Skip to content

port code to tensorflow2.0#87

Open
punitvara wants to merge 1 commit intomicrosoft:masterfrom
punitvara:full_tf2.0
Open

port code to tensorflow2.0#87
punitvara wants to merge 1 commit intomicrosoft:masterfrom
punitvara:full_tf2.0

Conversation

@punitvara
Copy link
Contributor

Compile and run code with tf2.0. Eager execution is
disabled. Tested code with python 3.5

Signed-off-by: Punit Vara punitvara@gmail.com

Compile and run code with tf2.0. Eager execution is
disabled. Tested code with python 3.5

Signed-off-by: Punit Vara <punitvara@gmail.com>
@punitvara
Copy link
Contributor Author

punitvara commented May 20, 2019

Either #86 or this PR should get merged. Preferably this one. @adityakusupati Can you please review this ?

@adityakusupati
Copy link
Contributor

adityakusupati commented May 24, 2019

I will try to set it up next week and test it out. @metastableB and @harsha-simhadri any thoughts? I think this is important for tf 2.0 transition.

@adityakusupati
Copy link
Contributor

@punitvara , a couple of things

  1. Bonsai example runs fine, but FastCells don't
  2. Can you isolate the changes you made from the original files, this diff is very cumbersome to follow?
@punitvara
Copy link
Contributor Author

Ok sure. I will push in tf directory once review is done I will create new PR with tf 2.0

@metastableB
Copy link
Contributor

@adityakusupati I am of the opinion that we should hold out on this PR for a little bit longer till we have a clear view on what we are going to focus on; eager or static compilation. If you feel like going ahead with supporting static graphs with TF2.0, please do so. You can get a clean diff w.r.t tf/ directory by using the native diff command as explained here. You will have to go through it one file at a tile.

@punitvara Thanks for your contributions!

@cdknorow
Copy link

cdknorow commented Oct 8, 2021

@metastableB I've been doing some work to do a port to tf2 for some functions (not using tf1.compat, but full tf2 with eager execution) I wouldn't mind putting together a PR for the work as I go through. But only if there is interest on your side.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

4 participants