``` #!/usr/bin/python3 # coding:utf-8 # @Author: Lin Misaka # @File: net.py # @Data: 2020/11/30 # @IDE: PyCharm from os import listdir import numpy as np import matplotlib.pyplot as plt import pickle # 函數img2vector將圖像轉換為向量 def img2vector(filename): returnVect = np.zeros((1, 1024)) fr = open(filename) for i in range(32): lineStr = fr.readline() for j in range(32): returnVect[0, 32 * i + j] = int(lineStr[j]) return returnVect # 讀取手寫字體txt數據 def handwritingData(dataPath): hwLabels = [] FileList = listdir(dataPath) # 1 獲取目錄內容 m = len(FileList) digitalmat = np.zeros((m, 1024)) for i in range(m): # 2 從文件名解析分類數字 fileNameStr = FileList[i] fileStr = fileNameStr.split('.')[0] # take off .txt classNumStr = int(fileStr.split('_')[0]) hwLabels.append(classNumStr) #print(digitalmat[i, :].shape)#(1024,) #print(img2vector(dataPath + '/%s' % fileNameStr).shape) # (1, 1024) #print(digitalmat[i].shape) # (1024,) digitalmat[i] = img2vector(dataPath + '/%s' % fileNameStr) return digitalmat, hwLabels # diff = True求導 def Sigmoid(x, diff=False): def sigmoid(x): # sigmoid函數 return 1 / (1 + np.exp(-x)) def dsigmoid(x): f = sigmoid(x) return f * (1 - f) if (diff == True): return dsigmoid(x) return sigmoid(x) # diff = True求導 def SquareErrorSum(y_hat, y, diff=False): if (diff == True): return y_hat - y return (np.square(y_hat - y) * 0.5).sum() class Net(): def __init__(self): # X Input self.X = np.random.randn(1024, 1) self.W1 = np.random.randn(16, 1024) self.b1 = np.random.randn(16, 1) self.W2 = np.random.randn(16, 16) self.b2 = np.random.randn(16, 1) self.W3 = np.random.randn(10, 16) self.b3 = np.random.randn(10, 1) self.alpha = 0.01 #學習率 self.losslist = [] #用于作圖 def forward(self, X, y, activate): self.X = X self.z1 = np.dot(self.W1, self.X) + self.b1 self.a1 = activate(self.z1) self.z2 = np.dot(self.W2, self.a1) + self.b2 self.a2 = activate(self.z2) self.z3 = np.dot(self.W3, self.a2) + self.b3 self.y_hat = activate(self.z3) Loss = SquareErrorSum(self.y_hat, y) return Loss, self.y_hat def backward(self, y, activate): self.delta3 = activate(self.z3, True) * SquareErrorSum(self.y_hat, y, True) self.delta2 = activate(self.z2, True) * (np.dot(self.W3.T, self.delta3)) self.delta1 = activate(self.z1, True) * (np.dot(self.W2.T, self.delta2)) dW3 = np.dot(self.delta3, self.a2.T) dW2 = np.dot(self.delta2, self.a1.T) dW1 = np.dot(self.delta1, self.X.T) d3 = self.delta3 d2 = self.delta2 d1 = self.delta1 #update weight self.W3 -= self.alpha * dW3 self.W2 -= self.alpha * dW2 self.W1 -= self.alpha * dW1 self.b3 -= self.alpha * d3 self.b2 -= self.alpha * d2 self.b1 -= self.alpha * d1 def setLearnrate(self, l): self.alpha = l def save(self,path): obj = pickle.dumps(self) with open(path,"wb") as f: f.write(obj) def load(path): obj = None with open(path, "rb") as f: try: obj = pickle.load(f) except: print("IOError") return obj def train(self, trainMat, trainLabels, Epoch=5, bitch=None): for epoch in range(Epoch): acc = 0.0 acc_cnt = 0 label = np.zeros((10, 1))#先生成一個10x1是向量，減少運算。