20210329 LeNet-5数字识别 TensorFlow2.0 MINIST数据集

LeNet-5 数字识别 TensorFlow2.0

LeNet-5 由两个卷积层、两个池化层和两个全连接层组成，每个卷积层使用尺寸为５×５（每个滤波器有１个通道）的滤波器，第一层中有６个滤波 器，第 二 层 中 有 １６ 个 滤 波 器。在 每 次 卷 积 之后，采用 Ｓｉｇｍｏｉｄ函数进行激活，并且使用２×２的平均池化进行池化操作。

# Implementation of LeNet-5 in keras 
# [LeCun et al., 1998. Gradient based learning applied to document recognition]
# Some minor changes are made to the architecture like using ReLU activation instead of 
# sigmoid/tanh, max pooling instead of avg pooling and softmax output layer 
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from tensorflow import keras
# 导入数据
(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()
# 观察数据
print (x_train.shape)
plt.imshow(x_train[1000])
print (y_train[1000])
#数据处理，维度要一致
x_train = x_train.reshape((x_train.shape[0],28,28,1)).astype('float32') 
x_test = x_test.reshape((x_test.shape[0],28,28,1)).astype('float32') #-1代表那个地方由其余几个值算来的
x_train = x_train/255
x_test = x_test/255
x_train = np.pad(x_train, ((0,0),(2,2),(2,2),(0,0)), 'constant')
x_test = np.pad(x_test, ((0,0),(2,2),(2,2),(0,0)), 'constant')
print (x_train.shape)

#序贯模型（Sequential):单输入单输出
model = keras.Sequential()

from tensorflow.keras.layers import Conv2D
from tensorflow.keras.layers import MaxPooling2D
from tensorflow.keras.layers import Flatten
from tensorflow.keras.layers import Dense

#Layer 1
#Conv Layer 1
model.add(Conv2D(filters = 6, 
                 kernel_size = 5, 
                 strides = 1, 
                 activation = 'relu', 
                 input_shape = (32,32,1)))
#Pooling layer 1
model.add(MaxPooling2D(pool_size = 2, strides = 2))

#Layer 2
#Conv Layer 2
model.add(Conv2D(filters = 16, 
                 kernel_size = 5,
                 strides = 1,
                 activation = 'relu',
                 input_shape = (14,14,6)))
#Pooling Layer 2
model.add(MaxPooling2D(pool_size = 2, strides = 2))
#Flatten
model.add(Flatten())

#Layer 3
#Fully connected layer 1
model.add(Dense(units = 120, activation = 'relu'))

#Layer 4
#Fully connected layer 2
model.add(Dense(units = 84, activation = 'relu'))

#Layer 5
#Output Layer
model.add(Dense(units = 10, activation = 'softmax'))
model.compile(optimizer = 'adam',  loss=keras.losses.SparseCategoricalCrossentropy(), metrics = ['accuracy'])

model.summary()

#进行模型训练
historyLeNet = model.fit(x_train, y_train, steps_per_epoch=10, epochs=42)
#保存下模型（这个方法比较常用，也可以考虑适合部署的SavedModel 方式）
model.save('LeNet_5.h5')#保存模型，名字可以任取，但要由.h5后缀
#测试模型
model.evaluate(x_test, y_test)

#相关库的导入
from tensorflow import keras
import tensorflow.keras.layers as layers
import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np 
import cv2
#加载模型
model=tf.keras.models.load_model('LeNet_5.h5')
#自身数据加载（随便去百度或者那截张图，然后读取处理后就可以）

def output(y_pre , y):
    temp = np.argmax(y_pre)
    print ('预测结果为'+str(temp))
    print ('实际结果为'+str(y))
    if str(temp)==str(y):
        print('预测结果正确')
    else:
        print('预测结果错误')

def readnum(path):#读取自己的数字数据进行预测
    img = cv2.imread(path, 0)#灰度图读入
   # plt.imshow(img)
    #查看数据
    img = cv2.resize(img,(28,28))
#    cv2.waitKey(0)
#    cv2.destroyAllWindows()
    img = np.array(img)
    img = img.reshape((-1,28,28,1)).astype('float32')
    img = 1-img/255.0 #因为我的数字是相反颜色的所以有个反转
    return img
#预测并输出结果
path = 'D:\\0.jpg'
xTemp=readnum(path)
x_pre = np.pad(xTemp, ((0,0),(2,2),(2,2),(0,0)), 'constant')
y_pre = model.predict(x_pre)
output(y_pre, 8)