Support Sharding Overlap #8473
Merged
Support Sharding Overlap #8473
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
|
Thanks for your contribution! |
ZHUI
requested changes
May 21, 2024
| if self.config.use_flash_attention and get_env_device() != "gcu": | ||
| is_casual = is_casual_mask(attention_mask) |
There was a problem hiding this comment.
不要删除这个。if hasattr(self.config, "casual_mask")
paddlenlp/trainer/trainer.py
Outdated
| @@ -1908,6 +1907,13 @@ def get_expected_keys(inputs, keys): | |||
| self.optimizer = mix_precision_utils.MixPrecisionOptimizer(self.optimizer) | |||
| self.optimizer = fleet.distributed_optimizer(self.optimizer) | |||
|
|
|||
| if in_sharding_parallel_mode: | |||
| sharding_parallel_config = set(self.args.sharding_parallel_config.split(" ")) | |||
There was a problem hiding this comment.
training_args 文件里面处理好，就别在这里split了
llm/run_pretrain.py
Outdated
| @@ -223,6 +223,10 @@ class ModelArguments: | |||
| default=None, | |||
| metadata={"help": "num_hidden_layers."}, | |||
| ) | |||
| casual_mask: Optional[bool] = field( | |||
There was a problem hiding this comment.
Suggested change
| casual_mask: Optional[bool] = field( | |
| use_casual_mask: Optional[bool] = field( |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #8473 +/- ##
===========================================
- Coverage 54.29% 54.25% -0.05%
===========================================
Files 617 617
Lines 96339 96368 +29
===========================================
- Hits 52312 52288 -24
- Misses 44027 44080 +53 ☔ View full report in Codecov by Sentry. |
ZHUI
previously approved these changes
May 22, 2024
paddlenlp/trainer/trainer.py
Outdated
| @@ -1908,6 +1907,12 @@ def get_expected_keys(inputs, keys): | |||
| self.optimizer = mix_precision_utils.MixPrecisionOptimizer(self.optimizer) | |||
| self.optimizer = fleet.distributed_optimizer(self.optimizer) | |||
|
|
|||
| if in_sharding_parallel_mode: | |||
| if "split_param" in self.args.sharding_parallel_config: | |||
| self.optimizer._set_all_gather_overlap_forward(True, model) | |||
|
|
sneaxiy
previously approved these changes
May 23, 2024
SylarTiaNII
added a commit
to SylarTiaNII/PaddleNLP
that referenced
this pull request
May 24, 2024
This reverts commit 7aaa788.
wawltor
pushed a commit
that referenced
this pull request
May 24, 2024
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.
PR types
Performance optimization
PR changes
Models
Description
1.支持sharding overlap