ResNet18代码实现[通俗易懂]

大家好，又见面了，我是你们的朋友全栈君。

import tensorflow as tf

from tensorflow import keras

from tensorflow.keras import layers, Sequential, Model, datasets, optimizers

# 自定义的预处理函数

def preprocess(x, y):

# 调用此函数时会自动传入x,y对象，shape为[b,28,28],[b]

# 标准化到0-1

x = 2*tf.cast(x, dtype=tf.float32) / 255.-1

# 转成整型张量

y = tf.cast(y, dtype=tf.int32)

# 返回的x,y将替换传入的x,y参数，从而实现数据的预处理功能

return x, y

# 在线下载，加载CIFAR10数据集

(x,y),(x_test,y_test)= datasets.cifar10.load_data()

# 删除y的一个不必要的维度，[b,1] → [b]

y= tf.squeeze(y,axis= 1)

y_test= tf.squeeze(y_test, axis= 1)

# 打印训练集和测试集的形状

# print(x.shape,y.shape, x_test.shape, y_test.shape)

# 构建训练集对象，随机打乱，预处理，批量化

train_db= tf.data.Dataset.from_tensor_slices((x,y))

train_db= train_db.shuffle(1000).map(preprocess).batch(512)

# 构建测试集对象，预处理，批量化

test_db= tf.data.Dataset.from_tensor_slices((x_test,y_test))

test_db= test_db.map(preprocess).batch(512)

# 从训练集中采样一个Batch，并观察

sample= next(iter(train_db))

# print(‘sample:’,sample[0].shape,sample[1].shape,tf.reduce_min(sample[0]),tf.reduce_max(sample[0]))

class BasicBlock(layers.Layer):

# 残差模块

def __init__(self, filter_num, stride= 1):

super(BasicBlock, self).__init__()

#第一个卷积单元

self.conv1= layers.Conv2D(filter_num, kernel_size=(3,3), strides= stride, padding= ‘same’)

self.bn1= layers.BatchNormalization()

self.relu= layers.Activation(‘relu’)

# 第二个卷积单元

self.conv2= layers.Conv2D(filter_num, kernel_size=(3,3), strides= 1, padding= ‘same’ )

self.bn2= layers.BatchNormalization()

# 通过1*1卷积完成shape匹配

if stride != 1:

self.downsample= Sequential()

self.downsample.add(layers.Conv2D(filter_num, kernel_size= (1,1), strides= stride))

else: # shape匹配，直接短接

self.downsample= lambda x:x

def call(self, inputs, training= None):

# 前向计算函数

# [b,h,w,c], 通过第一个卷积单元

out= self.conv1(inputs)

out= self.bn1(out)

out= self.relu(out)

# 通过第二个卷积单元

out= self.conv2(out)

out= self.bn2(out)

# 通过identity模块

identity= self.downsample(inputs)

# 两条路径输出直接相加

output= layers.add([out,identity])

output= tf.nn.relu(output)

return output

class ResNet(Model):

def __init__(self, layer_dims, num_classes= 10): #[2,2,2,2]

super(ResNet, self).__init__()

# 根网络，预处理

self.stem= Sequential([

layers.Conv2D(64, kernel_size= (3,3), strides= (1,1)),

layers.BatchNormalization(),

layers.Activation(‘relu’),

layers.MaxPool2D(pool_size=(2,2), strides=(1,1), padding= ‘same’)

])

# 堆叠4个Block,每个Block包含了多个BasicBlock，设置步长不一样

self.layer1= self.build_resblock(64, layer_dims[0])

self.layer2= self.build_resblock(128, layer_dims[1], stride= 2)

self.layer3= self.build_resblock(256, layer_dims[2], stride= 2)

self.layer4= self.build_resblock(512, layer_dims[3], stride= 2)

# 通过Pooling层将高宽降低为1*1

self.avgpool= layers.GlobalAveragePooling2D()

# 最后连接一个全连接层分类

self.fc= layers.Dense(num_classes)

def build_resblock(self, filter_num, blocks, stride= 1):

# 辅助函数，堆叠filter_num个BasicBlock

res_blocks= Sequential()

# 只有第一个BasicBlock的步长可能不为1， 实现下采样

res_blocks.add(BasicBlock(filter_num, stride))

# 其他BasicBlock步长都为1

for _ in range(1, blocks):

res_blocks.add(BasicBlock(filter_num, stride= 1))

return res_blocks

def call(self, inputs, training= None):

# 前向计算函数：通过根网络

x= self.stem(inputs)

# 一次通过4个模块

x= self.layer1(x)

x= self.layer2(x)

x= self.layer3(x)

x= self.layer4(x)

# 通过池化层

x= self.avgpool(x)

# 通过全连接层

x= self.fc(x)

return x

def resnet18():

# 通过调整模块内部BasicBlock的数量和配置实现不同的ResNet

return ResNet([2,2,2,2])

# def resnet34():

# # 通过调整模块内部BasicBlock的数量和配置实现不同的ResNet

# return ResNet([3,4,6,3])

model = resnet18() # ResNet18网络

model.build(input_shape=(None, 32, 32, 3))

# model.summary() # 统计网络参数

def main():

optimizer = optimizers.Adam(learning_rate=1e-4)

for epoch in range(10):

for step, (x,y) in enumerate(train_db):

with tf.GradientTape() as tape:

# [b, 32, 32, 3] => [b, 1, 1, 512]

logits= model(x)

y_onehot = tf.one_hot(y, depth=10)

# compute loss

loss = tf.losses.categorical_crossentropy(y_onehot, logits, from_logits=True)

loss = tf.reduce_mean(loss)

# 对所有参数求梯度

grads= tape.gradient(loss, model.trainable_variables)

# 自动更新

optimizer.apply_gradients(zip(grads,model.trainable_variables))

if step %10 == 0:

print(epoch, step, ‘loss:’, float(loss))

total_num = 0

total_correct = 0

for x,y in test_db:

# out = model(x)

# out = tf.reshape(out, [-1, 512])

logits = model(x)

prob = tf.nn.softmax(logits, axis=1)

pred = tf.argmax(prob, axis=1)

pred = tf.cast(pred, dtype=tf.int32)

correct = tf.cast(tf.equal(pred, y), dtype=tf.int32)

correct = tf.reduce_sum(correct)

total_num += x.shape[0]

total_correct += int(correct)

acc = total_correct / total_num

print(epoch, ‘acc:’, acc)

if __name__ == ‘__main__’:

main()

发布者：全栈程序员栈长，转载请注明出处：https://javaforall.cn/141598.html原文链接：https://javaforall.cn