Python实现PCA简化数据

最新推荐文章于 2023-02-20 18:49:55 发布

杨鑫newlfe

最新推荐文章于 2023-02-20 18:49:55 发布

阅读量331

点赞数

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/u012965373/article/details/90723210

Python 同时被 2 个专栏收录

714 篇文章 ¥9.90 ¥99.00

订阅专栏

Machine Learning

64 篇文章 ¥9.90 ¥99.00

订阅专栏

本文深入探讨了Python中PCA（主成分分析）的实现，详细解释了PCA的基本原理，通过实例展示了如何使用Python的sklearn库进行数据降维，帮助读者理解PCA在机器学习中的应用及其简化复杂数据集的能力。

# -*- coding:utf-8 -*-
__author__ = 'yangxin_ryan'
from numpy import *
import matplotlib.pyplot as plt


class PCA(object):

    def load_data_set(self, file_name, delim='\t'):
        fr = open(file_name)
        string_arr = [line.strip().split(delim) for line in fr.readlines()]
        dat_arr = [list(map(float, line)) for line in string_arr]
        return mat(dat_arr)

    def pca(self, data_mat, top_n_feat=9999999):
        mean_vals = mean(data_mat, axis=0)
        mean_removed = data_mat - mean_vals
        cov_mat = cov(mean_removed, rowvar=0)
        eig_vals, eig_vects = linalg.eig(mat(cov_mat))
        eig_val_ind = argsort(eig_vals)
        eig_val_ind = eig_val_ind[:-(top_n_feat + 1): -1]
        red_eig_vects = eig_vects[:, eig_

了解本专栏