Zach Mueller

@SunMarc updating cuda drivers will solve this :) (Tested on the 4090's): No need for fancy env settings etc, just do `python myscript.py` ```python import torch from accelerate.utils import send_to_device...

@vladmandic can you just do a try/catch in what gradio is running? I need an example of your gradio script and how it's being utilized for more ideas.

@vladmandic your attempt there is right, you need to call it purly pythonic and avoid subprocess when possible, that is the only way.

When trying to catch something like this, yes. We cannot guarantee that accelerate is free of subprocess, as certain configurations require us launching it like so. I don't believe we...

Hi @vladmandic, I found another issue requesting the same thing. I'll look into this here relatively soon (promise not months out like before), and we'll put in explicit exceptions for...

@polyrand try installing via: ``` pip install fastinference[onnxcpu] ``` Instead.

@mrwyattii this is because we're integrating with `Accelerate` to handle all the distributed code in `Trainer`. You can see the code we use to set everything up on the `Accelerate`...

Doing so will make `tests/deepspeed/test_deepspeed.py::TestDeepSpeedWithLauncher::test_basic_distributed_zero3_fp16` fail, with the same error as stated. Please try running with: `CUDA_VISIBLE_DEVICES="0,1" RUN_SLOW="yes" ACCELERATE_USE_DEEPSPEED="yes" pytest -sv tests/deepspeed/test_deepspeed.py -k test_basic_distributed` to replicate

Can you try installing via main? Aka: `pip install git+https://github.com/huggingface/accelerate`? And ideally can you tell us the output of `accelerate env`?

@codetomakechange out of curiosity have you found a way to do this *without* accelerate launch? (aka native torch-xla?) If not that's okay, will look into this soon