assisted_decoding called directly inside generate triggering warning to use when it shouldn't

[ofirzaf]

System Info

• transformers version: 4.39.1
• Python version: 3.10.14
• Huggingface_hub version: 0.22.0
• Safetensors version: 0.4.2
• Accelerate version: not installed
• Accelerate config: not found
• PyTorch version (GPU?): 2.2.1+cpu (False)
• Tensorflow version (GPU?): not installed (NA)
• Flax version (CPU?/GPU?/TPU?): not installed (NA)
• Jax version: not installed
• JaxLib version: not installed
• Using GPU in script?: No
• Using distributed or parallel set-up in script?: No

Who can help?

@gante

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

out = model.generate(
    **inputs,
    max_new_tokens=128,
    streamer=TextStreamer(tokenizer=tokenizer, skip_special_tokens=True),
    pad_token_id=tokenizer.eos_token_id,
    prompt_lookup_num_tokens=3,
)

This triggers the following warning:

Calling `_assisted_decoding` directly is deprecated and will be removed in v4.41. Use `generate` or a custom generation loop instead.

This happens probably due to the following call which should be changed to self._assisted_decoding(...)

transformers/src/transformers/generation/utils.py

Line 1509 in 7eb3ba8

result = self.assisted_decoding(

Expected behavior

Warning shouldn't be triggered

[zucchini-nlp]

Thanks for opening an issue. Yes, there is a PR fixing this behavior. You can update transformers with
!pip install --upgrade git+https://github.com/huggingface/transformers.git

after the PR is merged 🤗