Ho, ho, ho!!! Tis the season of holidays and for my Christmas present I received,

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[4], line 1
----> 1 pipe = pipeline(TASK, model=MODEL_NAME, torch_dtype=torch.bfloat16, device_map=0)

...
    raise ValueError(
ValueError: Using a `device_map`, `tp_plan`, `torch.device` context manager or setting `torch.set_default_device(device)` requires `accelerate`. You can install it with `pip install accelerate`

when trying to run an LLM from Hugging Face. Wutdafuk? Umm... Wutdafuk :/

The solution seems simple. Just install the accelerate package. Only problem is I already have it installed and I have checked and ensured that Im using the correct conda env.

And solution? Just had to update the accelerate package version. I didnt initially try this because I have used the older version before in a project and it worked fine and somehow now it breaks. Weird. But whatever. Remember kids, always update your packages.

Merry Christmas!!!