用于生成one_hot格式的label for i in range(len(trainMat)):#可以用batch，數據較少，一次訓練所有數據集 X = trainMat[i, :].reshape((1024, 1)) #生成輸入 labelidx = trainLabels[i] label[labelidx][0] = 1.0 Loss, y_hat = self.forward(X, label, Sigmoid)#前向傳播 self.backward(label, Sigmoid)#反向傳播 label[labelidx][0] = 0.0#還原為0向量 acc_cnt += int(trainLabels[i] == np.argmax(y_hat)) acc = acc_cnt / len(trainMat) self.losslist.append(Loss) print("epoch:%d,loss:%02f,accrucy : %02f%%" % (epoch, Loss, acc*100)) self.plotLosslist(self.losslist, "Loss:Init->randn,alpha=0.01") def plotLosslist(self, Loss, title): font = {'family': 'simsun', 'weight': 'bold', 'size': 20, } m = len(Loss) X = range(m) # plt.figure(1) plt.subplots(nrows=1, ncols=1, figsize=(10, 8)) plt.subplot(111) plt.title(title, font) plt.plot(X, Loss) plt.xlabel(r'Epoch', font) plt.ylabel(u'Loss', font) plt.show() def test(self, testMat, testLabels, bitch=None): acc = 0.0 acc_cnt = 0 label = np.zeros((10, 1))#先生成一個10x1是向量，減少運算。用于生成one_hot格式的label if(bitch == None): bitch = len(testMat) for i in range(bitch):#可以用batch，數據較少，一次訓練所有數據集 X = testMat[i, :].reshape((1024, 1)) #生成輸入 labelidx = testLabels[i] label[labelidx][0] = 1.0 Loss, y_hat = self.forward(X, label, Sigmoid)#前向傳播 label[labelidx][0] = 0.0#還原為0向量 acc_cnt += int(testLabels[i] == np.argmax(y_hat)) acc = acc_cnt / bitch print("test num: %d, accrucy : %05.3f%%"%(bitch,acc*100)) # 讀取訓練數據 trainDataPath = "./trainingDigits" trainMat, trainLabels = handwritingData(trainDataPath) testDataPath = "./testDigits" testMat, testLabels = handwritingData(testDataPath) net = Net() net.setLearnrate(0.01) net.train(trainMat, trainLabels, Epoch=20) #net.train(trainMat, trainLabels, Epoch=200) net.save("hr.model") net.test(testMat, testLabels) newmodel = Net.load("hr.model") newmodel.test(testMat, testLabels) ``` . >>> ===================== RESTART: E:\012digits210201\net.py ===================== epoch:0,loss:0.884695,accrucy : 29.452055% epoch:1,loss:0.803985,accrucy : 40.924658% epoch:2,loss:0.801575,accrucy : 48.630137% epoch:3,loss:0.804054,accrucy : 55.308219% epoch:4,loss:0.801643,accrucy : 61.815068% epoch:5,loss:0.794812,accrucy : 66.780822% epoch:6,loss:0.785229,accrucy : 70.376712% epoch:7,loss:0.773085,accrucy : 74.657534% epoch:8,loss:0.758082,accrucy : 78.253425% epoch:9,loss:0.740774,accrucy : 80.479452% epoch:10,loss:0.724316,accrucy : 83.390411% epoch:11,loss:0.710766,accrucy : 85.787671% epoch:12,loss:0.700182,accrucy : 87.328767% epoch:13,loss:0.691778,accrucy : 88.869863% epoch:14,loss:0.684741,accrucy : 89.897260% epoch:15,loss:0.678793,accrucy : 90.924658% epoch:16,loss:0.673885,accrucy : 91.438356% epoch:17,loss:0.670210,accrucy : 91.438356% epoch:18,loss:0.667625,accrucy : 91.780822% epoch:19,loss:0.665471,accrucy : 92.123288% test num: 276, accrucy : 94.565% test num: 276, accrucy : 94.565% >>>