本文是深度学习课程的实验报告
使用了MLP/LeNet/AlexNet/GoogLeNet/ResNet五个深度神经网络模型结构和MNIST、Fashion MNIST、HWDB1三个不同的数据集，所用的开发框架为tensorflow2。 本文的数据集和.ipynb文件可在此处下载
实验结果
实验结果如下表所示
模型在不同数据集上的准确度
MNIST | Fashion MNIST | HWDB1 | |
MLP | 97.76% | 87.73% | 84.17% |
LeNet | 98.68% | 85.82% | 91.33% |
AlexNet | 98.91% | 90.57% | 89.67% |
GoogLeNet | 99.27% | 90.27% | 91.50% |
ResNet | 99.21% | 91.35% | 93.67% |
导入相关库
import os
import warnings
import gzip
import numpy as np
import tensorflow as tf
import
环境设置
warnings.filterwarnings("ignore")
os.environ["TF_CPP_MIN_LOG_LEVEL"] = '2' # 只显示 warning 和 Error
os.environ['CUDA_VISIBLE_DEVICES'] = '-1' # 用CPU训练
logging.disable(30) # 不显示warning
加载mnist数据
def load_mnist():
with np.load(r'./datasets/mnist.npz', allow_pickle=True) as f:
x_train, y_train = f['x_train'], f['y_train']
x_test, y_test = f['x_test'], f['y_test']
return (x_train, y_train), (x_test, y_test)
加载fashion_mnist数据
def load_fashion_mnist():
dirname = os.path.join('datasets', 'fashion-mnist')
files = ['train-labels-idx1-ubyte.gz', 'train-images-idx3-ubyte.gz',
't10k-labels-idx1-ubyte.gz', 't10k-images-idx3-ubyte.gz']
paths = []
for fname in files:
paths.append(os.path.join(dirname, fname))
with gzip.open(paths[0], 'rb') as lbpath:
y_train = np.frombuffer(lbpath.read(), np.uint8, offset=8)
with gzip.open(paths[1], 'rb') as imgpath:
x_train = np.frombuffer(imgpath.read(), np.uint8,
offset=16).reshape(len(y_train), 28, 28)
with gzip.open(paths[2], 'rb') as lbpath:
y_test = np.frombuffer(lbpath.read(), np.uint8, offset=8)
with gzip.open(paths[3], 'rb') as imgpath:
x_test = np.frombuffer(imgpath.read(), np.uint8,
offset=16).reshape(len(y_test), 28, 28)
return (x_train, y_train), (x_test, y_test)
加载HWDB1数据
# 图像预处理/设置为24*24大小并归一化
def preprocess_image(image):
image_size = 24
img_tensor = tf.image.decode_jpeg(image, channels=1)
img_tensor = tf.image.resize(img_tensor, [image_size, image_size])
img_tensor /= 255.0 # normalize to [0,1] range
return img_tensor
def load_and_preprocess_image(path):
image = tf.io.read_file(path)
return preprocess_image(image)
def load_HWDB1():
root_path = './datasets/HWDB1'
train_images = []
train_labels = []
test_images = []
test_labels = []
temp = []
with open(os.path.join(root_path, 'train.txt'), 'r') as f:
for line in f:
line = line.strip('\n')
if line is not '':
imgpath = line
label = line.split('\\')[1]
train_images.append(imgpath)
train_labels.append(int(label))
for item in train_images:
img = load_and_preprocess_image(item)
temp.append(img)
x_train = np.array(temp)
y_train = np.array(train_labels)
temp = []
with open(os.path.join(root_path, 'test.txt'), 'r') as f:
for line in f:
line = line.strip('\n')
if line is not '':
imgpath = line
label = line.split('\\')[1]
test_images.append(imgpath)
test_labels.append(int(label))
for item in test_images:
img = load_and_preprocess_image(item)
temp.append(img)
x_test = np.array(temp)
y_test = np.array(test_labels)
return (x_train, y_train), (x_test, y_test)
定义mnist数据装载器
class MNISTLoader():
def __init__(self):
(self.train_data, self.train_label), (self.test_data,
self.test_label) = load_mnist()
# MNIST中的图像默认为uint8（0-255的数字）。以下代码将其归一化到0-1之间的浮点数，并在最后增加一维作为颜色通道
self.train_data = np.expand_dims(self.train_data.astype(
np.float32) / 255.0, axis=-1) # [60000, 28, 28, 1]
self.test_data = np.expand_dims(self.test_data.astype(
np.float32) / 255.0, axis=-1) # [10000, 28, 28, 1]
self.train_label = self.train_label.astype(np.int32) # [60000]
self.test_label = self.test_label.astype(np.int32) # [10000]
