libORF-master
所属分类：matlab编程
开发工具：matlab
文件大小：338KB
下载次数：43
上传日期：2015-07-06 16:52:21
上 传 者：
longbowking
说明： 针对各种机器学习，深度学习领域的一个matlab工具包
(A machine learning library focused on deep learning.Following algorithms and models are provided along with some static utility classes:
- Naive Bayes, Linear Regression, Logistic Regression, Softmax Regression, Linear Support Vector Machine, Non-Linear Support Vector Machine (with RBF kernel),
Feed-forward Neural Network, Embedding Neural Network, Convolutional Neural Network, Sparse Autoencoders, Denoising Autoencoders,
Contractive Autoencoders, Stacked Sparse Autoencoders, Self-Taught Learner and Restricted Boltzmann Machines are tested with this version.
- Rest of the methods are not tested hence not supplied and the progress is as follows:
+ Deep Belief Nets with Restricted Boltzmann Machines (not tested)
+ Bayes Nets (tested- refactoring)
+ Hidden Markov Models (tested- refactoring)
+ Conditional Random Fields (work in progress))
文件列表:
.directory (122, 2015-03-11)
@Color (0, 2015-03-11)
@Color\Color.m (2204, 2015-03-11)
@ContractiveAutoencoder (0, 2015-03-11)
@ContractiveAutoencoder\ContractiveAutoencoder.m (21265, 2015-03-11)
@ContractiveAutoencoder\private (0, 2015-03-11)
@ContractiveAutoencoder\private\computeNumericalGradient.m (658, 2015-03-11)
@ContractiveAutoencoder\private\contractiveAutoencoderCostBGD.m (5086, 2015-03-11)
@ContractiveAutoencoder\private\contractiveAutoencoderCostSGD.m (3365, 2015-03-11)
@ContractiveAutoencoder\private\dNonLinearity.m (453, 2015-03-11)
@ContractiveAutoencoder\private\nonLinearity.m (451, 2015-03-11)
@ConvUtils (0, 2015-03-11)
@ConvUtils\.svn (0, 2015-03-11)
@ConvUtils\.svn\all-wcprops (175, 2015-03-11)
@ConvUtils\.svn\entries (327, 2015-03-11)
@ConvUtils\.svn\text-base (0, 2015-03-11)
@ConvUtils\.svn\text-base\ConvUtils.m.svn-base (1886, 2015-03-11)
@ConvUtils\ConvUtils.m (1886, 2015-03-11)
@ConvUtils\private (0, 2015-03-11)
@ConvUtils\private\maxoutFprop.c (3940, 2015-03-11)
@ConvUtils\private\maxoutFprop.mexa64 (12849, 2015-03-11)
@DenoisingAutoencoder (0, 2015-03-11)
@DenoisingAutoencoder\DenoisingAutoencoder.m (30556, 2015-03-11)
@DenoisingAutoencoder\private (0, 2015-03-11)
@DenoisingAutoencoder\private\computeNumericalGradient.m (658, 2015-03-11)
@DenoisingAutoencoder\private\dNonLinearity.m (453, 2015-03-11)
@DenoisingAutoencoder\private\denoisingAutoencoderCostBGD.m (4082, 2015-03-11)
@DenoisingAutoencoder\private\denoisingAutoencoderCostSGD.m (3876, 2015-03-11)
@DenoisingAutoencoder\private\nonLinearity.m (451, 2015-03-11)
@Edge (0, 2015-03-11)
@Edge\Edge.m (2633, 2015-03-11)
@EmbeddingNeuralNet (0, 2015-03-11)
@EmbeddingNeuralNet\EmbeddingNeuralNet.m (30791, 2015-03-11)
@EmbeddingNeuralNet\private (0, 2015-03-11)
@EmbeddingNeuralNet\private\dNonLinearity.m (590, 2015-03-11)
@EmbeddingNeuralNet\private\embeddingNeuralNetCost.m (4230, 2015-03-11)
@EmbeddingNeuralNet\private\feedForwardENN.m (1709, 2015-03-11)
... ...
################## libORF - library On Random Fields ####################
version 1.0 beta
In this repository you can find implementations of various machine learning algorithm, using Object Oriented Matlab.
The library is written for self-educational purposes, without considering speed and scalability.
Since the implementations are in Matlab, speed and scalability is an issue, where I tried to manage some of them by making use of GPU or writing mex functions. Therefore, you may experience some scalability problems.
Following algorithms and models are provided along with some static utility classes:
- Naive Bayes, Linear Regression, Logistic Regression, Softmax Regression, Linear Support Vector Machine, Non-Linear Support Vector Machine (with RBF kernel),
Feed-forward Neural Network, Embedding Neural Network, Convolutional Neural Network, Sparse Autoencoders, Denoising Autoencoders,
Contractive Autoencoders, Stacked Sparse Autoencoders, Self-Taught Learner and Restricted Boltzmann Machines are tested with this version.
- Rest of the methods are not tested hence not supplied and the progress is as follows:
+ Deep Belief Nets with Restricted Boltzmann Machines (not tested)
+ Bayes Nets (tested - refactoring)
+ Hidden Markov Models (tested - refactoring)
+ Conditional Random Fields (work in progress)
PREREQUISITES
- What you may need is an optimizer function if you want to use some external optimizer such as minfunc (by Mark Schmidt), which is included along with the toolbox http://www.di.ens.fr/~mschmidt/Software/minFunc.html.
- For running examples you need sample datasets (currently maintained also in github) which can be downloaded from the link below ~80mb. If the link is broken feel free to contact me. After downloading data archive, extract all the files into libORF/data/ folder.https://www.dropbox.com/s/kagevr66n69yhhm/data.rar
USAGE:
libORF is written using object oriented matlab, just instantiate and call appropriate methods.
A full documentation is not supplied with this beta version, but the code is well commented, for me at least :)
Please refer to example scripts for sample usage in the main directory by simply calling run_.m
If you use libORF in your work, please cite the web-page: www.ceng.metu.edu.tr/~e1697481/libORF.html, or buy me a beer.
25 May 2014
Orhan Firat
近期下载者：
相关文件：
收藏者：