diff options
-rw-r--r-- | extensions-builtin/ScuNET/scripts/scunet_model.py | 55 | ||||
-rw-r--r-- | extensions-builtin/SwinIR/scripts/swinir_model.py | 63 | ||||
-rw-r--r-- | modules/gfpgan_model.py | 10 | ||||
-rw-r--r-- | modules/modelloader.py | 25 | ||||
-rw-r--r-- | modules/realesrgan_model.py | 4 | ||||
-rw-r--r-- | modules/upscaler_utils.py | 74 | ||||
-rw-r--r-- | requirements_versions.txt | 2 |
7 files changed, 114 insertions, 119 deletions
diff --git a/extensions-builtin/ScuNET/scripts/scunet_model.py b/extensions-builtin/ScuNET/scripts/scunet_model.py index 5f3dd08b..f799cb76 100644 --- a/extensions-builtin/ScuNET/scripts/scunet_model.py +++ b/extensions-builtin/ScuNET/scripts/scunet_model.py @@ -3,12 +3,11 @@ import sys import PIL.Image import numpy as np import torch -from tqdm import tqdm import modules.upscaler from modules import devices, modelloader, script_callbacks, errors - from modules.shared import opts +from modules.upscaler_utils import tiled_upscale_2 class UpscalerScuNET(modules.upscaler.Upscaler): @@ -40,47 +39,6 @@ class UpscalerScuNET(modules.upscaler.Upscaler): scalers.append(scaler_data2) self.scalers = scalers - @staticmethod - @torch.no_grad() - def tiled_inference(img, model): - # test the image tile by tile - h, w = img.shape[2:] - tile = opts.SCUNET_tile - tile_overlap = opts.SCUNET_tile_overlap - if tile == 0: - return model(img) - - device = devices.get_device_for('scunet') - assert tile % 8 == 0, "tile size should be a multiple of window_size" - sf = 1 - - stride = tile - tile_overlap - h_idx_list = list(range(0, h - tile, stride)) + [h - tile] - w_idx_list = list(range(0, w - tile, stride)) + [w - tile] - E = torch.zeros(1, 3, h * sf, w * sf, dtype=img.dtype, device=device) - W = torch.zeros_like(E, dtype=devices.dtype, device=device) - - with tqdm(total=len(h_idx_list) * len(w_idx_list), desc="ScuNET tiles") as pbar: - for h_idx in h_idx_list: - - for w_idx in w_idx_list: - - in_patch = img[..., h_idx: h_idx + tile, w_idx: w_idx + tile] - - out_patch = model(in_patch) - out_patch_mask = torch.ones_like(out_patch) - - E[ - ..., h_idx * sf: (h_idx + tile) * sf, w_idx * sf: (w_idx + tile) * sf - ].add_(out_patch) - W[ - ..., h_idx * sf: (h_idx + tile) * sf, w_idx * sf: (w_idx + tile) * sf - ].add_(out_patch_mask) - pbar.update(1) - output = E.div_(W) - - return output - def do_upscale(self, img: PIL.Image.Image, selected_file): devices.torch_gc() @@ -104,7 +62,16 @@ class UpscalerScuNET(modules.upscaler.Upscaler): _img[:, :, :h, :w] = torch_img # pad image torch_img = _img - torch_output = self.tiled_inference(torch_img, model).squeeze(0) + with torch.no_grad(): + torch_output = tiled_upscale_2( + torch_img, + model, + tile_size=opts.SCUNET_tile, + tile_overlap=opts.SCUNET_tile_overlap, + scale=1, + device=devices.get_device_for('scunet'), + desc="ScuNET tiles", + ).squeeze(0) torch_output = torch_output[:, :h * 1, :w * 1] # remove padding, if any np_output: np.ndarray = torch_output.float().cpu().clamp_(0, 1).numpy() del torch_img, torch_output diff --git a/extensions-builtin/SwinIR/scripts/swinir_model.py b/extensions-builtin/SwinIR/scripts/swinir_model.py index aae159af..8a555c79 100644 --- a/extensions-builtin/SwinIR/scripts/swinir_model.py +++ b/extensions-builtin/SwinIR/scripts/swinir_model.py @@ -4,11 +4,11 @@ import sys import numpy as np import torch from PIL import Image -from tqdm import tqdm from modules import modelloader, devices, script_callbacks, shared -from modules.shared import opts, state +from modules.shared import opts from modules.upscaler import Upscaler, UpscalerData +from modules.upscaler_utils import tiled_upscale_2 SWINIR_MODEL_URL = "https://github.com/JingyunLiang/SwinIR/releases/download/v0.0/003_realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN.pth" @@ -71,7 +71,7 @@ class UpscalerSwinIR(Upscaler): else: filename = path - model = modelloader.load_spandrel_model( + model_descriptor = modelloader.load_spandrel_model( filename, device=self._get_device(), dtype=devices.dtype, @@ -79,10 +79,10 @@ class UpscalerSwinIR(Upscaler): ) if getattr(opts, 'SWIN_torch_compile', False): try: - model = torch.compile(model) + model_descriptor.model.compile() except Exception: logger.warning("Failed to compile SwinIR model, fallback to