Installation

Create and activate environment

conda create -n detanet_env python=3.10 -y

conda activate detanet_env

torch installation

conda install pytorch==2.4.1 torchvision==0.19.1 torchaudio==2.4.1 pytorch-cuda=12.4 -c pytorch -c nvidia

install torch-geometric

pip install torch_geometric

pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.4.0+cu124.html

install dependencies

pip install nequip wandb pandas e3nn rdkit pot scikit-learn ipykernel

---

---

Used Versions

nequip==0.15.0 wandb==0.22.2 pandas==2.3.3 e3nn==0.5.8 rdkit==2025.9.1 pot==0.9.6.post1 scikit-learn==1.7.2 ipykernel==6.30.1

Code Derived from

DetaNet

Cite:

DetaNet:

Zou, Zihan & Zhang, Yujin & Liang, Lijun & Wei, Mingzhi & Leng, Jiancai & Jiang, Jun & Luo, Yi & Hu, Wei. (2023). A deep learning model for predicting selected organic molecular spectra. Nature Computational Science. 3. 1-8. 10.1038/s43588-023-00550-y.