使用model.save()保存所需的内容，清除tf会话，然后再次加载以解决此问题(如@NatthaphonHongcharoen所建议(在注释中):

Saving the desired with model.save() , clearing tf session and then loading it again fixed the issue(as suggested @NatthaphonHongcharoen (in the comments):

model.save(model_file)
del model
keras.backend.clear_session()
model = keras.models.load_model(model_file)

现在在tensorflow中导出的图仅显示所需的图层，冻结的图将生成较小的.pb文件.

Now the exported graph in tensorflow shows only the desired layers and the frozen graph generates a smaller .pb file.

然而，即使使用基础模型的单层并添加编译器仍然说的Dense层，仍然存在另一个问题

However another issue persists, even using a single layer of the base model and adding a Dense layer the compiler still says

设备上的内存不足，无法运行模型.

Not enough on-device memory to run model.

但这是一个没有直接在这个问题中问到的问题.

But that is a different issue not directly asked in this question.