keras实现用resnet18非预训练模型对四种天气分类，并打乱数据

以下是使用Keras实现对四种天气分类的代码，其中使用了ResNet18非预训练模型，并对数据进行了打乱。

import tensorflow as tf
from tensorflow.keras.layers import Input, Dense, Flatten
from tensorflow.keras.applications.resnet50 import ResNet50
from tensorflow.keras.models import Model
from tensorflow.keras.preprocessing.image import ImageDataGenerator

# 设置超参数
batch_size = 32
epochs = 10
num_classes = 4
input_shape = (224, 224, 3)

# 创建一个ResNet18模型（不包括头部）
resnet = ResNet50(include_top=False, input_tensor=Input(shape=input_shape))

# 添加自定义的头部
x = resnet.output
x = Flatten()(x)
x = Dense(256, activation='relu')(x)
x = Dense(num_classes, activation='softmax')(x)

# 创建一个新的模型
model = Model(inputs=resnet.input, outputs=x)

# 编译模型
model.compile(loss='categorical_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])

# 加载数据并进行数据增强
train_datagen = ImageDataGenerator(
    rescale=1./255,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True)

train_generator = train_datagen.flow_from_directory(
    'data/train',
    target_size=input_shape[:2],
    batch_size=batch_size,
    class_mode='categorical')

validation_datagen = ImageDataGenerator(rescale=1./255)

validation_generator = validation_datagen.flow_from_directory(
    'data/validation',
    target_size=input_shape[:2],
    batch_size=batch_size,
    class_mode='categorical')

# 训练模型
model.fit(
    train_generator,
    steps_per_epoch=len(train_generator),
    epochs=epochs,
    validation_data=validation_generator,
    validation_steps=len(validation_generator))

# 评估模型
test_datagen = ImageDataGenerator(rescale=1./255)

test_generator = test_datagen.flow_from_directory(
    'data/test',
    target_size=input_shape[:2],
    batch_size=batch_size,
    class_mode='categorical')

score = model.evaluate(test_generator, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])

在上面的代码中，我们使用了`ResNet50`模型，并在其顶部添加了自定义的头部。我们还使用`ImageDataGenerator`类来进行数据增强，并对数据进行了打乱。最后，我们训练模型并评估其性能。