(Google Colab) ImportError: cannot import name '_expand_mask' from 'transformers.models.bloom.modeling_bloom'

[phatjkk]

Hi !
Thank you for your excellent work!

I got this error when trying to run the model on Google Colab. I use the example code on README file:

from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer  
  
model_path = "vinai/PhoGPT-7B5-Instruct"  
  
config = AutoConfig.from_pretrained(model_path, trust_remote_code=True)  
config.init_device = "cuda"
# config.attn_config['attn_impl'] = 'triton' # Enable if "triton" installed!
  
model = AutoModelForCausalLM.from_pretrained(  
    model_path, config=config, torch_dtype=torch.bfloat16, trust_remote_code=True  
)
# If your GPU does not support bfloat16:
# model = AutoModelForCausalLM.from_pretrained(model_path, config=config, torch_dtype=torch.float16, trust_remote_code=True)
model.eval()  
  
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)  
  
PROMPT = "### Câu hỏi:\n{instruction}\n\n### Trả lời:"  
  
input_prompt = PROMPT.format_map(  
    {"instruction": "Làm thế nào để cải thiện kỹ năng quản lý thời gian?"}  
)  
  
input_ids = tokenizer(input_prompt, return_tensors="pt")  
  
outputs = model.generate(  
    inputs=input_ids["input_ids"].to("cuda"),  
    attention_mask=input_ids["attention_mask"].to("cuda"),  
    do_sample=True,  
    temperature=1.0,  
    top_k=50,  
    top_p=0.9,  
    max_new_tokens=1024,  
    eos_token_id=tokenizer.eos_token_id,  
    pad_token_id=tokenizer.pad_token_id  
)  
  
response = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]  
response = response.split("### Trả lời:")[1]

The error:

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
[<ipython-input-8-2cf92868f260>](https://localhost:8080/#) in <cell line: 13>()
     11 # )
     12 # If your GPU does not support bfloat16:
---> 13 model = AutoModelForCausalLM.from_pretrained(model_path, config=config, torch_dtype=torch.float16, trust_remote_code=True)
     14 model.eval()
     15 

11 frames
[~/.cache/huggingface/modules/transformers_modules/vinai/PhoGPT-7B5-Instruct/8083375bebd52681090be6ebaf8bae7aee491f73/hf_prefixlm_converter.py](https://localhost:8080/#) in <module>
     13 import torch
     14 from transformers.models.bloom.modeling_bloom import BaseModelOutputWithPastAndCrossAttentions, BloomForCausalLM, BloomModel, CausalLMOutputWithCrossAttentions, CrossEntropyLoss
---> 15 from transformers.models.bloom.modeling_bloom import _expand_mask as _expand_mask_bloom
     16 from transformers.models.bloom.modeling_bloom import _make_causal_mask as _make_causal_mask_bloom
     17 from transformers.models.bloom.modeling_bloom import logging

ImportError: cannot import name '_expand_mask' from 'transformers.models.bloom.modeling_bloom' (/usr/local/lib/python3.10/dist-packages/transformers/models/bloom/modeling_bloom.py)

I don't know how to fix it, but I found the same issue at : https://huggingface.co/mosaicml/mpt-7b/discussions/83

Can you tell me how to run it?
Thanks you!

[datquocnguyen]

It's a mismatch with the latest version of transformers. You might want to use an older version of transformers (e.g., 4.33).

I will make a fix to work with the latest version of transformers soon.

[datquocnguyen]

@phatjkk code updated, please help check whether it works fine in your side with the latest transformers.

[kxgong]

check

Hi, I met this problem with 4.36.0/1/2. I update my transformers to use mixtral-8x7b.

How can I solve this issue. Thank you.

[kxgong]

@phatjkk code updated, please help check whether it works fine in your side with the latest transformers.

My older version is 4.31.0 and it was no problem.

[megiandoni]

Also have this issue with 4.36.2

[datquocnguyen]

Please clear your cache of PhoGPT, and re-download it.
https://huggingface.co/vinai/PhoGPT-7B5-Instruct/blob/main/hf_prefixlm_converter.py was updated without any bloom modeling.

[kxgong]

Please clear your cache of PhoGPT, and re-download it. https://huggingface.co/vinai/PhoGPT-7B5-Instruct/blob/main/hf_prefixlm_converter.py was updated without any bloom modeling.

Thank you, I will try this.