读取sklearn中mnist_784离线数据集，并展示像素特征重要性热力图

“mnist_784”离线数据集：有监督的手写数字数据集

一、读取“mnist_784”离线数据集，并展示一个案例

import matplotlib.pyplot as plt
from scipy.io import loadmat

if __name__ == '__main__':
    mnist = loadmat('C:\\Users\\yunyun\\datasets\\mnist-original.mat')
    X = mnist['data'].T
    plt.imshow(X[10000].reshape(28, 28))
    plt.show()

展示结果如下图：

二、对“mnist_784”数据集进行随机森林集成分析

import matplotlib
import matplotlib.pyplot as plt
from scipy.io import loadmat
from sklearn.ensemble import RandomForestClassifier

if __name__ == '__main__':
    mnist = loadmat('C:\\Users\\yunyun\\datasets\\mnist-original.mat')
    rf_clf = RandomForestClassifier(n_estimators=500, n_jobs=-1)
    rf_clf.fit(mnist['data'].T, mnist['label'].reshape(70000, ))

    # 绘制热力图
    image = rf_clf.feature_importances_.reshape(28, 2