Thursday, April 25, 2024

Easily Fine-Tune Any Model for Free Locally - No Machine Learning Required

 This video is a hands-on step-by-step tutorial to install AutoTrain by hugging face on local system and then fine-tune any model on custom dataset.





$ conda create -n autotrain python=3.10

$ conda activate autotrain

$ pip install autotrain-advanced

$ conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia

$ conda install -c "nvidia/label/cuda-12.1.0" cuda-nvcc

$ export HF_TOKEN=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

$ autotrain app --host 0.0.0.0 --port 8000


Dataset Used:


text,label

"This movie was fantastic!",positive

"The food at that restaurant was terrible.",negative

"I love this song!",positive

"The customer service was excellent.",positive

"I couldn't finish reading this book.",negative

1 comment:

dianenicoyann said...

Hi Fahd!

I just tried your code on ubuntu 22.04.
It is not working and i get dependency issues when running the autotrain app.

ImportError: cannot import name 'get_full_repo_name' from 'huggingface_hub' (/home/aiqos/miniconda3/envs/autotrain/lib/python3.10/site-packages/huggingface_hub/__init__.py)