tgm dlm
1.0.0

這是AAAI 2024論文的代碼:具有擴散語言模型的文本引導分子生成。
cd TGMDLMCODE; pip install -e improved-diffusion/; pip install -e transformers/ 。scibert中。cd improved-diffusion; cd scriptspython process_text.py -i train_val_256; python process_text.py -i testpython train.pypython train_correct_withmask.py以下細節對於您實際訓練此模型之前就重要了!
python text_sample.py; python post_sample.py最終文件OURMODEL_OUTPUT.txt是我們的輸出。您可以通過運行ev.py來評估除text2mol以外的所有指標。有關Text2mol,請轉到Molt5以獲取更多詳細信息。
如果您使用代碼,請引用我們的論文:
@article{gong2024text,
title={Text-Guided Molecule Generation with Diffusion Language Model},
author={Gong, Haisong and Liu, Qiang and Wu, Shu and Wang, Liang},
volume={38},
url={https://ojs.aaai.org/index.php/AAAI/article/view/27761},
DOI={10.1609/aaai.v38i1.27761},
number={1},
journal={Proceedings of the AAAI Conference on Artificial Intelligence},
year={2024},
month={Mar.},
pages={109-117}
}
此代碼基於https://github.com/xiangli1999/diffusion-lm和https://github.com/blender-nlp/molt5