JIT", exc_info=True) - return model + return model_descriptor def _get_device(self): return devices.get_device_for('swinir') @@ -110,14 +110,14 @@ def upscale( w_pad = (w_old // window_size + 1) * window_size - w_old img = torch.cat([img, torch.flip(img, [2])], 2)[:, :, : h_old + h_pad, :] img = torch.cat([img, torch.flip(img, [3])], 3)[:, :, :, : w_old + w_pad] - output = inference( + output = tiled_upscale_2( img, model, - tile=tile, + tile_size=tile, tile_overlap=tile_overlap, - window_size=window_size, scale=scale, device=device, + desc="SwinIR tiles", ) output = output[..., : h_old * scale, : w_old * scale] output = output.data.squeeze().float().cpu().clamp_(0, 1).numpy() @@ -129,53 +129,6 @@ def upscale( return Image.fromarray(output, "RGB") -def inference( - img, - model, - *, - tile: int, - tile_overlap: int, - window_size: int, - scale: int, - device, -): - # test the image tile by tile - b, c, h, w = img.size() - tile = min(tile, h, w) - assert tile % window_size == 0, "tile size should be a multiple of window_size" - sf = scale - - stride = tile - tile_overlap - h_idx_list = list(range(0, h - tile, stride)) + [h - tile] - w_idx_list = list(range(0, w - tile, stride)) + [w - tile] - E = torch.zeros(b, c, h * sf, w * sf, dtype=devices.dtype, device=device).type_as(img) - W = torch.zeros_like(E, dtype=devices.dtype, device=device) - - with tqdm(total=len(h_idx_list) * len(w_idx_list), desc="SwinIR tiles") as pbar: - for h_idx in h_idx_list: - if state.interrupted or state.skipped: - break - - for w_idx in w_idx_list: - if state.interrupted or state.skipped: - break - - in_patch = img[..., h_idx: h_idx + tile, w_idx: w_idx + tile] - out_patch = model(in_patch) - out_patch_mask = torch.ones_like(out_patch) - - E[ - ..., h_idx * sf: (h_idx + tile) * sf, w_idx * sf: (w_idx + tile) * sf - ].add_(out_patch) - W[ - ..., h_idx * sf: (h_idx + tile) * sf, w_idx * sf: (w_idx + tile) * sf - ].add_(out_patch_mask) - pbar.update(1) - output = E.div_(W) - - return output - - def on_ui_settings(): import gradio as gr diff --git a/modules/gfpgan_model.py b/modules/gfpgan_model.py index 48f8ad5e..445b0409 100644 --- a/modules/gfpgan_model.py +++ b/modules/gfpgan_model.py @@ -3,6 +3,8 @@ from __future__ import annotations import logging
import os
+import torch
+
from modules import (
devices,
errors,
@@ -25,7 +27,7 @@ class FaceRestorerGFPGAN(face_restoration_utils.CommonFaceRestoration): def get_device(self):
return devices.device_gfpgan
- def load_net(self) -> None:
+ def load_net(self) -> torch.Module:
for model_path in modelloader.load_models(
model_path=self.model_path,
model_url=model_url,
@@ -34,13 +36,13 @@ class FaceRestorerGFPGAN(face_restoration_utils.CommonFaceRestoration): ext_filter=['.pth'],
):
if 'GFPGAN' in os.path.basename(model_path):
- net = modelloader.load_spandrel_model(
+ model = modelloader.load_spandrel_model(
model_path,
device=self.get_device(),
expected_architecture='GFPGAN',
).model
- net.different_w = True # see https://github.com/chaiNNer-org/spandrel/pull/81
- return net
+ model.different_w = True # see https://github.com/chaiNNer-org/spandrel/pull/81
+ return model
raise ValueError("No GFPGAN model found")
def restore(self, np_image):
diff --git a/modules/modelloader.py b/modules/modelloader.py index 0b89d682..a7194137 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -1,8 +1,9 @@ from __future__ import annotations +import importlib import logging import os -import importlib +from typing import TYPE_CHECKING from urllib.parse import urlparse import torch @@ -10,6 +11,8 @@ import torch from modules import shared from modules.upscaler import Upscaler, UpscalerLanczos, UpscalerNearest, UpscalerNone +if TYPE_CHECKING: + import spandrel logger = logging.getLogger(__name__) @@ -140,19 +143,19 @@ def load_spandrel_model( *, device: str | torch.device | None, half: bool = False, - dtype: str | None = None, + dtype: str | torch.dtype | None = None, expected_architecture: str | None = None, -): +) -> spandrel.ModelDescriptor: import spandrel - model = spandrel.ModelLoader(device=device).load_from_file(path) - if expected_architecture and model.architecture != expected_architecture: + model_descriptor = spandrel.ModelLoader(device=device).load_from_file(path) + if expected_architecture and