Seers is a simple Python library for vision language model training and education. It provides training recipes and prediction code for cleanly training and deploying vision language models.
Please follow Python environment setup instructions for setting up your environment.
pip install git+https://github.com/facebookresearch/schedule_free.git
pip install peft
git clone https://github.com/mkturkcan/seers.git
cd seers/seers/A simple model example is given at HuggingFace.