self.num_train_data, self.num_test_data = self.train_data.shape[
0], self.test_data.shape[0]
def get_batch(self, batch_size):
# 从数据集中随机取出batch_size个元素并返回
index = np.random.randint(0, self.num_train_data, batch_size)
return self.train_data[index, :], self.train_label[index]
定义fashion_mnist数据装载器
class Fashion_MNISTLoader():
def __init__(self):
(self.train_data, self.train_label), (self.test_data,
self.test_label) = load_fashion_mnist()
self.train_data = np.expand_dims(
self.train_data.astype(np.float32) / 255.0, axis=-1)
self.test_data = np.expand_dims(
self.test_data.astype(np.float32) / 255.0, axis=-1)
self.train_label = self.train_label.astype(np.int32)
self.test_label = self.test_label.astype(np.int32)
self.num_train_data, self.num_test_data = self.train_data.shape[
0], self.test_data.shape[0]
def get_batch(self, batch_size):
# 从数据集中随机取出batch_size个元素并返回
index = np.random.randint(0, self.num_train_data, batch_size)
return self.train_data[index, :], self.train_label[index]
定义HWDB1数据装载器
class HWDB1Loader():
def __init__(self):
(self.train_data, self.train_label), (self.test_data,
self.test_label) = load_HWDB1()
self.train_label = self.train_label.astype(np.int32)
self.test_label = self.test_label.astype(np.int32)
self.num_train_data, self.num_test_data = self.train_data.shape[
0], self.test_data.shape[0]
def get_batch(self, batch_size):
# 从数据集中随机取出batch_size个元素并返回
index = np.random.randint(0, self.num_train_data, batch_size)
return self.train_data[index, :], self.train_label[index]
定义多层感知机网络结构
两个全连接层
class MLP(tf.keras.Model):
def __init__(self):
super().__init__()
self.flatten = tf.keras.layers.Flatten() # Flatten层将除第一维（batch_size）以外的维度展平
self.dense1 = tf.keras.layers.Dense(units=100, activation=tf.nn.relu)
self.dense2 = tf.keras.layers.Dense(units=10)
def call(self, inputs): # [batch_size, 28, 28, 1]
x = self.flatten(inputs) # [batch_size, 784]
x = self.dense1(x) # [batch_size, 100]
x = self.dense2(x) # [batch_size, 10]
output = tf.nn.softmax(x)
return
定义LeNet网络结构
from tensorflow.keras.layers import Conv2D, BatchNormalization, Activation, MaxPool2D, Dropout, Flatten, Dense
from tensorflow.keras import Model
class LeNet(Model):
def __init__(self):
super(LeNet, self).__init__()
# 原LeNet输入是 32x32，这个Mnist数据输入是28x28
self.c1 = Conv2D(filters=6, kernel_size=(5, 5), activation='sigmoid')
self.p1 = MaxPool2D(pool_size=(2, 2), strides=2)
self.c2 = Conv2D(filters=16, kernel_size=(5, 5),activation='sigmoid')
self.p2 = MaxPool2D(pool_size=(2, 2), strides=2)
self.flatten = Flatten()
self.f1 = Dense(120, activation='sigmoid')
self.f2 = Dense(84, activation='sigmoid')
self.f3 = Dense(10, activation='softmax')
# 前向传播
def call(self, inputs): # [batch_size, 28, 28, 1]
x = self.c1(inputs) # [batch_size, 24, 24, 6] 卷积前后计算公式：输出大小=[(输入大小-卷积核大小+2*Padding）/步长]+1
x = self.p1(x) # [batch_size, 12, 12, 6]
x = self.c2(x) # [batch_size, 8, 8, 16]
x = self.p2(x) # [batch_size, 4, 4, 16]
x = self.flatten(x) # [batch_size, 256]
x = self.f1(x) # [batch_size, 120]
x = self.f2(x) # [batch_size, 84]
y = self.f3(x) # [batch_size, 10]
return
定义AlexNet网络结构
AlexNet创新点：
1.激活函数使用relu
2.卷积之后引入标准化层(BN层)
3.使用了Dropout防止过拟合
from tensorflow.keras import layers, models, Model, Sequential
class AlexNet(Model):
def __init__(self):
super(AlexNet, self).__init__()
# 依次建立卷积层/标准化层/激活层/池化层，一共8层
# 96通道，卷积核改成3x3(原始是11x11)
self.c1 = Conv2D(filters=96, kernel_size=(3, 3))
self.b1 = BatchNormalization()
# ReLu激活函数
self.a1 = Activation('relu')
