terngrad icon indicating copy to clipboard operation
terngrad copied to clipboard

Published 20 hours ago • verified publisher icon wenwei202

Why we have to build tensorflow from source?

Open GeKeShi opened this issue 5 years ago • 4 comments

Can we install the official TensorFlow binary version and run your code based on it? Does that mean you changed the source code of TensorFlow? I find that your algorithm just inserts some tensor operation in the computation graph, but have not changed the source code of TensorFlow.

GeKeShi avatar Mar 30 '19 09:03 GeKeShi

I have this question too. Have you figured out the necessary to build from source?

woskii avatar Apr 15 '19 02:04 woskii

I have this question too. Have you figured out the necessary to build from source?

no, I haven't, the author implements the code in official Pytorch code, it means it will make no sense to build the tensorflow version.

GeKeShi avatar Apr 18 '19 06:04 GeKeShi

I have this question too. Have you figured out the necessary to build from source?

no, I haven't, the author implements the code in official Pytorch code, it means it will make no sense to build the tensorflow version.

Can you post a link of the pytorch implementation? Thanks a lot.

woskii avatar Apr 18 '19 08:04 woskii

I have this question too. Have you figured out the necessary to build from source?

no, I haven't, the author implements the code in official Pytorch code, it means it will make no sense to build the tensorflow version.

Can you post a link of the pytorch implementation? Thanks a lot.

The author post that link in the README of this repo

GeKeShi avatar May 06 '19 02:05 GeKeShi