手写数字识别

mnist手写数字识别

一、BP神经网络

from tensorflow.keras.datasets import mnist
import numpy as np
from PIL import Image
from tensorflow.keras.utils import to_categorical
from tensorflow.keras import models,Sequential
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras.layers import Dense

data_path = r'mnist.npz'

(train_images,train_labels),(test_images,test_labels) = mnist.load_data(data_path)

train_images = train_images.reshape((60000,28*28)).astype('float')
test_images = test_images.reshape((10000,28*28)).astype('float')

train_images, test_images = train_images / 255.0,test_images / 255.0  # 归一化

train_labels = to_categorical(train_labels)
test_labels = to_categorical(test_labels)

model = Sequential([
    Dense(units=784,activation='relu',input_shape=(28*28,)),
    Dense(units=64,activation='relu'),
    Dense(units=10,activation='softmax')
])

model.compile(optimizer=RMSprop(lr=0.001),loss='categorical_crossentropy',metrics=['accuracy'])

model.fit(train_images,train_labels,epochs=5)

model.summary()

model.save('model.h5')

二、CNN卷积神经网络

from tensorflow.keras.datasets import mnist
import matplotlib.pyplot as pit
from tensorflow.keras.utils import to_categorical
from tensorflow.keras import models,layers,regularizers
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras import backend as K

# 加载本地数据集
data_path = r'mnist.npz'

# 处理数据
(train_images,train_labels),(test_images,test_labels) = mnist.load_data(data_path)


train_images = train_images.reshape((60000,28,28,1)).astype('float')
test_images = test_images.reshape((10000,28,28,1)).astype('float')

train_labels = to_categorical(train_labels)
test_labels = to_categorical(test_labels)

pit.imshow(train_images[0])
pit.show()

# 初始化模型
model = models.Sequential()
# 卷积层
model.add(layers.Conv2D(32, (3, 3),activation="relu",input_shape=(28, 28, 1)))
# 池化层
model.add(layers.MaxPooling2D((2, 2)))
# 卷积层
model.add(layers.Conv2D(64, (3, 3),activation="relu"))
# 池化层
model.add(layers.MaxPooling2D((2, 2)))
# 卷积层
model.add(layers.Conv2D(128, (3, 3),activation="relu"))
# 全连接层
model.add(layers.Flatten())
model.add(layers.Dense(64, activation='relu'))
model.add(layers.Dense(10, activation='softmax'))
#查看模型结构
model.summary()

model.compile(optimizer='adam',
              loss='categorical_crossentropy',
              metrics=['accuracy'])
history = model.fit(train_images,train_labels, epochs=5,
          batch_size=128,validation_data=(test_images,test_labels))

# 保存模型
model.save('minst_model.h5')

三、数据集

https://download.csdn.net/download/qq_43810466/89605106