vLLM NemoTokenizerGroup bugfix

[janekl]

What does this PR do ?

For NemoTokenizerGroup in the base class BaseTokenizerGroup from vLLM there is an abstract from_config method that needs to be implemented. This change mocks the implementation to make scripts/deploy/nlp/deploy_vllm_triton.py functional.

There error fixed here is:

INFO 09-27 15:02:31 llm_engine.py:176] Initializing an LLM engine (v0.5.3) with config: model='/tmp/tmpioclfxxh', speculative_config=None, tokenizer=None, skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, rope_scaling=None, rope_theta=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=512, download_dir=None, load_format=<class 'nemo.export.vllm.model_loader.NemoModelLoader'>, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto, quantization_param_path=None, device_config=cuda, decoding_config=None, observability_config=None, seed=0, served_model_name=/data/export/LLAMA2-7B-fp8-sft.nemo, use_v2_block_manager=False, enable_prefix_caching=False)
ERROR 09-27 15:02:31 deploy_vllm_triton.py:169] An error has occurred while setting up or serving the model. Error message: An error has occurred during the model export. Error message: Can't instantiate abstract class NemoTokenizerGroup with abstract method from_config

CC @apanteleev

Collection: NLP

Changelog

• Add specific line by line info of high level changes in this PR.

Usage

• You can potentially add a usage example below

# Add a code snippet demonstrating how to use this 

GitHub Actions CI

The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.

The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

• New Feature
• Bugfix
• Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

• Related to vLLM Export Support #9381

[janekl]

FYI @HuiyingLi

[janekl]

I'm closing this as it's covered by #10697