qq_41665310的博客

机器学习之用bp神经网络实现手写数字识别代码如下# 使用三层bp神经网络采用梯度下降法训练数字手写识别import scipy.io as sioimport numpy as npfrom sklearn.metrics import classification_report###################################### 计算给定theta下的输出 #def feedforward(theta1,theta2,X):

机器学习之用逻辑回归实现手写数字识别import numpy as npimport matplotlib as pltimport scipy.io as siofrom sklearn.metrics import classification_report, accuracy_scoreplt.switch_backend('agg')if __name__ == '__main__': data = sio.loadmat('ex3data1.mat') X = dat