# 最大池化3*3池化层，步长为2
self.p1 = MaxPool2D(pool_size=(3, 3), strides=2)
self.c2 = Conv2D(filters=256, kernel_size=(3, 3))
self.b2 = BatchNormalization()
self.a2 = Activation('relu')
self.p2 = MaxPool2D(pool_size=(3, 3), strides=2)
# 384个通道，3*3卷积核，使用全零填充，激活函数为ReLu
self.c3 = Conv2D(filters=384, kernel_size=(3, 3), padding='same', activation='relu')
self.c4 = Conv2D(filters=384, kernel_size=(3, 3), padding='same', activation='relu')
self.c5 = Conv2D(filters=256, kernel_size=(3, 3), padding='same', activation='relu')
self.p3 = MaxPool2D(pool_size=(3, 3), strides=2)
# 将像素拉直
self.flatten = Flatten()
# 设置全连接层2048个神经元，激活函数为ReLu
self.f1 = Dense(2048, activation='relu')
# 舍弃50%的神经元
self.d1 = Dropout(0.5)
self.f2 = Dense(2048, activation='relu')
self.d2 = Dropout(0.5)
# 全连接层，神经元为10，采用softmax激活函数
self.f3 = Dense(10, activation='softmax')
# 前向传播
def call(self, inputs): # [batch_size, 28, 28, 1]
x = self.c1(inputs) # [batch_size, 26, 26, 96] 卷积前后计算公式：输出大小=[(输入大小-卷积核大小+2*Padding）/步长]+1
x = self.b1(x) # [batch_size, 26, 26, 96]
x = self.a1(x) # [batch_size, 26, 26, 96]
x = self.p1(x) # [batch_size, 13, 13, 96] 池化3x3，步长为2 (26-3)/2+1=13
x = self.c2(x) # [batch_size, 11, 11, 256] 13-3+1=11
x = self.b2(x) # [batch_size, 11, 11, 256]
x = self.a2(x) # [batch_size, 11, 11, 256]
x = self.p2(x) # [batch_size, 5, 5, 256] (11-3)/2+1=5
x = self.c3(x) # [batch_size, 5, 5, 384] padding=same 输出大小=[输入/步长]
x = self.c4(x) # [batch_size, 5, 5, 384]
x = self.c5(x) # [batch_size, 5, 5, 256]
x = self.p3(x) # [batch_size, 2, 2, 256] (5-3)/2+1=2
x = self.flatten(x) # [batch_size, 1024]
x = self.f1(x) # [batch_size, 2048]
x = self.d1(x) # [batch_size, 2048]
x = self.f2(x) # [batch_size, 2048]
x = self.d2(x) # [batch_size, 2048]
y = self.f3(x) # [batch_size, 10]
return
定义GoogLeNet网络结构
GoogLeNet-Inception 共四个版本，这里选择了v3
inception模块结构图：
GoogLeNet创新点：
v1:引入了Inception结构,并使用1x1卷积和来压缩通道数（减少参数量。Inception作用：代替人工确定卷积层中的过滤器类型或者确定是否需要创建卷积层和池化层，即：不需要人为的决定使用哪个过滤器，是否需要池化层等，由网络自行决定这些参数，可以给网络添加所有可能值，将输出连接起来，网络自己学习它需要什么样的参数。
v2：引入BN层，BN作用：加速网络训练/防止梯度消失。
v3：(1)将Inception内部的BN层推广到外部。(2)优化了网络结构，将较大的二维卷积拆成两个较小的一维卷积，比如将3x3拆成1x3和3x1。这样节省了大量参数，加速运算，并增加了一层非线性表达能力。
v4：引入残差结构。
from tensorflow.keras import layers, models, Model, Sequential
from tensorflow.keras.layers import Conv2D, BatchNormalization, Activation, MaxPool2D, Dropout, Flatten, Dense, ReLU, GlobalAveragePooling2D
# 构建单个卷积组
class ConvBNRelu(Model):
# 默认卷积核边长是3，步长为1，使用全零填充
def __init__(self, ch, kernelsz=3, strides=1, padding='same'):
super(ConvBNRelu, self).__init__()
# 设置sequence
self.model = tf.keras.models.Sequential([
# 卷积核个数/卷积核尺寸/卷积步长/是否全零填充
Conv2D(ch, kernelsz, strides=strides, padding=padding),
BatchNormalization(), # 标准化BN层
Activation('relu') # ReLU激活函数
])
def call(self, x):
# 在training=False时，BN通过整个训练集计算均值、方差去做批归一化，training=True时，通过当前batch的均值、方差去做批归一化。
x = self.model(x, training=False)
return x
# 构建单个inception模块
class InceptionBlk(Model):
def __init__(self, ch, strides=1):
super(InceptionBlk, self).__init__()
self.ch = ch
self.strides = strides
# 设置各卷积的内容，按照inception的结构依次设置
self.c1 = ConvBNRelu(ch, kernelsz=1, strides=strides)
self.c2_1 = ConvBNRelu(ch, kernelsz=1, strides=strides)
self.c2_2 = ConvBNRelu(ch, kernelsz=3, strides=1)
self.c3_1 = ConvBNRelu(ch, kernelsz=1, strides=strides)
self.c3_2 = ConvBNRelu(ch, kernelsz=5, strides=1)
# 3*3卷积核，步长为1，全零填充
self.p4_1 = MaxPool2D(3, strides=1, padding='same')
self.c4_2 = ConvBNRelu(ch, kernelsz=1, strides=strides)
# 前向传播
def call(self, x):
x1 = self.c1(x)
x2_1 = self.c2_1(x)
x2_2 = self.c2_2(x2_1)
