• Python: 3.6

pip install -r requirements.txt

python main.py --data_dir data --dataset_name L1 --result_dir results

• data_dir contains raw as subdir that contains fasta files.
• After finish running, results will be save in result_dir/log and result_dir/model.
• If data files are placed as below tree, running is simple: python main.py

.
├── README.md
├── config
│   └── dataset_metadata.json
├── data
│   ├── processed
│   └── raw
│       ├── L1.fna
│       ├── L2.fna
│       ├── L3.fna
├── main.py

The deep clustering algorithm (in adec.py) is based on paper Adversarial Deep Embedded Clustering: on a better trade-off between Feature Randomness and Feature Drift