Dimensionality Reduction

--Hands-on Machine Learning with Scikit-Learn and TensorFlow -Chapter 8

Introduction

降维 pros：有助于加快训练速度；有助于数据可视化。cons：可能会导致重要信息丢失。
Two main approaches to dimensionality:projection and manifold learning
Three popular dimensionality reduction techniques:PCA,Kernel PCA, and LLE

Two main approaches for Dimensionalty Reduction

Projection 在实际问题当中，训练数据通常是非均匀的分布在整个维度里面。有很多特征是连续的，但是有一些特征非常相似。结果这些训练数据在低纬度空间中挨得非常近。
Manifold Learning （流形学习）

PCA(Principal Component Analysis)

pca是迄今为止最流行的降维算法。首先定义一个超平面，然后将数据投影到上面去。
pca降维应当注意点地方： 2.1 preserving the variance

2.2 the axis minimizes the mean squared distance between the original dataset and projection onto the axis

3.　Principal Components:PCA identifies the axis that accounts for the largest amount of variance in the training set.

the unit vector that defines the i(th) axis is called i(th) principal component.

如何找到训练数据的主成分？Singular Value Decomposition(SVD)

PAC默认数据集是以愿数据为中心的。Sklearn 的pca 包已经将数据集中化处理了。而用其他方法构造pca时候，不要忘记首先集中化处理数据。（centering the data）

在降维的时候，一定要尽可能的保证更大的方差。

from sklearn.decomposition import PCApca=PCA(n_components=2)
X2D=pcd.fit_transform(X)pca.explained_variance_ratio_

选择合适的维数

转载于:https://www.cnblogs.com/makino/p/9626871.html

Dimensionality Reduction相关推荐

文献记录(part17)--VARCLUST: clustering variables using dimensionality reduction
学习笔记,仅供参考,有错必究关键词:子空间聚类,降维,主成分分析,贝叶斯信息准则,k-centroids VARCLUST: clustering variables using dimension ...
An Indexable Time Series Dimensionality Reduction Method for Maximum Deviation Reduction and Simi...
An Indexable Time Series Dimensionality Reduction Method for Maximum Deviation Reduction and Simi... ...
Dimensionality Reduction - Principle Component Analysis problem formulation
摘要: 本文是吴恩达 (Andrew Ng)老师<机器学习>课程,第十五章<降维>中第117课时<主成分分析问题规划>的视频原文字幕.为本人在视频学习过程中记录下来 ...
sklearn - Dimensionality reduction
参考: 1.http://scikit-learn.org/stable/ 2.http://scikit-learn.org/stable/modules/decomposition.html#de ...
[论文翻译]A Global Geometric Framework for Nonlinear Dimensionality Reduction
论文题目:A Global Geometric Framework for Nonlinear Dimensionality Reduction 论文来源:Science 290, 2319 (200 ...
Review of Feature Selection, Dimensionality Reduction and Classification for Chronic Disease Diagnos
Review of Feature Selection, Dimensionality Reduction and Classification for Chronic Disease Diagnos ...
【论文阅读】Dimensionality Reduction by Learning an Invariant Mapping
1.背景对比学习算是比较早就已经提出了一种技术.其中,早期比较有名的一篇文章就是Lecun等在<Dimensionality Reduction by Learning an Invarian ...
【论文翻译】：Nonlinear Dimensionality Reduction by Locally Linear Embedding
[论文题目]:Nonlinear Dimensionality Reduction by Locally Linear Embedding [论文来源]:Nonlinear Dimensionalit ...
【论文翻译】Nonlinear Dimensionality Reduction by Locally Linear Embedding
论文题目:Nonlinear Dimensionality Reduction by Locally Linear Embedding 论文来源:http://science.sciencemag.o ...

Dimensionality Reduction

Dimensionality Reduction相关推荐

最新文章

热门文章