如何在 PCA 之后使用 2 个矩阵重建 Python 中 10 个分量的数据？

Question

Help please,. I've seen some anawers here, but they didn't help me. I need to reconstruct the initial data. having 2 matrixes and using first ten principal components. First matrix (Z) (X_reduced_417)- result of applying sklearn.decomposition.PCA. Second matrix (X_loadings_417) (F) is weight matrix? Answer is Initial data = Z*F+mean_matrix. How to use sklearn to find Z?

import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import sklearn.datasets, sklearn.decomposition

df_loadings = pd.read_csv('X_loadings_417.csv', header=None)
df_reduced = pd.read_csv('X_reduced_417.csv', header=None) ```

Answer 1

import pandas as pd
import numpy as np

# Load the df_loadings and df_reduced matrices from the CSV files
df_loadings = pd.read_csv("X_loadings_417.csv", header=None)
df_reduced = pd.read_csv("X_reduced_417.csv", header=None)

# Convert the DataFrames to numpy arrays
F = df_loadings.values
Z = df_reduced.values

# The mean of the original data is needed to reconstruct the data
mean_matrix = np.mean(X, axis=0)

# Reconstruct the original data using the first ten principal components
X_reconstructed = Z[:,:10].dot(F[:10,:]) + mean_matrix