stable-diffusion-webui/modules/esrgan_model.py

161 lines
6 KiB
Python
Raw Normal View History

2022-09-04 15:54:12 +00:00
import os
import numpy as np
import torch
from PIL import Image
2022-09-26 14:29:50 +00:00
from basicsr.utils.download_util import load_file_from_url
2022-09-04 15:54:12 +00:00
import modules.esrgam_model_arch as arch
from modules import shared, modelloader, images, devices
2022-09-26 14:29:50 +00:00
from modules.paths import models_path
from modules.upscaler import Upscaler, UpscalerData
2022-09-26 14:29:50 +00:00
from modules.shared import opts
2022-09-04 15:54:12 +00:00
def fix_model_layers(crt_model, pretrained_net):
# this code is adapted from https://github.com/xinntao/ESRGAN
if 'conv_first.weight' in pretrained_net:
return pretrained_net
if 'model.0.weight' not in pretrained_net:
is_realesrgan = "params_ema" in pretrained_net and 'body.0.rdb1.conv1.weight' in pretrained_net["params_ema"]
if is_realesrgan:
raise Exception("The file is a RealESRGAN model, it can't be used as a ESRGAN model.")
else:
raise Exception("The file is not a ESRGAN model.")
crt_net = crt_model.state_dict()
load_net_clean = {}
for k, v in pretrained_net.items():
if k.startswith('module.'):
load_net_clean[k[7:]] = v
else:
load_net_clean[k] = v
pretrained_net = load_net_clean
tbd = []
for k, v in crt_net.items():
tbd.append(k)
# directly copy
for k, v in crt_net.items():
if k in pretrained_net and pretrained_net[k].size() == v.size():
crt_net[k] = pretrained_net[k]
tbd.remove(k)
crt_net['conv_first.weight'] = pretrained_net['model.0.weight']
crt_net['conv_first.bias'] = pretrained_net['model.0.bias']
for k in tbd.copy():
if 'RDB' in k:
ori_k = k.replace('RRDB_trunk.', 'model.1.sub.')
if '.weight' in k:
ori_k = ori_k.replace('.weight', '.0.weight')
elif '.bias' in k:
ori_k = ori_k.replace('.bias', '.0.bias')
crt_net[k] = pretrained_net[ori_k]
tbd.remove(k)
crt_net['trunk_conv.weight'] = pretrained_net['model.1.sub.23.weight']
crt_net['trunk_conv.bias'] = pretrained_net['model.1.sub.23.bias']
crt_net['upconv1.weight'] = pretrained_net['model.3.weight']
crt_net['upconv1.bias'] = pretrained_net['model.3.bias']
crt_net['upconv2.weight'] = pretrained_net['model.6.weight']
crt_net['upconv2.bias'] = pretrained_net['model.6.bias']
crt_net['HRconv.weight'] = pretrained_net['model.8.weight']
crt_net['HRconv.bias'] = pretrained_net['model.8.bias']
crt_net['conv_last.weight'] = pretrained_net['model.10.weight']
crt_net['conv_last.bias'] = pretrained_net['model.10.bias']
return crt_net
class UpscalerESRGAN(Upscaler):
def __init__(self, dirname):
self.name = "ESRGAN"
self.model_url = "https://github.com/cszn/KAIR/releases/download/v1.0/ESRGAN.pth"
self.model_name = "ESRGAN_4x"
self.scalers = []
self.user_path = dirname
self.model_path = os.path.join(models_path, self.name)
super().__init__()
model_paths = self.find_models(ext_filter=[".pt", ".pth"])
scalers = []
if len(model_paths) == 0:
scaler_data = UpscalerData(self.model_name, self.model_url, self, 4)
scalers.append(scaler_data)
for file in model_paths:
if "http" in file:
name = self.model_name
else:
name = modelloader.friendly_name(file)
scaler_data = UpscalerData(name, file, self, 4)
self.scalers.append(scaler_data)
def do_upscale(self, img, selected_model):
model = self.load_model(selected_model)
if model is None:
return img
model.to(devices.device_esrgan)
img = esrgan_upscale(model, img)
return img
def load_model(self, path: str):
if "http" in path:
filename = load_file_from_url(url=self.model_url, model_dir=self.model_path,
file_name="%s.pth" % self.model_name,
progress=True)
2022-09-04 15:54:12 +00:00
else:
filename = path
if not os.path.exists(filename) or filename is None:
print("Unable to load %s from %s" % (self.model_path, filename))
return None
pretrained_net = torch.load(filename, map_location='cpu' if shared.device.type == 'mps' else None)
crt_model = arch.RRDBNet(3, 3, 64, 23, gc=32)
pretrained_net = fix_model_layers(crt_model, pretrained_net)
crt_model.load_state_dict(pretrained_net)
crt_model.eval()
return crt_model
2022-09-04 15:54:12 +00:00
def upscale_without_tiling(model, img):
img = np.array(img)
img = img[:, :, ::-1]
img = np.moveaxis(img, 2, 0) / 255
img = torch.from_numpy(img).float()
img = img.unsqueeze(0).to(devices.device_esrgan)
2022-09-04 15:54:12 +00:00
with torch.no_grad():
output = model(img)
output = output.squeeze().float().cpu().clamp_(0, 1).numpy()
output = 255. * np.moveaxis(output, 0, 2)
output = output.astype(np.uint8)
output = output[:, :, ::-1]
return Image.fromarray(output, 'RGB')
def esrgan_upscale(model, img):
if opts.ESRGAN_tile == 0:
2022-09-04 15:54:12 +00:00
return upscale_without_tiling(model, img)
grid = images.split_grid(img, opts.ESRGAN_tile, opts.ESRGAN_tile, opts.ESRGAN_tile_overlap)
2022-09-04 15:54:12 +00:00
newtiles = []
scale_factor = 1
for y, h, row in grid.tiles:
newrow = []
for tiledata in row:
x, w, tile = tiledata
output = upscale_without_tiling(model, tile)
scale_factor = output.width // tile.width
newrow.append([x * scale_factor, w * scale_factor, output])
newtiles.append([y * scale_factor, h * scale_factor, newrow])
newgrid = images.Grid(newtiles, grid.tile_w * scale_factor, grid.tile_h * scale_factor, grid.image_w * scale_factor, grid.image_h * scale_factor, grid.overlap * scale_factor)
output = images.combine_grid(newgrid)
2022-09-04 15:54:12 +00:00
return output