sklearn PCA实践

sklearn PCA实践


PCA无需样本标签，属于无监督学习降维；LDA需要样本标签，属于有监督学习降维。
[url]https://blog.csdn.net/ainimao6666/article/details/64933677[/url]


[url]https://blog.csdn.net/Huangyi_906/article/details/76438885[/url]

class sklearn.decomposition.PCA(n_components=None, copy=True, whiten=False, svd_solver=’auto’, tol=0.0, iterated_power=’auto’, random_state=None)

其中参数：
n_components=None：指定降维后的维数。如果给定数在（0,1）之间，则为降维后占原维数的百分比。默认自动选择。


属性：
components_ :主成分组数
explained_variance_ratio_:每个主成分占方差比例
n_components_ :一个整数，指示主成分有多少个元素。


方法：
fit(x):训练模型
transform(x): 执行降维
fit_transform(x): 训练并降维
inverse_transform(x): 逆向操作，把降维的数据逆向转换回原来数据。


例子
[url]https://blog.csdn.net/puredreammer/article/details/52255025[/url]
[url]https://www.jianshu.com/p/8642d5ea5389[/url]

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets.samples_generator import make_classification
from sklearn.decomposition import PCA
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis

from mpl_toolkits.mplot3d import Axes3D


X = np.array([[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]])

pca = PCA(n_components=2)#n_components表示结果取几维

pca.fit(X) #训练模型
print(pca.explained_variance_ratio_) #每维对一致性的比率，越高表示这个维度更能表示降维后的效果[0.99244289 0.00755711]

X2 =  pca.transform(X) #执行降维
print('X2==')
print(X2)
# [[ 1.38340578  0.2935787 ]
#  [ 2.22189802 -0.25133484]
#  [ 3.6053038   0.04224385]
#  [-1.38340578 -0.2935787 ]
#  [-2.22189802  0.25133484]
#  [-3.6053038  -0.04224385]]

X4 =  pca.fit_transform(X)#训练并降维
print('X4==')
print(X4)
# [[ 1.38340578  0.2935787 ]
#  [ 2.22189802 -0.25133484]
#  [ 3.6053038   0.04224385]
#  [-1.38340578 -0.2935787 ]
#  [-2.22189802  0.25133484]
#  [-3.6053038  -0.04224385]]


X3 =  pca.inverse_transform(X2)#逆向操作，把降维的数据逆向转换回原来数据。
print('X3==')
print(X3)
# [[-1. -1.]
#  [-2. -1.]
#  [-3. -2.]
#  [ 1.  1.]
#  [ 2.  1.]
#  [ 3.  2.]]

pca = PCA(n_components=1)#降为一维
X5 =  pca.fit_transform(X)
print('X5==')
print(X5)
# [[ 1.38340578]
#  [ 2.22189802]
#  [ 3.6053038 ]
#  [-1.38340578]
#  [-2.22189802]
#  [-3.6053038 ]]


X6 =  pca.inverse_transform(X5)#逆向操作，把降维的数据逆向转换回原来数据。
print('X6==')
print(X6)
# [[-1.15997501 -0.75383654]
#  [-1.86304424 -1.21074232]
#  [-3.02301925 -1.96457886]
#  [ 1.15997501  0.75383654]
#  [ 1.86304424  1.21074232]
#  [ 3.02301925  1.96457886]]

#==============================================
xSrc = X[:,0];
ySrc = X[:,1];
print (xSrc)
# [-1 -2 -3  1  2  3]
print (ySrc)
# [-1 -1 -2  1  1  2]
plt.scatter(xSrc, ySrc,marker='o',c='r',alpha=0.5) #原坐标点的图

xInverse = X6[:,0];
yInverse = X6[:,1];
plt.scatter(xInverse, yInverse,marker='o',c='b',alpha=0.5) #降为一维后再变为二维的图

# x2Inverse = X4[:,0];
# y2Inverse = X4[:,1];
# plt.scatter(x2Inverse, y2Inverse,marker='o',c='g',alpha=0.5) #降为二维后再变为二维的图

plt.show()

[img]http://dl2.iteye.com/upload/attachment/0130/7987/f886532d-6e86-35ec-a433-52e2cb279ec9.png[/img]