Huggingface Image Classifier

Fine-tune the Vision Transformer (ViT) using LoRA and Optuna for hyperparameter search.

Installation

To install the environment, you need to have Anaconda installed.

Navigate to the Folder

Replace [path_to] with the path to where you have downloaded the folder.

cd [path_to]/huggingface_image_classifier

Install Dependencies

conda env create -f env.yml

Activate the Environment

conda activate hugging_face

Start the Pipeline

To get help for main.py, you can start with:

python main.py -h

Run the main script with:

python main.py