x3_1 = self.c3_1(x)
x3_2 = self.c3_2(x3_1)
x4_1 = self.p4_1(x)
x4_2 = self.c4_2(x4_1)
# 将4个部分叠加起来，深度为3
x = tf.concat([x1, x2_2, x3_2, x4_2], axis=3)
return x
class Inception(Model):
# 设置默认ch=16，就是16个卷积核
def __init__(self, num_blocks, num_classes, init_ch=16, **kwargs):
super(Inception, self).__init__(**kwargs)
self.in_channels = init_ch
self.out_channels = init_ch
self.num_blocks = num_blocks
self.init_ch = init_ch
# 直接为初始化中的值（3*3卷积核，步长为1）
self.c1 = ConvBNRelu(init_ch)
# 调用定义的sequence
self.blocks = tf.keras.models.Sequential()
# 后面调用时，num_blocks=2，即有2个block
for block_id in range(num_blocks):
for layer_id in range(2):
if layer_id == 0:
# 第1层，执行inception block，步长为2
block = InceptionBlk(self.out_channels, strides=2)
else:
# 第2层，执行inception block，步长为1
block = InceptionBlk(self.out_channels, strides=1)
self.blocks.add(block)
# enlarger out_channels per block
# 因为步长不一样，所以深度加深，保证输出特征抽取中信息的承载量一致
self.out_channels *= 2
# 最终经过inception后变为128个通道的数据，送入平均池化
# 平均池化层
self.p1 = GlobalAveragePooling2D()
# num_classes为分类数量
self.f1 = Dense(num_classes, activation='softmax')
def call(self, x):
# 执行4个结构
x = self.c1(x)
x = self.blocks(x)
x = self.p1(x)
y = self.f1(x)
return
定义ResNet网络结构
这里选择了ResNet18
class ResnetBlock(Model):
def __init__(self, filters, strides=1, residual_path=False):
super(ResnetBlock, self).__init__()
self.filters = filters
self.strides = strides
self.residual_path = residual_path
self.c1 = Conv2D(filters, (3, 3), strides=strides, padding='same', use_bias=False)
self.b1 = BatchNormalization()
self.a1 = Activation('relu')
self.c2 = Conv2D(filters, (3, 3), strides=1, padding='same', use_bias=False)
self.b2 = BatchNormalization()
# residual_path为True,对输入进行下采样，即用1*1的卷积核，保证x和F(x)维度相同，顺利相加。
if residual_path:
self.down_c1 = Conv2D(filters, (1, 1), strides=strides, padding='same', use_bias=False)
self.down_b1 = BatchNormalization()
self.a2 = Activation('relu')
def call(self, inputs):
residual = inputs # residual等于输入本身, 即residual=x
# 将输入通过卷积、BN层、激活层、计算F（x）
x = self.c1(inputs)
x = self.b1(x)
x = self.a1(x)
x = self.c2(x)
y = self.b2(x)
if self.residual_path:
residual = self.down_c1(inputs)
residual = self.down_b1(residual)
out = self.a2(y + residual) # 最后输出的是两部分的和，即F(x)或F(x)+Wx，再使用relu激活函数
return out
class ResNet18(Model):
def __init__(self, block_list, initial_filters=64): # block_list表示每个block有几个卷积层
super(ResNet18, self).__init__()
self.num_block = len(block_list) # 共有几个block
self.block_list = block_list
self.out_filters = initial_filters
# 原论文输入是224*224，3通道的图像，采用7*7的卷积核，这里对参数进行修改
self.c1 = Conv2D(self.out_filters, (3, 3), strides=1, padding='same', use_bias=False)
self.b1 = BatchNormalization()
self.a1 = Activation('relu')
self.blocks = tf.keras.Sequential()
# 构建ResNet网络结构
# 后面调用为[2,2,2,2]，即总共四个resnetblock，每个block有2个卷积层
for block_id in range(len(block_list)): # 第几个resnetblock
for layer_id in range(block_list[block_id]): # 这个block有几个卷积层
if block_id != 0 and layer_id == 0: # 对除第一个block以外每个block的输入进行下采样
block = ResnetBlock(self.out_filters, strides=2, residual_path=True)
else:
block = ResnetBlock(self.out_filters, residual_path=False)
self.blocks.add(block)
# self.out_filters *= 2 # 下一个block的卷积核数为2倍
self.p1 = tf.keras.layers.GlobalAveragePooling2D()
self.f1 = tf.keras.layers.Dense(
10, activation='softmax', kernel_regularizer=tf.keras.regularizers.l2())
def call(self, inputs):
x = self.c1(inputs)
x = self.b1(x)
x = self.a1(x)
x = self.blocks(x)
x = self.p1(x)
y = self.f1(x)
return
训练——MLP-Mnist
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = MLP()
