手写数字识别：Python+BP神经网络

此项目及使用方法已上传至Github，请移步至https://github.com/choutsugi/handwritten-number-recognizer查看。

1.简述

编程语言：Python3.6
界面设计：PyQt5
识别方法：BP神经网络
训练数据：Mnist训练集
手写数字获取：画图板

2.环境配置

此项目使用python3.6.3版本，请做参考，更高或更低版本可能不适用本项目（无法安装依赖或程序异常退出）。
项目依赖：requirement.txt

numpy==1.12.0
Pillow==4.2.0
PyQt5==5.11.3
PyQt5-sip==4.19.19
PyQt5-stubs==5.11.3.0
sip==4.19.8

3.文件目录

项目文件列表：

文件/文件夹	用途
Mnist文件夹	手写数字训练集
Resource文件夹	界面设计相关图像
tmp文件夹	图像处理临时文件
bp_train.py	神经网络训练
Main.py	主程序
MainWidget.py	交互界面
NN.py	神经网络定义
PaintBoard.py	画板
recognize.py	识别算法
weights.npy	权重

4.主程序

main.py

from MainWidget import MainWidget
from PyQt5.QtWidgets import QApplication
from PyQt5.Qt import *
import sys

def main():
    app = QApplication(sys.argv)
    app.setWindowIcon(QIcon('./Resource/icon/xmf.ico'))
    mainWidget = MainWidget()  # 新建一个主界面
    mainWidget.show()  # 显示主界面

    exit(app.exec_())  # 进入消息循环

if __name__ == '__main__':
    main()

5.神经网络定义

NN.py

import numpy as np

def LabelBinarizer(label):
    relabel=np.zeros([len(label),10],dtype=np.int32)
    for i in range(len(label)):
        relabel[i][label[i]]=1
    return relabel

def tanh(x):
    return np.tanh(x)

def tanh_deriv(x):
    return 1.0 - np.tanh(x)*np.tanh(x)

def logistic(x):
    return 1/(1 + np.exp(-x))

def logistic_derivative(x):
    return logistic(x)*(1-logistic(x))

class NeuralNetwork:
    def __init__(self, layers, activation='tanh'):
 
        if activation == 'logistic':
            self.activation = logistic
            self.activation_deriv = logistic_derivative
        elif activation == 'tanh':
            self.activation = tanh
            self.activation_deriv = tanh_deriv

        self.weights = []
        for i in range(1, len(layers) - 1):
            self.weights.append((2*np.random.random((layers[i - 1] + 1, layers[i] + 1))-1)*0.25)
            self.weights.append((2*np.random.random((layers[i] + 1, layers[i + 1]))-1)*0.25)

    def fit(self, X, y, learning_rate=0.15, epochs=60000):
        X = np.atleast_2d(X)
        temp = np.ones([X.shape[0], X.shape[1]+1])
        temp[:, 0:-1] = X  
        X = temp
        y = np.array(y)

        for k in range(epochs):
            print(k+1)
            i = np.random.randint(X.shape[0])
            a = [X[i]]

            for l in range(len(self.weights)):  
                a.append(self.activation(np.dot(a[l], self.weights[l])))  
            error = y[i] - a[-1]  
            deltas = [error * self.activation_deriv(a[-1])] 
            
            for l in range(len(a) - 2, 0, -1): 
                deltas.append(deltas[-1].dot(self.weights[l].T)*self.activation_deriv(a[l]))
            deltas.reverse()
            for i in range(len(self.weights)):
                layer = np.atleast_2d(a[i])
                delta = np.atleast_2d(deltas[i])
                self.weights[i] += learning_rate * layer.T.dot(delta)
        np.save("weights",self.weights)

    def predict(self, x):
        self.weights=np.load("weights.npy")
        x = np.array(x)
        temp = np.ones(x.shape[0]+