add restart sampler

modules/sd_samplers_kdiffusion.py

@@ -1,3 +1,5 @@
+# export PIP_CACHE_DIR=/scratch/dengm/cache
+# export XDG_CACHE_HOME=/scratch/dengm/cache
 from collections import deque
 import torch
 import inspect
@@ -30,12 +32,76 @@ samplers_k_diffusion = [
     ('DPM++ 2M Karras', 'sample_dpmpp_2m', ['k_dpmpp_2m_ka'], {'scheduler': 'karras'}),
     ('DPM++ SDE Karras', 'sample_dpmpp_sde', ['k_dpmpp_sde_ka'], {'scheduler': 'karras', "second_order": True, "brownian_noise": True}),
     ('DPM++ 2M SDE Karras', 'sample_dpmpp_2m_sde', ['k_dpmpp_2m_sde_ka'], {'scheduler': 'karras', "brownian_noise": True}),
+    ('Restart (new)', 'restart_sampler', ['restart'], {'scheduler': 'karras', "second_order": True}),
 ]
 
+
+@torch.no_grad()
+def restart_sampler(model, x, sigmas, extra_args=None, callback=None, disable=None, s_noise=1., restart_list = {0.1: [10, 2, 2]}):
+    """Implements restart sampling in Restart Sampling for Improving Generative Processes (2023)"""
+    '''Restart_list format: {min_sigma: [ restart_steps, restart_times, max_sigma]}'''
+    
+    from tqdm.auto import trange, tqdm
+    extra_args = {} if extra_args is None else extra_args
+    s_in = x.new_ones([x.shape[0]])
+    step_id = 0
+    
+    from k_diffusion.sampling import to_d, append_zero
+
+    def heun_step(x, old_sigma, new_sigma):
+        nonlocal step_id
+        denoised = model(x, old_sigma * s_in, **extra_args)
+        d = to_d(x, old_sigma, denoised)
+        if callback is not None:
+            callback({'x': x, 'i': step_id, 'sigma': new_sigma, 'sigma_hat': old_sigma, 'denoised': denoised})
+        dt = new_sigma - old_sigma
+        if new_sigma == 0:
+            # Euler method
+            x = x + d * dt
+        else:
+            # Heun's method
+            x_2 = x + d * dt
+            denoised_2 = model(x_2, new_sigma * s_in, **extra_args)
+            d_2 = to_d(x_2, new_sigma, denoised_2)
+            d_prime = (d + d_2) / 2
+            x = x + d_prime * dt
+        step_id += 1
+        return x
+    # print(sigmas)
+    temp_list = dict()
+    for key, value in restart_list.items():
+        temp_list[int(torch.argmin(abs(sigmas - key), dim=0))] = value
+    restart_list = temp_list
+        
+            
+    def get_sigmas_karras(n, sigma_min, sigma_max, rho=7., device='cpu'):
+        ramp = torch.linspace(0, 1, n).to(device)
+        min_inv_rho = (sigma_min ** (1 / rho))
+        max_inv_rho = (sigma_max ** (1 / rho))
+        if isinstance(min_inv_rho, torch.Tensor):
+            min_inv_rho = min_inv_rho.to(device)
+        if isinstance(max_inv_rho, torch.Tensor):
+            max_inv_rho = max_inv_rho.to(device)
+        sigmas = (max_inv_rho + ramp * (min_inv_rho - max_inv_rho)) ** rho
+        return append_zero(sigmas).to(device)
+
+    for i in trange(len(sigmas) - 1, disable=disable):
+        x = heun_step(x, sigmas[i], sigmas[i+1])
+        if i + 1 in restart_list:
+            restart_steps, restart_times, restart_max = restart_list[i + 1]
+            min_idx = i + 1
+            max_idx = int(torch.argmin(abs(sigmas - restart_max), dim=0))
+            sigma_restart = get_sigmas_karras(restart_steps, sigmas[min_idx], sigmas[max_idx], device=sigmas.device)[:-1] # remove the zero at the end
+            for times in range(restart_times):
+                x = x + torch.randn_like(x) * s_noise * (sigmas[max_idx] ** 2 - sigmas[min_idx] ** 2) ** 0.5
+                for (old_sigma, new_sigma) in zip(sigma_restart[:-1], sigma_restart[1:]):
+                    x = heun_step(x, old_sigma, new_sigma)
+    return x
+
 samplers_data_k_diffusion = [
     sd_samplers_common.SamplerData(label, lambda model, funcname=funcname: KDiffusionSampler(funcname, model), aliases, options)
     for label, funcname, aliases, options in samplers_k_diffusion
-    if hasattr(k_diffusion.sampling, funcname)
+    if (hasattr(k_diffusion.sampling, funcname) or funcname == 'restart_sampler')
 ]
 
 sampler_extra_params = {
@@ -245,7 +311,7 @@ class KDiffusionSampler:
 
         self.model_wrap = denoiser(sd_model, quantize=shared.opts.enable_quantization)
         self.funcname = funcname
-        self.func = getattr(k_diffusion.sampling, self.funcname)
+        self.func = getattr(k_diffusion.sampling, self.funcname) if funcname != "restart_sampler" else restart_sampler
         self.extra_params = sampler_extra_params.get(funcname, [])
         self.model_wrap_cfg = CFGDenoiser(self.model_wrap)
         self.sampler_noises = None