The HLM is the method for using the language models to convert a helical filament into vectors. These filament vectors can be used for further processing.
python3
pytorch >= 1.7
numpy
umap-learn
scikit-learn
TO use the HLM-Bert, an additional hugging face environment needs to be installed from here: https://huggingface.co/docs/transformers/installation
running the jupyter note embedding.ipynb to get the embedding vector of each filament and separate them into different clusters using HLM_word2vec method, also, try to run HLM bert method, you can run the jupyter notebook BERT.ipynb
Or run:
$ python HLM.py --o your_output_directory --in_parts your_star_file
and use -h to check the possible command