Diffusion-based-Fluid-Super-resolution

Diffusion-based-Fluid-Super-resolution copied to clipboard

kmflow_re1000_rs256.yml doesn't work

3

Hi! I started directly from step 2 and ran `python main.py --config kmflow_re1000_rs256.yml --seed 1234 --sample_step 1 --t 240 --r 30` in the main directory of this repo, but got...

whw-alex

In `train_ddpm/runners/diffusion_tub.py` the loss is computed as `loss = loss_registry[config.model.type](model, x, t, e, b, x_offset.item(), x_scale.item())`. This calls `def conditional_noise_estimation_loss(model, x0: torch.Tensor, t: torch.LongTensor, e: torch.Tensor, b: torch.Tensor, x_scale, x_offset,...

jhbastek

Hi, I tried to use your code, but it seems that Py torch 1.7 is not the rights version since fft2 didn't exist in the mentioned version and you used...

moeindarman77

Hi, Maybe the physical residual (computed by the governing vorticity equation) should be calculated by the denoised one rather than the noised one. ` x = x0 * a.sqrt() +...

NightGlow0826