[BugFix] GPT inference error when pipeline_para_size > 1 and int8_mode != 0#750
Open
00why00 wants to merge 1 commit intoNVIDIA:mainfrom
Open
[BugFix] GPT inference error when pipeline_para_size > 1 and int8_mode != 0#75000why00 wants to merge 1 commit intoNVIDIA:mainfrom
00why00 wants to merge 1 commit intoNVIDIA:mainfrom
Conversation
00why00
changed the title
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The problem is caused by the empty tensor check in quant ops in
FasterTransformer/examples/pytorch/gpt/utils/gpt.py
Lines 46 to 47 in f8e42aa
When pipeline_para_size > 1, the current rank does not load the weights of all layers, resulting in quantization errors, and it is necessary to add is_load() check when traversing layers