nodes/yolain-ComfyUI-Easy-Use
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

add:differential diffusion for easy kSampelrInpainting

Browse Source
This commit is contained in:
yolain
2024-03-25 19:19:08 +08:00
parent 54ea3af9b8
commit 192181007a
4 changed files with 68 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.en.md
+6
View File
@@ -31,6 +31,12 @@
## Changelog
**v1.1.2 (2024/3/25)**
- `easy kSamplerInpainting` add *additional* widget，you can choose 'Differential Diffusion' or 'Only InpaintModelConditioning'
- Fixed `easy pipeEdit` error when add lora to prompt
- Fixed layerDiffuse xyplot bug
**v1.1.1 (2024/3/16)**
- The issue that the seed is 0 when a node with a seed control is added and **control before generate** is fixed for the first time run queue prompt.
README.md
+6
View File
@@ -34,6 +34,12 @@
## 更新日志
**v1.1.2 (2024/3/25)**
- `easy kSamplerInpainting` 增加 *additional* 属性，可设置成 Differential Diffusion 或 Only InpaintModelConditioning
- 修复 `easy pipeEdit` 提示词输入lora时报错
- 修复 layerDiffuse xyplot相关bug
**v1.1.1 (2024/3/21)**
- 修复首次添加含seed的节点且当前模式为control_before_generate时，seed为0的问题
__init__.py
+1 -1
View File
@@ -73,4 +73,4 @@ WEB_DIRECTORY = "./web"
__all__ = ['NODE_CLASS_MAPPINGS', 'NODE_DISPLAY_NAME_MAPPINGS', "WEB_DIRECTORY"]
print('\033[34mComfy-Easy-Use (v1.1.1): \033[92mLoaded\033[0m')
print('\033[34mComfy-Easy-Use (v1.1.2): \033[92mLoaded\033[0m')
py/easyNodes.py
+55 -43
View File
@@ -10,7 +10,7 @@ from urllib.request import urlopen
from PIL import Image
from server import PromptServer
from nodes import MAX_RESOLUTION, LatentFromBatch, RepeatLatentBatch, NODE_CLASS_MAPPINGS as ALL_NODE_CLASS_MAPPINGS, ConditioningSetMask, ConditioningConcat, CLIPTextEncode, VAEEncodeForInpaint
from nodes import MAX_RESOLUTION, LatentFromBatch, RepeatLatentBatch, NODE_CLASS_MAPPINGS as ALL_NODE_CLASS_MAPPINGS, ConditioningSetMask, ConditioningConcat, CLIPTextEncode, VAEEncodeForInpaint, InpaintModelConditioning
from .config import MAX_SEED_NUM, BASE_RESOLUTIONS, RESOURCES_DIR, INPAINT_DIR, FOOOCUS_STYLES_DIR, FOOOCUS_INPAINT_HEAD, FOOOCUS_INPAINT_PATCH
from .log import log_node_info, log_node_error, log_node_warn
from .wildcards import process_with_loras, get_wildcard_list, process
@@ -3054,6 +3054,7 @@ class samplerSimpleInpainting:
"image_output": (["Hide", "Preview", "Save", "Hide/Save", "Sender", "Sender/Save"],{"default": "Preview"}),
"link_id": ("INT", {"default": 0, "min": 0, "max": sys.maxsize, "step": 1}),
"save_prefix": ("STRING", {"default": "ComfyUI"}),
"additional": (["None", "Differential Diffusion", "Only InpaintModelConditioning"],{"default": "None"})
},
"optional": {
"model": ("MODEL",),
@@ -3072,62 +3073,73 @@ class samplerSimpleInpainting:
FUNCTION = "run"
CATEGORY = "EasyUse/Sampler"
def run(self, pipe, grow_mask_by, image_output, link_id, save_prefix, model=None, mask=None, patch=None, tile_size=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False):
def run(self, pipe, grow_mask_by, image_output, link_id, save_prefix, additional, model=None, mask=None, patch=None, tile_size=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False):
model = model if model is not None else pipe['model']
latent = pipe['samples'] if 'samples' in pipe else None
if 'noise_mask' in latent:
mask = latent['noise_mask']
if mask is not None:
positive = pipe['positive']
