This is a project about Condition-based Latent Diffusion Model, by using the deep learning method can learn the structure of some particles from the noise. It can recover the structure according to a specific function Pair Distribution Function.

The project holds the model files for CbLDM as well as the data used for training. Specifically, it includes files such as submodel.zip, data_train.7z, model_val_loss=0.13734.ckpt.

The CbLDM training code and other related files are available at https://github.com/hegwj/CbLDM