Hugging Face's logo

diffusers
/
ddpm-cifar10-32-demo

Unconditional Image Generation Diffusers PyTorch DDPMPipeline
Model card Files Community
ddpm-cifar10-32-demo / modeling_ddpm.py
patrickvonplaten's picture
patrickvonplaten
Duplicate from google/ddpm-cifar10-32
96eb466
raw history blame contribute delete
No virus
2.79 kB
# Copyright 2022 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from diffusers import DiffusionPipeline
import tqdm
import torch
class DDPM(DiffusionPipeline):
modeling_file = "modeling_ddpm.py"
def __init__(self, unet, noise_scheduler):
super().__init__()
self.register_modules(unet=unet, noise_scheduler=noise_scheduler)
def __call__(self, generator=None, torch_device=None):
torch_device = "cuda" if torch.cuda.is_available() else "cpu"
self.unet.to(torch_device)
# 1. Sample gaussian noise
image = self.noise_scheduler.sample_noise((1, self.unet.in_channels, self.unet.resolution, self.unet.resolution), device=torch_device, generator=generator)
for t in tqdm.tqdm(reversed(range(len(self.noise_scheduler))), total=len(self.noise_scheduler)):
# i) define coefficients for time step t
clip_image_coeff = 1 / torch.sqrt(self.noise_scheduler.get_alpha_prod(t))
clip_noise_coeff = torch.sqrt(1 / self.noise_scheduler.get_alpha_prod(t) - 1)
image_coeff = (1 - self.noise_scheduler.get_alpha_prod(t - 1)) * torch.sqrt(self.noise_scheduler.get_alpha(t)) / (1 - self.noise_scheduler.get_alpha_prod(t))
clip_coeff = torch.sqrt(self.noise_scheduler.get_alpha_prod(t - 1)) * self.noise_scheduler.get_beta(t) / (1 - self.noise_scheduler.get_alpha_prod(t))
# ii) predict noise residual
with torch.no_grad():
noise_residual = self.unet(image, t)
# iii) compute predicted image from residual
# See 2nd formula at https://github.com/hojonathanho/diffusion/issues/5#issue-896554416 for comparison
pred_mean = clip_image_coeff * image - clip_noise_coeff * noise_residual
pred_mean = torch.clamp(pred_mean, -1, 1)
prev_image = clip_coeff * pred_mean + image_coeff * image
# iv) sample variance
prev_variance = self.noise_scheduler.sample_variance(t, prev_image.shape, device=torch_device, generator=generator)
# v) sample x_{t-1} ~ N(prev_image, prev_variance)
sampled_prev_image = prev_image + prev_variance
image = sampled_prev_image
return image