negative = pipe['negative']
pixels = pipe["images"] if pipe and "images" in pipe else None
if pixels is None:
raise Exception("No Images found")
vae = pipe["vae"] if pipe and "vae" in pipe else None
if pixels is None:
raise Exception("No VAE found")
x = (pixels.shape[1] // 8) * 8
y = (pixels.shape[2] // 8) * 8
mask = torch.nn.functional.interpolate(mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])),
size=(pixels.shape[1], pixels.shape[2]), mode="bilinear")
pixels = pixels.clone()
if pixels.shape[1] != x or pixels.shape[2] != y:
x_offset = (pixels.shape[1] % 8) // 2
y_offset = (pixels.shape[2] % 8) // 2
pixels = pixels[:, x_offset:x + x_offset, y_offset:y + y_offset, :]
mask = mask[:, :, x_offset:x + x_offset, y_offset:y + y_offset]
if grow_mask_by == 0:
mask_erosion = mask
if additional != "None":
positive, negative, latent = InpaintModelConditioning().encode(positive, negative, pixels, vae, mask)
if additional == "Differential Diffusion":
cls = ALL_NODE_CLASS_MAPPINGS['DifferentialDiffusion']
if cls is not None:
model, = cls().apply(model)
else:
raise Exception("Differential Diffusion not found,please update comfyui")
else:
kernel_tensor = torch.ones((1, 1, grow_mask_by, grow_mask_by))
padding = math.ceil((grow_mask_by - 1) / 2)
if 'noise_mask' not in latent:
if pixels is None:
raise Exception("No Images found")
if vae is None:
raise Exception("No VAE found")
x = (pixels.shape[1] // 8) * 8
y = (pixels.shape[2] // 8) * 8
mask = torch.nn.functional.interpolate(mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])),
size=(pixels.shape[1], pixels.shape[2]), mode="bilinear")
mask_erosion = torch.clamp(torch.nn.functional.conv2d(mask.round(), kernel_tensor, padding=padding), 0,
1)
pixels = pixels.clone()
if pixels.shape[1] != x or pixels.shape[2] != y:
x_offset = (pixels.shape[1] % 8) // 2
y_offset = (pixels.shape[2] % 8) // 2
pixels = pixels[:, x_offset:x + x_offset, y_offset:y + y_offset, :]
mask = mask[:, :, x_offset:x + x_offset, y_offset:y + y_offset]
m = (1.0 - mask.round()).squeeze(1)
for i in range(3):
pixels[:, :, :, i] -= 0.5
pixels[:, :, :, i] *= m
pixels[:, :, :, i] += 0.5
t = vae.encode(pixels)
if grow_mask_by == 0:
mask_erosion = mask
else:
kernel_tensor = torch.ones((1, 1, grow_mask_by, grow_mask_by))
padding = math.ceil((grow_mask_by - 1) / 2)
latent = {"samples": t, "noise_mask": (mask_erosion[:, :, :x, :y].round())}
mask_erosion = torch.clamp(torch.nn.functional.conv2d(mask.round(), kernel_tensor, padding=padding), 0,
1)
# when patch was linked
if patch is not None:
worker = InpaintWorker(node_name="easy kSamplerInpainting")
model, = worker.patch(model, latent, patch)
m = (1.0 - mask.round()).squeeze(1)
for i in range(3):
pixels[:, :, :, i] -= 0.5
pixels[:, :, :, i] *= m
pixels[:, :, :, i] += 0.5
t = vae.encode(pixels)
latent = {"samples": t, "noise_mask": (mask_erosion[:, :, :x, :y].round())}
# when patch was linked
if patch is not None:
worker = InpaintWorker(node_name="easy kSamplerInpainting")
model, = worker.patch(model, latent, patch)
new_pipe = {
**pipe,
"model": model,
"positive": pipe['positive'],
"negative": pipe['negative'],
"vae": pipe['vae'],
"clip": pipe['clip'],
"positive": positive,
"negative": negative,
"vae": vae,
"samples": latent,
"images": pipe['images'],
"seed": pipe['seed'],
"loader_settings": pipe["loader_settings"],
}
else: