optimization.html.

[Muhammad Mohsin Khan]

Dear Calvo,

Thanks for replying. Let I see this App. Thank You

Answer 2

[Jorge Calvo]

I thought you would like to know that, in R2021b, we are included an example for training long short-term memory (LSTM) networks using Bayesian optimization in Experiment Manager:

https://www.mathworks.com/help/deeplearning/ug/exp-mgr-sequence-regression-example.html

I hope you find it helpful!

[Muhammad Mohsin Khan]

Dear Calvo. Waooo. This is amazing... Thank you so much.