Created
February 5, 2020 15:45
-
-
Save theWrongCode-dev/a0db6dc77674e9e31904fef7fdd6cd1f to your computer and use it in GitHub Desktop.
PCA Implemented using Python 3
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
class PCA: | |
def __init__(self,n_component=None): | |
"""Principal component analysis (PCA) implementation. | |
Transforms a dataset of possibly correlated values into n linearly | |
uncorrelated components. The components are ordered such that the first | |
has the largest possible variance and each following component as the | |
largest possible variance given the previous components. This causes | |
the early components to contain most of the variability in the dataset. | |
Parameters | |
---------- | |
n_components : int | |
""" | |
self.n_component =n_component | |
self.components_ =None | |
self.explained_variance_ = None | |
self.mean = None | |
def fit(self,data): | |
global values,vectors | |
#mean | |
meandata= np.mean(data.T,axis=1) | |
self.mean = meandata.round(2) | |
C = data - self.mean | |
# calculate covariance matrix of centered matrix | |
V = np.cov(C.T) | |
# eigendecomposition of covariance matrix | |
values, vectors = np.linalg.eig(V) | |
key = np.argsort(values)[::-1][:self.n_component] | |
values, vectors = values[key], vectors[:, key] | |
self.components_ = vectors[0:self.n_component] | |
self.explained_variance_ = values | |
@property | |
def variance_ratio(self): | |
if len(vectors)!=0: | |
s_squared = values** 2 | |
variance_ratio = s_squared / (s_squared).sum() | |
return variance_ratio[0:self.n_component] | |
def transform(self, data): | |
meandata= np.mean(data.T,axis=1) | |
self.mean = meandata.round(2) | |
C = data - self.mean | |
return np.dot(C, self.components_.T) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
pca implemented only using python