tensorflow.python.keras和tensorflow.keras有什么区别?

问题描述：

正如标题所述，它们是相同的api吗?当我在keras中打印layers模块时，结果如下所示:

As the title says, Are they the same api? When I print the layers module in keras, the result are shown as follow:

from tensorflow.keras import layers
print(layers)
from tensorflow.python.keras import layers
print(layers)

结果

<module 'tensorflow.python.keras.api._v1.keras.layers' from '/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/api/_v1/keras/layers/__init__.py'>
<module 'tensorflow.python.keras.layers' from '/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/layers/__init__.py'>

我们可以看到两个模块来自不同的来源.

We can see that two modules come from different source.

我从源代码中找到api模块，只有一个BUILD文件.

And I find the api module from source code, there is only a BUILD file.

两个模块之间是否存在关系，api生成器的机制是什么?

Is there a relation between two modules, what is the mechanism of the api generator?

答

tf.python.*下的任何内容都是私有的，仅用于开发，而不用于公共用途.

Anything under tf.python.* is private, intended for development only, rather than for public use.

不支持从tensorflow.python或任何其他模块(包括import tensorflow_core ...)导入，并且可能会突然中断.

Importing from tensorflow.python or any other modules (including import tensorflow_core...) is not supported, and can break unannounced.

因此，建议不要在tf.python.*中使用任何东西.

So, it is suggested not to use anything with tf.python.*.

tensorflow.python.keras和tensorflow.keras有什么区别?

相关推荐