登录
首页 » matlab » 最先提出深度学习算法hinton的自动编码器AutoEncoder

最先提出深度学习算法hinton的自动编码器AutoEncoder

于 2020-10-03 发布 文件大小:22680KB
0 2372
下载积分: 1 下载次数: 180

代码说明:

  最先提出深度学习算法hinton的自动编码器matlab代码,内容是:利用多层rbm进行自动编码的多层特征训练,然后使用梯度算法进行fine turn。可以进行特征提取,也可以进行分类。压缩包里已带有训练用签字图片数据。相应算法说明可以查看hinton于2006年发表在 science的文章(First proposed deep learning algorithm hinton automatic encoder matlab code that reads: multilayer rbm automatic encoding multi-features of the training, and then use the gradient algorithm fine turn. Feature extraction can be classified. With compression bag has been training with the signature image data. Note You can view the corresponding algorithm was published in 2006 in hinton science articles)

文件列表:

code
....\backprop.m,5594,2006-05-21
....\backpropclassify.m,5474,2006-06-20
....\backup ZIP
....\..........\Autoencoder_Code.tar,51200,2016-04-15
....\..........\minimize.m,8995,2016-04-15
....\..........\t10k-images-idx3-ubyte.gz,1648877,2016-04-15
....\..........\t10k-labels-idx1-ubyte.gz,4542,2016-04-15
....\..........\train-images-idx3-ubyte.gz,9912422,2016-04-15
....\..........\train-labels-idx1-ubyte.gz,28881,2016-04-15
....\CG_CLASSIFY.m,1853,2006-06-20
....\CG_CLASSIFY_INIT.m,1136,2006-06-20
....\CG_MNIST.m,2727,2006-06-20
....\converter.m,3011,2006-06-20
....\makebatches.m,4169,2006-06-20
....\minimize.m,8995,2016-04-15
....\mnistclassify.m,1902,2006-06-20
....\mnistdeepauto.m,2199,2006-06-20
....\mnistdisp.m,1084,2006-06-20
....\rbm.m,3914,2006-06-20
....\rbmhidlinear.m,3964,2006-06-20
....\README.txt,2934,2006-07-13
....\t10k-images.idx3-ubyte,7840016,1998-01-26
....\t10k-labels.idx1-ubyte,10008,1998-01-26
....\train-images.idx3-ubyte,47040016,1996-11-18
....\train-labels.idx1-ubyte,60008,1996-11-18

下载说明:请别用迅雷下载,失败请重下,重下不扣分!

发表评论

0 个回复

  • Wigner-Ville
    自己写的一个计算空间时频分布的程序,魏格纳威尔分布原代码(Wrote a frequency distribution calculating space program, Wigner distribution of the original code Weir)
    2021-04-23 22:08:47下载
    积分:1
  • AIRtoolsManual_3
    this file contains the user manual for Algebraic Reconstruction Technique (ART) matlab toolbox
    2011-12-24 22:25:09下载
    积分:1
  • near-field-inverse-problems
    本书详细介绍了基于近场测试数据的逆散射问题的相关知识及其具体应用,这是近场问题的一个很好的例子,在人们都关注高频远场问题的背景下,提出近场的问题,对于计算电磁学的发展很有启发意义,我们的工程实际中,除了高频远场的应用外,低频近场也占有很大的份额,基于进场的反演是一个新兴的课题,需要我们共同去研究!(Book introduced in detail based on near field test data of the inverse scattering problem related knowledge and its practical application, this is a very good example of near field problems, people are concerned about the background of far field problem of a high frequency in the, near field is proposed, for the development of electromagnetism calculation is instructive. In the actual engineering, in addition to the high-frequency far field of application, low frequency near-field also occupies a large share of, based on the inversion approach is a new and developing subject, we go to study!)
    2015-06-28 20:27:02下载
    积分:1
  • penniu
    可以实现模式识别领域的数据的分类及回归,wolf 方法计算李雅普诺夫指数,图像的光流法计算的matlab程序。( You can achieve data classification and regression pattern recognition, wolf calculated Lyapunov exponent, Image optical flow calculation matlab program.)
    2017-01-16 13:58:16下载
    积分:1
  • K-meanCluster
    How the K-mean Cluster work Step 1. Begin with a decision the value of k = number of clusters Step 2. Put any initial partition that classifies the data into k clusters. You may assign the training samples randomly, or systematically as the following: Take the first k training sample as single-element clusters Assign each of the remaining (N-k) training sample to the cluster with the nearest centroid. After each assignment, recomputed the centroid of the gaining cluster. Step 3 . Take each sample in sequence and compute its distance from the centroid of each of the clusters. If a sample is not currently in the cluster with the closest centroid, switch this sample to that cluster and update the centroid of the cluster gaining the new sample and the cluster losing the sample. Step 4 . Repeat step 3 until convergence is achieved, that is until a pass through the training sample causes no new assignments. (How the K-mean Cluster workStep 1. Begin with a decision the value of k = number of clusters Step 2. Put any initial partition that classifies the data into k clusters. You may assign the training samples randomly, or systematically as the following: Take the first k training sample as single-element clusters Assign each of the remaining (Nk) training sample to the cluster with the nearest centroid. After each assignment, recomputed the centroid of the gaining cluster. Step 3. Take each sample in sequence and compute its distance from the centroid of each of the clusters. If a sample is not currently in the cluster with the closest centroid, switch this sample to that cluster and update the centroid of the cluster gaining the new sample and the cluster losing the sample. Step 4. Repeat step 3 until convergence is achieved, that is until a pass through the training sample causes no new assignments.)
    2007-11-15 01:49:03下载
    积分:1
  • maiya
    基于MATLAB的信号处理,经过脉冲压缩时加窗与不加窗对比以及分析。(MATLAB-based signal processing, after pulse compression and non-add window add contrast as well as the analysis window.)
    2009-03-12 22:21:30下载
    积分:1
  • tuxingchuliyuanma
    基于MATLAB的图像处理的各种操作源码(MATLAB image processing based on various operating source )
    2009-03-23 16:19:55下载
    积分:1
  • Desktop
    ocr programing in matlab to help anyone need help
    2012-03-26 23:38:48下载
    积分:1
  • lvboquzaoxin
    用于消噪处理用于研究Mallet算法及滤波器设计分解的高频系数采用db9较好,即它的消失矩较大(for Signal Processing Algorithms for the study Mallet decomposition and filter design using high-frequency coefficients db9 better, that is, its larger vanishing moments)
    2007-05-24 18:33:21下载
    积分:1
  • divide
    对区间进行分段的一个功能函数接口需自己调整.(The interval of a function interface segments need to own adjustment. )
    2014-12-09 10:38:07下载
    积分:1
  • 696520资源总数
  • 104162会员总数
  • 42今日下载