https://github.com/heucoder/dimensionality_reduction_alo_codes

特征提取/数据降维:PCA、LDA、MDS、LLE、TSNE等降维算法的python实现
https://github.com/heucoder/dimensionality_reduction_alo_codes

data-reduction feature-extraction python

Last synced: about 2 months ago
JSON representation

特征提取/数据降维:PCA、LDA、MDS、LLE、TSNE等降维算法的python实现

• Host: GitHub
• URL: https://github.com/heucoder/dimensionality_reduction_alo_codes
• Owner: heucoder
• License: apache-2.0
• Created: 2019-06-09T13:43:11.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2023-05-10T05:58:24.000Z (over 1 year ago)
• Last Synced: 2024-07-22T03:36:52.285Z (2 months ago)
• Topics: data-reduction, feature-extraction, python
• Language: Python
• Homepage:
• Size: 2.11 MB
• Stars: 2,253
• Watchers: 43
• Forks: 618
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-machine-learning-resources - **[Tutorial

README

        
# DimensionalityReduction_alo_codes

**网上关于各种降维算法的资料参差不齐，同时大部分不提供源代码；在此通过借鉴资料实现了一些经典降维算法的Demo(python)，同时也给出了参考资料的链接。**

降维算法|资料链接|代码|展示|

---|---|---|---

PCA | [资料链接１](https://blog.csdn.net/u013719780/article/details/78352262) [资料链接２](https://blog.csdn.net/u013719780/article/details/78352262) [资料链接３](https://blog.csdn.net/weixin_40604987/article/details/79632888) | [PCA](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/PCA) | ![PCA](codes/PCA/PCA.png)

KPCA | [资料链接1](https://blog.csdn.net/u013719780/article/details/78352262) [资料链接2](https://blog.csdn.net/u013719780/article/details/78352262) [资料链接3](https://blog.csdn.net/weixin_40604987/article/details/79632888) |[KPCA](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/PCA) |![KPCA](codes/PCA/KPCA.png)

LDA | [资料链接１](https://blog.csdn.net/ChenVast/article/details/79227945) [资料链接2](https://www.cnblogs.com/pinard/p/6244265.html) | [LDA](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/LDA) | ![LDA](codes/LDA/LDA.png)

MDS | [资料链接１](https://blog.csdn.net/zhangweiguo_717/article/details/69663452?locationNum=10&fps=1) | [MDS](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/MDS) | ![MDS](codes/MDS/MDS_1.png) ![Tensor-MDS](codes/MDS/MDS_2.png)

ISOMAP | [资料链接１](https://blog.csdn.net/zhangweiguo_717/article/details/69802312) [资料链接２](http://www-clmc.usc.edu/publications/T/tenenbaum-Science2000.pdf) | [ISOMAP](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/ISOMAP) | ![ISOMAP](codes/ISOMAP/Isomap.png)

LLE | [资料链接１](https://blog.csdn.net/scott198510/article/details/76099630) [资料链接2](https://www.cnblogs.com/pinard/p/6266408.html?utm_source=itdadao&utm_medium=referral) | [LLE](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/LLE) |![LLE](codes/LLE/LLE.png)

TSNE | [资料链接１](http://bindog.github.io/blog/2018/07/31/t-sne-tips/) | [TSNE](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/T-SNE) |![TSNE](codes/T-SNE/T-SNE.png)

AutoEncoder |无　| |![AutoEncoder](codes/AutoEncoder/AutoEncoder.png)

FastICA | [资料链接１](https://blog.csdn.net/lizhe_dashuju/article/details/50263339) |[FastICA](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/ICA) |

SVD | [资料链接１](https://blog.csdn.net/m0_37870649/article/details/80547167) [资料链接2](https://www.cnblogs.com/pinard/p/6251584.html) | [SVD](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/SVD) |

LE | [资料链接1](https://blog.csdn.net/hustlx/article/details/50850342)[资料链接2](https://blog.csdn.net/jwh_bupt/article/details/8945083) | [LE](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/LE) | ![LE](codes/LE/LE_1.png)

LPP | [资料链接１](https://blog.csdn.net/qq_39187538/article/details/90402961) [资料链接２](https://blog.csdn.net/xiaohen123456/article/details/82288222) | [LPP](https://github.com/heucoder/dimensionality_reduction_alo_codes/tree/master/codes/LPP) | ![LPP](codes/LPP/LPP.png)

`环境: python3.6 ubuntu18.04(windows10)`

`需要的库: numpy sklearn tensorflow matplotlib`

- 每一个代码都可以单独运行，但是只是作为一个demo，仅供学习使用

- 其中AutoEncoder只是使用AutoEncoder简单的实现了一个PCA降维算法,自编码器涉及到了深度学习领域，其本身就是一个非常大领域

- LE算法的鲁棒性极差，对近邻的选择和数据分布十分敏感

- **2019.6.20添加了LPP算法，但是效果没有论文上那么好，有点迷，后续需要修改**