Refactor load weights #1603
Open
Refactor load weights #1603
+1,033
−1,365
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
grimoire
force-pushed
the
refactor-load-weights
branch
from
53fe863
to
d7138f1
Compare
RunningLeon
reviewed
Jun 4, 2024
|
@zhulinJulia24 hi, could you start a full-scope test of all pytorch engine models using |
lvhan028
reviewed
Jun 11, 2024
| rank=rank, | ||
| world_size=world_size, | ||
| prefix='query_key_value') | ||
| rowwise_parallelize_linear(self.dense, |
There was a problem hiding this comment.
much better than previous version
lvhan028
reviewed
Jun 11, 2024
| logger = get_logger('lmdeploy') | ||
|
|
||
|
|
||
| def _get_weight_type(model_path: str, use_safetensors: bool = None): |
There was a problem hiding this comment.
use_safetensors can be {True, False, None}. Why not True or False?
There was a problem hiding this comment.
lvhan028
reviewed
Jun 11, 2024
| for name, param in mod.named_parameters(recurse=False): | ||
| dtype = param.dtype | ||
| if not loader.has(name): | ||
| logger.debug(f'rank [{rank}]' |
There was a problem hiding this comment.
How to invoke this condition?
There was a problem hiding this comment.
Some model might shared weight of token embedding, they do not safe redundant weight in checkpoint.
lvhan028
reviewed
Jun 11, 2024
lvhan028
reviewed
Jun 11, 2024
| @@ -160,204 +157,3 @@ def sync_qparam_to_context(context: Any, layer_id: str, qparams: dict): | |||
| context.set_output(layer_id, last_qparam) | |||
| else: | |||
| context.set_output(layer_id, qparams) | |||
|
|
|||
|
|
|||
| @torch.no_grad() | |||
lvhan028
reviewed
Jun 11, 2024
lvhan028
reviewed
Jun 11, 2024
lvhan028
approved these changes
Jun 11, 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.
optimization tp model loading.
requirement