# CDGS

**Repository Path**: cool_c0de/CDGS

## Basic Information

- **Project Name**: CDGS
- **Description**: from https://github.com/GRAPH-0/CDGS 20251111
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-11-11
- **Last Updated**: 2025-11-11

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# CDGS

[Conditional Diffusion Based on Discrete Graph Structures for Molecular Graph Generation](https://arxiv.org/abs/2301.00427) - AAAI 2023

The extension version: Learning Joint 2D & 3D Diffusion Models for
Complete Molecule Generation [[Paper]](https://arxiv.org/abs/2305.12347)  [[Code]](https://github.com/GRAPH-0/JODO).

## Dependencies

* pytorch 1.11
* PyG 2.1

For NSPDK evaluation:

```pip install git+https://github.com/fabriziocosta/EDeN.git --user```

Others see requirements.txt .


## Training 

### QM9

```shell
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_qm9_cdgs.py --mode train --workdir exp/vpsde_qm9_cdgs
``` 

* Set GPU id via `CUDA_VISIBLE_DEVICES`.
* `workdir` is the directory path to save checkpoints, which can be changed to `YOUR_PATH`. We provide the pretrained checkpoint in `exp/vpsde_qm9_cdgs`.
* More hyperparameters in the config file `configs/vp_qm9_cdgs.py`

### ZINC250k

```shell
# 256 hidden dimension
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_zinc_cdgs.py --mode train --workdir exp/vpsde_zinc_cdgs_256 --config.training.n_iters 2500000

# 128 hidden dimension
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_zinc_cdgs.py --mode train --workdir exp/vpsde_zinc_cdgs_128 --config.training.batch_size 128 --config.training.eval_batch_size 128 --config.training.n_iters 2500000
```

The pretrained checkpoints are provided in [Google Drive 256ch](https://drive.google.com/drive/folders/1bA_6ldtwF6gTMToZGG7w1dqmOpIuWZJd?usp=sharing) 
and [Google Drive 128ch](https://drive.google.com/drive/folders/1WRKkqJyJMue_evkqULRSftyHvRPqir6m?usp=share_link). 

## Sampling

### QM9

1. EM sampling with 1000 steps

```shell
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_qm9_cdgs.py --mode eval --workdir exp/vpsde_qm9_cdgs --config.eval.begin_ckpt 200 --config.eval.end_ckpt 200
```

* Add `--config.eval.nspdk` if apply NSPDK evaluation.
* Change iteration steps through `--config.model.num_scales YOUR_STEPS`.
* Change sampling batch size `--config.eval.batch_size` to control GPU memory usage.

2. DPM-Solver examples

```shell
# Order 3; 50 step
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_qm9_cdgs.py --mode eval --workdir exp/vpsde_qm9_cdgs --config.eval.begin_ckpt 200 --config.eval.end_ckpt 200 --config.sampling.method dpm3 --config.sampling.ode_step 50

# Order 2; 20 step
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_qm9_cdgs.py --mode eval --workdir exp/vpsde_qm9_cdgs --config.eval.begin_ckpt 200 --config.eval.end_ckpt 200 --config.sampling.method dpm2 --config.sampling.ode_step 20

# Order 1; 10 step
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_qm9_cdgs.py --mode eval --workdir exp/vpsde_qm9_cdgs --config.eval.begin_ckpt 200 --config.eval.end_ckpt 200 --config.sampling.method dpm1 --config.sampling.ode_step 10
```

### ZINC250k

1. EM sampling examples

```shell
# 1000 steps
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_zinc_cdgs.py --mode eval --workdir exp/vpsde_zinc_cdgs_256 --config.eval.begin_ckpt 250 --config.eval.end_ckpt 250 --config.eval.batch_size 800

# 200 steps
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_zinc_cdgs.py --mode eval --workdir exp/vpsde_zinc_cdgs_256 --config.eval.begin_ckpt 250 --config.eval.end_ckpt 250 --config.eval.batch_size 800 --config.model.num_scales 200
```

2. DPM-Solver examples

```shell
# Order 3; 50 step
CUDA_VISIBLE_DEVICES=0 python main.py --config configs/vp_zinc_cdgs.py --mode eval --workdir exp/vpsde_zinc_cdgs_256 --config.eval.begin_ckpt 250 --config.eval.end_ckpt 250 --config.eval.batch_size 800 --config.sampling.method dpm3 --config.sampling.ode_step 50
```

### Results 
We provide molecules generated by CDGS: [Google Drive](https://drive.google.com/drive/folders/1eafc2bMETEyUVXvD9fW5Fc72v70rvPzH?usp=share_link). 

## Citation

```bibtex
@article{huang2023conditional,
  title={Conditional Diffusion Based on Discrete Graph Structures for Molecular Graph Generation},
  author={Huang, Han and Sun, Leilei and Du, Bowen and Lv, Weifeng},
  journal={arXiv preprint arXiv:2301.00427},
  year={2023}
}
```