Make initialization of tokenizer and detokenizer optional #3748
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
GeauxEric
force-pushed
the
optional-tokenizer
branch
from
69e732d
to
50cc4b7
Compare
|
@GeauxEric Thank you for the contribution! Could you update this branch with the changes merged in the main branch? Happy to review after the conflicts are resolved! |
GeauxEric
force-pushed
the
optional-tokenizer
branch
from
9c6d6cc
to
3a6a0e7
Compare
with a ',', it becomes a tuple, and deemed as True.
GeauxEric
force-pushed
the
optional-tokenizer
branch
from
3a6a0e7
to
676256f
Compare
@ywang96 |
|
Maybe others can help take a look? @simon-mo? |
ywang96
approved these changes
Apr 18, 2024
vllm/engine/arg_utils.py
Outdated
| @@ -95,6 +96,10 @@ def add_cli_args( | |||
| type=str, | |||
| default=EngineArgs.tokenizer, | |||
| help='name or path of the huggingface tokenizer to use') | |||
| parser.add_argument( | |||
| '--skip_tokenizer_init', | |||
There was a problem hiding this comment.
format nit
Suggested change
| '--skip_tokenizer_init', | |
| '--skip-tokenizer-init', |
Comment on lines
+343
to
+346
| eos_token_id = None | ||
| if self.tokenizer: | ||
| eos_token_id = self.tokenizer.get_lora_tokenizer( | ||
| lora_request).eos_token_id |
There was a problem hiding this comment.
I still think it's worth having an warning here about eos_token_id being None.
|
Could you please take another look and consider merging/approving it when you have a moment? |
ywang96
approved these changes
Apr 21, 2024
There was a problem hiding this comment.
@GeauxEric Sorry for the delayed review and thank you for addressing the comments and contributing this PR!
robertgshaw2-neuralmagic
pushed a commit
to neuralmagic/nm-vllm
that referenced
this pull request
Apr 21, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
ZackBradshaw
pushed a commit
to ZackBradshaw/vllm
that referenced
this pull request
Apr 22, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
ZackBradshaw
pushed a commit
to ZackBradshaw/vllm
that referenced
this pull request
Apr 22, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
ZackBradshaw
pushed a commit
to ZackBradshaw/vllm
that referenced
this pull request
Apr 22, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
xjpang
pushed a commit
to xjpang/vllm
that referenced
this pull request
Apr 25, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
robertgshaw2-neuralmagic
pushed a commit
to neuralmagic/nm-vllm
that referenced
this pull request
Apr 26, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
alexeykondrat
pushed a commit
to alexeykondrat/ci-vllm
that referenced
this pull request
May 1, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
z103cb
pushed a commit
to z103cb/opendatahub_vllm
that referenced
this pull request
May 7, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
Temirulan
pushed a commit
to Temirulan/vllm-whisper
that referenced
this pull request
Sep 6, 2024
…ct#3748) Co-authored-by: Yun Ding <yunding@nvidia.com> Co-authored-by: Roger Wang <ywang@roblox.com>
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.
Add a flag to initialize the LLM engine to disable tokenizer.
If tokenizer is disabled, so does detokenizer.
FIX #3635
FIX #3647