实现pca降维-Python实现

PCA:主成分分析-Python实现，X:[2500,784],把X降到50维

def pca(X=np.array([]), no_dims=50):
    """
        Runs PCA on the NxD array X in order to reduce its dimensionality to
        no_dims dimensions.
    """

    print("Preprocessing the data using PCA...")
    (n, d) = X.shape
    X = X - np.tile(np.mean(X, 0), (n, 1)) # np.mean(X,0)在列上求均值
    (l, M) = np.linalg.eig(np.dot(X.T, X))
    Y = np.dot(X, M[:, 0:no_dims])
    return Y

np.linalg.eig(np.dot(X.T, X)):
eig方法：计算方阵的特征值和右特征向量。
l:(784,):特征值
M:(784,784)：右特征向量
Y = X*M

补：

参考：

https://blog.csdn.net/baimafujinji/article/details/79407488

https://blog.csdn.net/qq_24464989/article/details/79834564?utm_medium=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase&depth_1-utm_source=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase