File Exchange

image thumbnail

Deep Learning Toolbox For Classification and Regression

version 11.11.0.0 (300 KB) by BERGHOUT Tarek
deep learning Toolbox with multiple types stacked autoencoders is presented in this work with a full toolbox of convolutional neural net

81 Downloads

Updated 23 Apr 2019

View License

Using Deep learning toolbox based Extreme Learning Machine you can reconstruct any training or testing inputs by mapping them into another space with another dimensions where new intresting features can apear evry case. And later you will have the possibility to decrease or increase the dimensionality of your inputs using stacked auto-encoders, and this is means that this time you can tune all the parameters of the hidden layers entirely not only the output weights beta as mentioned in basic ELM theories for single hidden layer feedforward neural network.
Tuning all the parameters of the hidden layers using stacked auto encoders allows you to guard the best networks generalization and approximation.
concerning stacked autoencoders, I have developed many types of them listed as follow:
- Restricted Boltzmann machines.
- Autoencoders.
- Sparse autoencoders.
- Denoising autoencoders.
- Contractive autoencoders.
- any type you want,can be prepared for you, just contact me via: berghouttarek@gmail.com
you have just to add one line in the function DBN_ELM with any type you want, and then your model will be ready to train.
In addition to that a full toolbox of convolutional neural net is availabal, it can be yours, just contact me.
the code also is manged for the user to be able to add any type of stacked AEs he needs or activation function and other parameters to train his deep net.
for any referances you can chek the ones that i used in this work :
[1] J. Cao, K. Zhang, M. Luo, C. Yin, and X. Lai, “Extreme learning machine and adaptive sparse representation for image classification,” Neural Networks, vol. 81, no. 61773019, pp. 91–102, 2016.
[2] Hongming Zhou, Guang-Bin Huang, Zhiping Lin, Han Wang, and Yeng Chai Soh, “Stacked Extreme Learning Machines,” IEEE Trans. Cybern., vol. 45, no. 9, pp. 2013–2025, 2015.
[3] L. le Cao, W. bing Huang, and F. chun Sun, “Building feature space of extreme learning machine with sparse denoising stacked-autoencoder,” Neurocomputing, vol. 174, pp. 60–71, 2016.
[4] N. Zhang, S. Ding, and Z. Shi, “Denoising Laplacian multi-layer extreme learning machine,” Neurocomputing, vol. 171, pp. 1066–1074, 2016.
[5] C. M. Vong, “Local Receptive Fields Based Extreme Learning Machine,” IEEE Comput. Intell. Mag., vol. 10, no. 2, pp. 18–29, 2015.

Cite As

BERGHOUT Tarek (2019). Deep Learning Toolbox For Classification and Regression (https://www.mathworks.com/matlabcentral/fileexchange/69925-deep-learning-toolbox-for-classification-and-regression), MATLAB Central File Exchange. Retrieved .

Comments and Ratings (17)

the SAEs_TB function is missing

smsinks

Missing function

Note:
the SAEs_TB functions are not included by the author but instead are for sale from the author via email.

for any type of auto encoders you can download these ones in the links
bellow they are free and available you are only need to modify them to fit the corrunt code , or just read the discription verry well , i mentiond that any type you want,can be prepared for you if you contact me, those who contacted me they didn't leave with empty hands, any way thanks for comments .
https://la.mathworks.com/matlabcentral/fileexchange/66080-autoencoders?s_tid=prof_contriblnk
https://la.mathworks.com/matlabcentral/fileexchange/71257-contractive-autoencoders?s_tid=prof_contriblnk
https://la.mathworks.com/matlabcentral/fileexchange/71115-denoising-autoencoders?s_tid=prof_contriblnk

Sa Mo

why you define "s" in DBN_ELM that is unused? (line 72)
SAEs_TB is a function which is missing in this folder.
Error in DBN_ELM (line 30)
[betaW]=SAEs_TB(train_set,Network_architecture,activF);%training

Error in example (line 14)
[training_Accuracy,Testing_Accuracy]=DBN_ELM(s1,s2,t1,t2,Network_architecture,ActivF,SAEs_type,net_type);% training process

Cyrus

The author hasn't submitted the necessary files to run the example. This code is just a waste of time and should be removed from this online library.

read the discription carefully please

SAEs_TB is a function which is missing in this folder,

Thank you Berghout Tarek for sharing this code. I also got the same error, SAEs_TB is a function which is missing in this folder, Can Berghout Tarek please help in this regards.

Faqih Romi

I have tried running the example.m code, did not change the code, and the comment appears as below, why did the comment appear?

Undefined function or variable 'SAEs_TB'.

Error in DBN_ELM (line 30)
[betaW]=SAEs_TB(train_set,Network_architecture,activF);%training

Error in example (line 14)
[training_Accuracy,Testing_Accuracy]=DBN_ELM(s1,s2,t1,t2,Network_architecture,ActivF,SAEs_type,net_type);% training process

thank u Rahul

Thanks for the information. Highly informative author.

Thanks a lot Tarek for sharing the code.

thanks a lot selina

This toolbox is very good, and the author is very enthusiastic.

Noor Abbas

Jahetbe

Updates

11.11.0.0

work with a full toolbox of convolutional neural net is availabal.

11.10.0.0

new features and comments

11.9.0.0

new features added

11.8.0.0

new deep learning tool box

11.7.0.0

new features added

11.6.0.0

new version

11.5.0.0

new features added

11.3.0.0

new descriptif image

11.2.0.0

picture

11.1.0.0

Just some changes

MATLAB Release Compatibility
Created with R2013b
Compatible with R2014b to any release
Platform Compatibility
Windows macOS Linux

ELM-DBN_demo