上面的代码改进后出现WARNING:tensorflow:Model was constructed with shape (None, 224, 224, 3) for input KerasTensor(type_spec=TensorSpec(shape=(None, 224, 224, 3), dtype=tf.float32, name='input_5'), name='input_5', description="created by layer 'input_5'"), but it was called on an input with incompatible shape (None, 100, 125, 3).

这个警告是因为在构建模型时，我们定义了输入张量的形状为 (None, 224, 224, 3)，但实际使用时输入张量的形状为 (None, 100, 125, 3)，两者不匹配。这是因为我们将 MobileNetV2 模型的输入形状定义为 (224, 224, 3)，而在新模型中我们定义了新的输入形状为 (100, 125, 3)。

虽然这个警告不会影响模型的训练和预测，但为了消除警告，我们可以将 MobileNetV2 模型的输入形状与新模型的输入形状匹配。具体来说，我们可以使用 `tf.keras.layers.experimental.preprocessing` 中的 `Resizing` 层将输入张量的大小调整为 (224, 224)。修改后的代码如下所示：

from tensorflow.keras import layers, models
from tensorflow.keras.layers.experimental.preprocessing import Resizing

# Define your new input layer with the desired shape
new_input = layers.Input(shape=(100, 125, 3))

# Load the MobileNetV2 model with the top layer removed
mobile = tf.keras.applications.MobileNetV2(input_shape=(224, 224, 3), include_top=False, weights='imagenet')

# Freeze the weights of the MobileNetV2 layers
for layer in mobile.layers:
    layer.trainable = False

# Resize the input to match the input shape of the MobileNetV2 model
resized_input = Resizing(224, 224)(new_input)

# Get the output of the MobileNetV2 model
x = mobile(resized_input)

# Add a global average pooling layer and a dense layer for classification
x = layers.GlobalAveragePooling2D()(x)
x = layers.Dense(128, activation='relu')(x)
predictions = layers.Dense(10, activation='softmax')(x)

# Create the new model
new_model = models.Model(inputs=new_input, outputs=predictions)

# Compile the new model
new_model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])

# Print the summary of the new model
new_model.summary()

在这个修改后的代码中，我们使用 `Resizing` 层将输入张量的大小调整为 (224, 224)，然后将调整后的张量作为 MobileNetV2 模型的输入。这样可以保证输入形状的匹配，并消除警告。