The current setting is to train two unets (one to estimate the residuals and one to estimate the noise), which can be used to explore partially path-independent generation process.
1. Load Pre-trained Models
Download the pre-trained models (two unets, deresidual+denoising) for partially path-independent generation process.
cd ./RDDM/experiments/0_Partially_path-independent_generation
cp model-100.pt ./results/sample/
Then, you should set the path of celeba dataset.
python test.py
2. Differences in code compared to other tasks: Other tasks need to modify
a) [self.alphas_cumsum[t]*self.num_timesteps, self.betas_cumsum[t]*self.num_timesteps]] -> [t,t] (in L852 and L1292).
b) For image restoration, generation=False in L120, convert_to_ddim=False in L640 and L726.
c) uncomment L726 for simultaneous removal of residuals and noise.
d) modify the corresponding experimental settings (see Table 4 in the Appendix).