data_loader = MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
测试——MLP-Mnist
准确率：97.76%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.977600
训练——MLP-Fashion_MNIST
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = MLP()
data_loader = Fashion_MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
print("epoch %d" % (e + 1))
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
print("batch %d: loss %f" % (batch_index, loss.numpy()))
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
测试——MLP-Fashion_MNIST
准确率：87.73%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.877300
训练——MLP-HWDB1
进行100轮训练
num_epochs = 100
batch_size = 50
learning_rate = 0.001
model = MLP()
data_loader = HWDB1Loader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
测试——MLP-HWDB1
准确率：84.17%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.841667
训练——LeNet-Mnist
训练10轮
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = LeNet()
data_loader = MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
测试——LeNet-Mnist
准确率：98.68%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.986800
训练——LeNet-Fashion_MNIST
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = LeNet()
data_loader = Fashion_MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
测试——LeNet-Fashion_MNIST
准确率：85.82%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.858200
训练——LeNet-HWDB1
进行100轮训练
num_epochs = 100
batch_size = 50
learning_rate = 0.001
model = LeNet()
data_loader = HWDB1Loader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
测试——LeNet-HWDB1
准确率：91.33%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.913333
训练——AlexNet-Mnist
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = AlexNet()
data_loader = MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 0.067972
epoch 2: loss 0.008088
epoch 3: loss 0.002195
epoch 4: loss 0.000097
epoch 5: loss 0.008648
epoch 6: loss 0.002917
epoch 7: loss 0.024030
epoch 8: loss 0.000092
epoch 9: loss 0.003540
epoch 10: loss 0.003601
测试——AlexNet-Mnist
准确率：98.91%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.989100
训练——AlexNet-Fashion_MNIST
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = AlexNet()
data_loader = Fashion_MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 0.512883
epoch 2: loss 0.372691
epoch 3: loss 0.317794
epoch 4: loss 0.209880
epoch 5: loss 0.259238
epoch 6: loss 0.246677
epoch 7: loss 0.325049
epoch 8: loss 0.085606
epoch 9: loss 0.231111
epoch 10: loss 0.108106
测试——AlexNet-Fashion_MNIST
准确率:90.57%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.905700
训练——AlexNet-HWDB1
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = AlexNet()
data_loader = HWDB1Loader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 2.154262
epoch 2: loss 1.550519
epoch 3: loss 1.117614
epoch 4: loss 0.796395
epoch 5: loss 0.482055
epoch 6: loss 0.261059
epoch 7: loss 0.598217
epoch 8: loss 0.408028
epoch 9: loss 0.527755
epoch 10: loss 0.232926
测试——AlexNet-HWDB1
准确率：89.67%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.896667
训练——GoogLeNet-Mnist
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = Inception(num_blocks=2, num_classes=10)
data_loader = MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 0.226245
epoch 2: loss 0.046722
epoch 3: loss 0.043534
epoch 4: loss 0.005478
epoch 5: loss 0.005200
epoch 6: loss 0.020319
epoch 7: loss 0.087346
epoch 8: loss 0.008595
epoch 9: loss 0.035517
epoch 10: loss 0.025345
测试——GoogLeNet-Mnist