model_descriptor.architecture != expected_architecture: logger.warning( - f"Model {path!r} is not a {expected_architecture!r} model (got {model.architecture!r})", + f"Model {path!r} is not a {expected_architecture!r} model (got {model_descriptor.architecture!r})", ) if half: - model = model.model.half() + model_descriptor.model.half() if dtype: - model = model.model.to(dtype=dtype) - model.eval() - logger.debug("Loaded %s from %s (device=%s, half=%s, dtype=%s)", model, path, device, half, dtype) - return model + model_descriptor.model.to(dtype=dtype) + model_descriptor.model.eval() + logger.debug("Loaded %s from %s (device=%s, half=%s, dtype=%s)", model_descriptor, path, device, half, dtype) + return model_descriptor diff --git a/modules/realesrgan_model.py b/modules/realesrgan_model.py index 65f2e880..4d35b695 100644 --- a/modules/realesrgan_model.py +++ b/modules/realesrgan_model.py @@ -36,14 +36,14 @@ class UpscalerRealESRGAN(Upscaler): errors.report(f"Unable to load RealESRGAN model {path}", exc_info=True)
return img
- mod = modelloader.load_spandrel_model(
+ model_descriptor = modelloader.load_spandrel_model(
info.local_data_path,
device=self.device,
half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling),
expected_architecture="ESRGAN", # "RealESRGAN" isn't a specific thing for Spandrel
)
return upscale_with_model(
- mod,
+ model_descriptor,
img,
tile_size=opts.ESRGAN_tile,
tile_overlap=opts.ESRGAN_tile_overlap,
diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 39f78a0b..8e413854 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -6,7 +6,7 @@ import torch import tqdm from PIL import Image -from modules import devices, images +from modules import images, shared logger = logging.getLogger(__name__) @@ -17,7 +17,7 @@ def upscale_without_tiling(model, img: Image.Image): img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 img = torch.from_numpy(img).float() - model_weight = next(iter(model.parameters())) + model_weight = next(iter(model.model.parameters())) img = img.unsqueeze(0).to(device=model_weight.device, dtype=model_weight.dtype) with torch.no_grad(): @@ -68,3 +68,73 @@ def upscale_with_model( overlap=grid.overlap * scale_factor, ) return images.combine_grid(newgrid) + + +def tiled_upscale_2( + img, + model, + *, + tile_size: int, + tile_overlap: int, + scale: int, + device, + desc="Tiled upscale", +): + # Alternative implementation of `upscale_with_model` originally used by + # SwinIR and ScuNET. It differs from `upscale_with_model` in that tiling and + # weighting is done in PyTorch space, as opposed to `images.Grid` doing it in + # Pillow space without weighting. + b, c, h, w = img.size() + tile_size = min(tile_size, h, w) + + if tile_size <= 0: + logger.debug("Upscaling %s without tiling", img.shape) + return model(img) + + stride = tile_size - tile_overlap + h_idx_list = list(range(0, h - tile_size, stride)) + [h - tile_size] + w_idx_list = list(range(0, w - tile_size, stride)) + [w - tile_size] + result = torch.zeros( + b, + c, + h * scale, + w * scale, + device=device, + ).type_as(img) + weights = torch.zeros_like(result) + logger.debug("Upscaling %s to %s with tiles", img.shape, result.shape) + with tqdm.tqdm(total=len(h_idx_list) * len(w_idx_list), desc=desc) as pbar: + for h_idx in h_idx_list: + if shared.state.interrupted or shared.state.skipped: + break + + for w_idx in w_idx_list: + if shared.state.interrupted or shared.state.skipped: + break + + in_patch = img[ + ..., + h_idx : h_idx + tile_size, + w_idx : w_idx + tile_size, + ] + out_patch = model(in_patch) + + result[ + ..., + h_idx * scale : (h_idx + tile_size) * scale, + w_idx * scale : (w_idx + tile_size) * scale, + ].add_(out_patch) + + out_patch_mask = torch.ones_like(out_patch) + + weights[ + ..., + h_idx * scale : (h_idx + tile_size) * scale, + w_idx * scale : (w_idx + tile_size) * scale, + ].add_(out_patch_mask) + + pbar.update(1) + + output = result.div_(weights) + + return output diff --git a/requirements_versions.txt b/requirements_versions.txt index 1e0ccafa..2a922f28 100644 --- a/requirements_versions.txt +++ b/requirements_versions.txt @@ -12,7 +12,7 @@ inflection==0.5.1 jsonmerge==1.8.0
kornia==0.6.7
lark==1.1.2
-numpy==1.23.5
+numpy==1.26.2
omegaconf==2.2.3
open-clip-torch==2.20.0
piexif==1.1.3
|