Denoising Diffusion Implicit Models (DDIM)

DDIM Implementation: Jiaming Song, Chenlin Meng and Stefano Ermon, Stanford

Modified into Conditional DDIM by Hengtong.

The code has been tested on PyTorch 1.12.

Training is exactly the same as DDPM with the following:

python main.py --config {DATASET}.yml --exp {PROJECT_PATH} --doc {MODEL_NAME} --ni --cond

the new added --cond is the option for conditional ddim.

python main.py --config {DATASET}.yml --exp {PROJECT_PATH} --doc {MODEL_NAME} --sample --fid --timesteps {STEPS} --eta {ETA} --ni

where

• ETA controls the scale of the variance (0 is DDIM, and 1 is one type of DDPM).
• STEPS controls how many timesteps used in the process.
• MODEL_NAME finds the pre-trained checkpoint according to its inferred path.

Use --sequence option instead.

The above two cases contain some hard-coded lines specific to producing the image, so modify them according to your needs.

This implementation is based on / inspired by:

• https://github.com/ermongroup/ddim (the DDIM Pytorch repo),