准确率：99.27%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.992700
训练——GoogLeNet-Fashion_MNIST
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = Inception(num_blocks=2, num_classes=10)
data_loader = Fashion_MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 0.555443
epoch 2: loss 0.369227
epoch 3: loss 0.424934
epoch 4: loss 0.332159
epoch 5: loss 0.254867
epoch 6: loss 0.216396
epoch 7: loss 0.330004
epoch 8: loss 0.130716
epoch 9: loss 0.225775
epoch 10: loss 0.072902
测试——GoogLeNet-Fashion_MNIST
准确率：90.27%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.902700
训练——GoogLeNet-HWDB1
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = Inception(num_blocks=2, num_classes=10)
data_loader = HWDB1Loader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 2.175095
epoch 2: loss 1.401912
epoch 3: loss 1.073104
epoch 4: loss 0.582406
epoch 5: loss 0.561393
epoch 6: loss 0.342880
epoch 7: loss 0.225490
epoch 8: loss 0.534456
epoch 9: loss 0.142884
epoch 10: loss 0.083469
测试——GoogLeNet-HWDB1
准确率:91.50%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.915000
训练——ResNet-Mnist
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = ResNet18([2, 2, 2, 2])
data_loader = MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 0.009949
epoch 2: loss 0.026236
epoch 3: loss 0.016156
epoch 4: loss 0.000139
epoch 5: loss 0.003694
epoch 6: loss 0.004449
epoch 7: loss 0.001600
epoch 8: loss 0.001913
epoch 9: loss 0.001122
epoch 10: loss 0.005298
测试——ResNet-Mnist
准确率：99.21%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.992100
训练——ResNet-Fashion_MNIST
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = ResNet18([2, 2, 2, 2])
data_loader = Fashion_MNISTLoader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
print("epoch %d" % (e + 1))
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
print("batch %d: loss %f" % (batch_index, loss.numpy()))
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
测试——ResNet-Fashion_MNIST
准确率：91.35%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.913500
训练——ResNet-HWDB1
进行10轮训练
num_epochs = 10
batch_size = 50
learning_rate = 0.001
model = ResNet18([2, 2, 2, 2])
data_loader = HWDB1Loader()
optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)
num_batches = int(data_loader.num_train_data // batch_size)
for e in range(num_epochs):
for batch_index in range(num_batches):
X, y = data_loader.get_batch(batch_size)
with tf.GradientTape() as tape:
y_pred = model(X)
loss = tf.keras.losses.sparse_categorical_crossentropy(
y_true=y, y_pred=y_pred)
loss = tf.reduce_mean(loss)
grads = tape.gradient(loss, model.variables)
optimizer.apply_gradients(grads_and_vars=zip(grads, model.variables))
print("epoch %d: loss %f" % (e + 1, loss.numpy()))
epoch 1: loss 2.286184
epoch 2: loss 1.170036
epoch 3: loss 0.718102
epoch 4: loss 0.766253
epoch 5: loss 0.271660
epoch 6: loss 0.121803
epoch 7: loss 0.105720
epoch 8: loss 0.029223
epoch 9: loss 0.422725
epoch 10: loss 0.005966
测试——ResNet-HWDB1
准确率：93.67%
sparse_categorical_accuracy = tf.keras.metrics.SparseCategoricalAccuracy()
num_batches = int(data_loader.num_test_data // batch_size)
for batch_index in range(num_batches):
start_index, end_index = batch_index * \
batch_size, (batch_index + 1) * batch_size
y_pred = model.predict(data_loader.test_data[start_index: end_index])
sparse_categorical_accuracy.update_state(
y_true=data_loader.test_label[start_index: end_index], y_pred=y_pred)
print("test accuracy: %f" % sparse_categorical_accuracy.result())
test accuracy: 0.936667
