Add a "no_conversion" flow to torch-tensorrt #1360
Closed
Conversation
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
Adds a "no_conversion" option to torch-tensorrt which when enabled will replace the standard conversion and and engine insertion with an embedded function call for each convertible segment. This allows inspection of the partition without running conversion and the possibility to convert each engine individually in subsequent runs. Fixes # (issue) Please delete options that are not relevant and/or add your own. - Bug fix (non-breaking change which fixes an issue) - New feature (non-breaking change which adds functionality) - Breaking change (fix or feature that would cause existing functionality to not work as expected) - This change requires a documentation update - [ ] My code follows the style guidelines of this project (You can use the linters) - [ ] I have performed a self-review of my own code - [ ] I have commented my code, particularly in hard-to-understand areas and hacks - [ ] I have made corresponding changes to the documentation - [ ] I have added tests to verify my fix or my feature - [ ] New and existing unit tests pass locally with my changes - [ ] I have added the relevant labels to my PR in so that relevant reviewers are notified
github-actions
bot
added
component: api [Python]
github-actions
bot
requested review from
andi4191,
bowang007,
narendasan and
peri044
Collaborator
|
@bowang007 Could something like this solve our graph stitching issues? Say if the way partitioning works is it creates a bunch of methods but a method is either 100% PyTorch or 100% TRT. May make things like collections way easier too. |
|
This PR has not seen activity for 90 days, Remove stale label or comment or this will be closed in 10 days |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Adds a "no_conversion" option to torch-tensorrt which when enabled will replace the standard conversion and engine insertion with an embedded function call for each convertible segment. This allows inspection of the partition without running conversion and the possibility to convert each engine individually in subsequent runs.
Future work:
Allow this flow to run without a GPU to enable TRT convertibility/partitioning linting flows on host machines
Partition without running shape propagation when in the no-convert flow
Fixes # (#1361)
Please delete options that are not relevant and/or add your own.
New feature (non-breaking change which adds functionality)
My code follows the style guidelines of this project (You can use the linters)
I have performed a self-review of my own code
I have commented my code, particularly in hard-to-understand areas and hacks
I have made corresponding changes to the documentation
I have added tests to verify my fix or my feature
New and existing unit tests pass locally with my changes
I have added the relevant labels to my PR in so that relevant reviewers are notified