- Create new conda environement:
conda env create -f environment.yml
- Activate conda environment:
conda activate cl-distilled
- Install pip requirements:
pip install -r requirements.txt
- Download training data:
./scripts/download_training_data.sh
- Download evaluation data:
./scripts/download_eval_data.sh
- Make cl-distilled env available for Jupyter:
python -m ipykernel install --user --name=cl-distilled
And we're good to go! Distilled Face :)
- Source code for modules required by DistilFACE can be found in:
src/distilface/modules/pooler.py
src/distilface/modules/similarity.py
- Main DistilFACE model implementation can be found in the notebooks in folder:
- Section 2 of
notebooks/training/*.ipynb
- Hyperparameter tuning results can be found in all the notebooks in:
- Section 4 of
notebooks/training/*.ipynb
- Install required binaries:
$ apt-get update && apt-get install texlive-latex-base texlive-fonts-recommended texlive-fonts-extra texlive-bibtex-extra
- Generate LaTeX Report:
make report
Referenced from Bibliography_management_with_bibtex