TF.SQRT（）调用后面会发生什么？
我试图了解tf.sqrt（）呼叫（我的意思是呼叫堆）背后会发生什么。我知道gen_math_ops.py是在TF汇编过程中生成的，其中具有SQRT函数。但是似乎SQRT在两个地方实现，
- https://github.com/tensorflow/tensorflow/blob/r2.9/tensorflow/core/core/math_ops.cc.cc
- https://github.com/tensorflow/tensorflow/tensorflow/tensorflow/blob/r.r.rblob/rblob/tensorflow/core/core/core/kernels/kernels/cwise_op_op_op_sqrt.cc.ccccc.ccc.ccc <一下/a>
任何人都可以让我知道它们是如何连接的，以及如何在幕后执行SQRT？
谢谢你！
I am trying to understand what happens behind tf.sqrt() call (I mean the call-stack). I got to know that gen_math_ops.py is generated during TF compilation which has a sqrt function in it. But it seems that sqrt in implemented in two places,
- https://github.com/tensorflow/tensorflow/blob/r2.9/tensorflow/core/ops/math_ops.cc
- https://github.com/tensorflow/tensorflow/blob/r2.9/tensorflow/core/kernels/cwise_op_sqrt.cc
Can anyone please let me know how these are connected and how sqrt is executed behind the scenes?
Thank you!
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论