298977887/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2025-03-14 21:47:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Don't use numpy for calculating sigmas.

Browse Source
This commit is contained in:
comfyanonymous 2024-02-07 18:52:51 -05:00
parent 7daad468ec
commit c661a8b118
2 changed files with 5 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
comfy/ldm/modules/diffusionmodules/util.py
View File

@ -98,7 +98,7 @@ def make_beta_schedule(schedule, n_timestep, linear_start=1e-4, linear_end=2e-2,
alphas = torch.cos(alphas).pow(2)
alphas = alphas / alphas[0]
betas = 1 - alphas[1:] / alphas[:-1]
betas = np.clip(betas, a_min=0, a_max=0.999)
betas = torch.clamp(betas, min=0, max=0.999)
elif schedule == "squaredcos_cap_v2": # used for karlo prior
# return early
@ -113,7 +113,7 @@ def make_beta_schedule(schedule, n_timestep, linear_start=1e-4, linear_end=2e-2,
betas = torch.linspace(linear_start, linear_end, n_timestep, dtype=torch.float64) ** 0.5
else:
raise ValueError(f"schedule '{schedule}' unknown.")
return betas.numpy()
return betas
def make_ddim_timesteps(ddim_discr_method, num_ddim_timesteps, num_ddpm_timesteps, verbose=True):

8
comfy/model_sampling.py
View File

@ -1,5 +1,4 @@
import torch
import numpy as np
from comfy.ldm.modules.diffusionmodules.util import make_beta_schedule
import math
@ -42,8 +41,7 @@ class ModelSamplingDiscrete(torch.nn.Module):
else:
betas = make_beta_schedule(beta_schedule, timesteps, linear_start=linear_start, linear_end=linear_end, cosine_s=cosine_s)
alphas = 1. - betas
alphas_cumprod = torch.tensor(np.cumprod(alphas, axis=0), dtype=torch.float32)
# alphas_cumprod_prev = np.append(1., alphas_cumprod[:-1])
alphas_cumprod = torch.cumprod(alphas, dim=0)
timesteps, = betas.shape
self.num_timesteps = int(timesteps)
@ -58,8 +56,8 @@ class ModelSamplingDiscrete(torch.nn.Module):
self.set_sigmas(sigmas)
def set_sigmas(self, sigmas):
self.register_buffer('sigmas', sigmas)
self.register_buffer('log_sigmas', sigmas.log())
self.register_buffer('sigmas', sigmas.float())
self.register_buffer('log_sigmas', sigmas.log().float())
@property
def sigma_min(self):