Create a new conda environment, and install the dependencies:
conda create -n lqdm
conda activate lqdm
conda install python pytorch torchvision -c pytorch -y
python3 -m pip install -r requirements.txt
Download the data and model checkpoint from here. Create a new directory datasets/ from the project root, and place the ZIP file inside. Create a new directory checkpoints/ from the project root, and place the checkpoint file inside.
python3 main.py fit -c configs/lqds.yaml --trainer.devices 4 --data.batch_size 4 --data.path
data/
This will train the model from scratch using the dataset from the config. Replace with the number of GPUs and the batch size based on available memory.
python3 main.py validate -c configs/lqds.yaml --trainer.devices 1 --data.batch_size 4 --data.path data/ --model.ckpt_path checkpoints/model.ckpt
This will evaluate metrics for a trained model checkpoint.