stable-diffusion-webui/modules/sd_vae_approx.py

62 lines
2 KiB
Python
Raw Normal View History

2022-12-24 19:39:00 +00:00
import os
import torch
from torch import nn
from modules import devices, paths
sd_vae_approx_model = None
class VAEApprox(nn.Module):
def __init__(self):
super(VAEApprox, self).__init__()
self.conv1 = nn.Conv2d(4, 8, (7, 7))
self.conv2 = nn.Conv2d(8, 16, (5, 5))
self.conv3 = nn.Conv2d(16, 32, (3, 3))
self.conv4 = nn.Conv2d(32, 64, (3, 3))
self.conv5 = nn.Conv2d(64, 32, (3, 3))
self.conv6 = nn.Conv2d(32, 16, (3, 3))
self.conv7 = nn.Conv2d(16, 8, (3, 3))
self.conv8 = nn.Conv2d(8, 3, (3, 3))
def forward(self, x):
extra = 11
x = nn.functional.interpolate(x, (x.shape[2] * 2, x.shape[3] * 2))
x = nn.functional.pad(x, (extra, extra, extra, extra))
for layer in [self.conv1, self.conv2, self.conv3, self.conv4, self.conv5, self.conv6, self.conv7, self.conv8, ]:
x = layer(x)
x = nn.functional.leaky_relu(x, 0.1)
return x
def model():
global sd_vae_approx_model
if sd_vae_approx_model is None:
model_path = os.path.join(paths.models_path, "VAE-approx", "model.pt")
2022-12-24 19:39:00 +00:00
sd_vae_approx_model = VAEApprox()
if not os.path.exists(model_path):
model_path = os.path.join(paths.script_path, "models", "VAE-approx", "model.pt")
sd_vae_approx_model.load_state_dict(torch.load(model_path, map_location='cpu' if devices.device.type != 'cuda' else None))
2022-12-24 19:39:00 +00:00
sd_vae_approx_model.eval()
sd_vae_approx_model.to(devices.device, devices.dtype)
return sd_vae_approx_model
def cheap_approximation(sample):
# https://discuss.huggingface.co/t/decoding-latents-to-rgb-without-upscaling/23204/2
coefs = torch.tensor([
[0.298, 0.207, 0.208],
[0.187, 0.286, 0.173],
[-0.158, 0.189, 0.264],
[-0.184, -0.271, -0.473],
]).to(sample.device)
x_sample = torch.einsum("lxy,lr -> rxy", sample, coefs)
return x_sample