keras如何管理卷积层的权重？

from keras.layers import Conv2D

conv_layer = Conv2D(filters=32, kernel_size=(3, 3), activation='relu', kernel_initializer='he_normal')

from keras.models import Sequential
from keras.optimizers import Adam

model = Sequential()
model.add(conv_layer)
model.compile(optimizer=Adam(), loss='categorical_crossentropy')
model.fit(x_train, y_train, epochs=10, batch_size=32)

# 保存权重
model.save_weights('weights.h5')

# 加载权重
model.load_weights('weights.h5')

from keras.optimizers import SGD

model.compile(optimizer=SGD(learning_rate=0.01), loss='categorical_crossentropy')

from keras.layers import Dropout

model.add(Dropout(0.5))