Repository navigation
Add implicit batch experimental - #34293
Merged
tensorflow-copybara merged 23 commits intoDec 16, 2019
Merged
tensorflow-copybara merged 23 commits into
tensorflow-copybara merged 23 commits into
Conversation
aaroey
requested changes
Nov 18, 2019
pooyadavoodi
force-pushed
the
add_implicit_batch_experimental
branch
from
November 19, 2019 22:32
7e7a5b3 to
84fd786
Compare
aaroey
requested changes
Nov 20, 2019
added 15 commits
December 4, 2019 11:33
Use buildEngineWithConfig from TensorRT 6.0 Enable custom rewriter_config_template
This model can be loaded and executed in tests to ensure backwards compatibility across TF versions.
Contributor
|
@pooyadavoodi Can you please resolve conflicts? Thanks! |
Member
|
Also @sanjoy |
Set use_implicit_batch to true in case it is not present in graph Add parenthesese for readability
pooyadavoodi
force-pushed
the
add_implicit_batch_experimental
branch
from
December 6, 2019 17:54
73cb4f1 to
0c2048a
Compare
Author
Done I've added the commits from #34660 otherwise it will conflict with the master once #34660 is merged. I am hoping github will take care of duplicated commits. |
aaroey
previously approved these changes
Dec 11, 2019
This allows to create a rewriter_config that has all the non-TRT optimizers off.
aaroey
requested changes
Dec 12, 2019
added 2 commits
December 12, 2019 13:47
Disabling optimizers should happen after CopyFrom the temaplte otherwise the template can overwrite the disablement.
This is fine because CopyFrom of rewriter_config_template will overwrite the disablement of optimizers and thus we can skip them.
aaroey
previously approved these changes
Dec 12, 2019
Contributor
|
@pooyadavoodi Could you please address Ubuntu Sanity errors? Thanks! |
aaroey
approved these changes
Dec 16, 2019
tensorflow-bot
Bot
added
kokoro:force-run
tensorflow-copybara
pushed a commit
that referenced
this pull request
Dec 16, 2019
…mental PiperOrigin-RevId: 285784123 Change-Id: I4df92ec74bdc834284cd5695184a8a58b7774f9c
tensorflow-copybara
pushed a commit
that referenced
this pull request
Dec 16, 2019
#34293). PiperOrigin-RevId: 285813192 Change-Id: I38560c42489af14be40a99b78642eb8109437810
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
use_implicit_batch. This is the first step in adding dynamic shapes to TF-TRT, so it still doesn't work but it provides the scaffolding for the next PRs.use_implicit_batchuntil we take it out of experimental.