Fix for triton dynamic batching requirement #890
Merged
Conversation
This file contains 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
|
Hello @rajeevsrao, I ran into an issue when apply converted TRT engine to Triton and apply dynamic batching. I think the code could use a simple fix. Original code always output a static engine when only 1 optimization profile is provided. |
according to the definition of batch size in bert demo: `parser.add_argument("-b", "--batch-size", default=[], action="append", help="Batch size(s) to optimize for. The engine will be usable with any batch size below this, but may not be optimal for smaller sizes. Can be specified multiple times to optimize for more than one batch size.", type=int)`
The min batch size in the optimization profile should always start with 1, instead of increasing by 1 for each separate optimization profile such as [1, 2, 3]
There is no point to output a static batch optimization profile based on the batch size definition.
Signed-off-by: DougM <mengdong0427@gmail.com>
|
Thanks @mengdong - will review. |
|
@rajeevsrao could you merge this so that we can use bert engine with bs>1 and multiple optimization profiles in Triton correctly? Thanks! |
|
thanks! |
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Outdated suggestions cannot be applied.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.
according to the definition of batch size in bert demo:
parser.add_argument("-b", "--batch-size", default=[], action="append", help="Batch size(s) to optimize for. The engine will be usable with any batch size below this, but may not be optimal for smaller sizes. Can be specified multiple times to optimize for more than one batch size.", type=int)The min batch size in the optimization profile should always start with 1, instead of increasing by 1 for each separate optimization profile such as [1, 2, 3]
There is no point to output a static batch optimization profile based on the batch size definition.