Add calibration based INT8 quantization to TensorRT EP #5842
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
stevenlix
changed the title
jywu-msft
reviewed
Nov 18, 2020
jywu-msft
reviewed
Nov 18, 2020
jywu-msft
reviewed
Nov 18, 2020
| static const std::string kDumpSubgraphs = "ORT_TENSORRT_DUMP_SUBGRAPHS"; | ||
| static const std::string kEngineCacheEnable = "ORT_TENSORRT_ENGINE_CACHE_ENABLE"; | ||
| static const std::string kEngineCachePath = "ORT_TENSORRT_ENGINE_CACHE_PATH"; | ||
| static const std::string kCachePath = "ORT_TENSORRT_CACHE_PATH"; |
There was a problem hiding this comment.
this is a compatibility breaking change.
users who are already using previous name, may not know it has changed.
i think it would be safer to add another check to throw error if the previous name is found set in the environment variables.
or we don't change the name at all.
There was a problem hiding this comment.
you can handle this in a subsequent PR?
jywu-msft
approved these changes
Nov 20, 2020
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Outdated suggestions cannot be applied.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.
Native TRT calibration table is only for the models that can run as a whole on native TRT
ORT generated calibration table can also work with models that only its subgraphs can run on TRT. Those subgraphs will run in INT8 precision in TRT EP if possible