This repository offers various extension nodes for ComfyUI. Nodes here have different characteristics compared to those in the ComfyUI Impact Pack. The Impact Pack has become too large now...
License: GNU General Public License v3.0
I am running a simple workflow based on your example, but have added in a variable to help keep the seeds aligned. I am using a simple mask to find the face and then using the new regional seed to shift that. However, this does not appear to be working as expected. Just sanity checking what I have here. Workflow attachment in next comments.
Tried updating the nodes used. Switching the version of pillow. Changing FaceDetailer input images to different sizes.
Impact Pack version:4.12
img2texture 1.1.0
Pillow 10.0.1 & 0.9.5
Hey, I was messing around with image refiner last night and I noticed that it was encountering a few errors for example see exhibit 1 below and also noticed that after fixing it I encountered an issue of a missing function from comfyui's main model management set I don't think this is something related to my local install but it is possible will
exhibit 1
Traceback (most recent call last):
File "e:\AI\Stable-Diffusion\ComfyUI\venv\Lib\site-packages\aiohttp\web_protocol.py", line 433, in _handle_request
resp = await request_handler(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "e:\AI\Stable-Diffusion\ComfyUI\venv\Lib\site-packages\aiohttp\web_app.py", line 504, in _handle
resp = await handler(request)
^^^^^^^^^^^^^^^^^^^^^^
File "e:\AI\Stable-Diffusion\ComfyUI\venv\Lib\site-packages\aiohttp\web_middlewares.py", line 117, in impl
return await handler(request)
^^^^^^^^^^^^^^^^^^^^^^
File "e:\AI\Stable-Diffusion\ComfyUI\server.py", line 46, in cache_control
response: web.Response = await handler(request)
^^^^^^^^^^^^^^^^^^^^^^
File "E:\AI\Stable-Diffusion\ComfyUI\custom_nodes\ComfyUI-Workflow-Component\image_refiner\custom_server.py", line 69, in imagerefiner_generate
result = ir.generate(base_pil.convert('RGB'), mask_pil, prompt_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\AI\Stable-Diffusion\ComfyUI\custom_nodes\ComfyUI-Workflow-Component\image_refiner\imagerefiner.py", line 174, in generate
input_data_all = prepare_input(class_def, merged_pil, mask_pil, prompt_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\AI\Stable-Diffusion\ComfyUI\custom_nodes\ComfyUI-Workflow-Component\image_refiner\imagerefiner.py", line 94, in prepare_input
model, clip, vae = load_checkpoint(v['checkpoint'])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\AI\Stable-Diffusion\ComfyUI\custom_nodes\ComfyUI-Workflow-Component\image_refiner\imagerefiner.py", line 38, in load_checkpoint
model, clip, vae, _ = comfy_nodes.CheckpointLoaderSimple().load_checkpoint(ckpt_name)
^^^^^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 4, got 3)
exhibit B
Traceback (most recent call last):
File "e:\AI\Stable-Diffusion\ComfyUI\venv\Lib\site-packages\aiohttp\web_protocol.py", line 433, in _handle_request
resp = await request_handler(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "e:\AI\Stable-Diffusion\ComfyUI\venv\Lib\site-packages\aiohttp\web_app.py", line 504, in _handle
resp = await handler(request)
^^^^^^^^^^^^^^^^^^^^^^
File "e:\AI\Stable-Diffusion\ComfyUI\venv\Lib\site-packages\aiohttp\web_middlewares.py", line 117, in impl
return await handler(request)
^^^^^^^^^^^^^^^^^^^^^^
File "e:\AI\Stable-Diffusion\ComfyUI\server.py", line 46, in cache_control
response: web.Response = await handler(request)
^^^^^^^^^^^^^^^^^^^^^^
File "E:\AI\Stable-Diffusion\ComfyUI\custom_nodes\ComfyUI-Workflow-Component\image_refiner\custom_server.py", line 69, in imagerefiner_generate
result = ir.generate(base_pil.convert('RGB'), mask_pil, prompt_data)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "E:\AI\Stable-Diffusion\ComfyUI\custom_nodes\ComfyUI-Workflow-Component\image_refiner\imagerefiner.py", line 177, in generate
comfy.model_management.unload_model()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: module 'comfy.model_management' has no attribute 'unload_model'
When randomization is turned on, together with control_before_generate, then when using GlobalSeed's value in Save Image node, the value prior to randomization is used for the filename.
Instead it should be the randomized value.
Here is an example workflow that demonstrates the issue:
Here, the seed value prior to randomization was "1", and it is used for the filename "1_00001_.png". However the randomized seed is 1054289837613396.
DWPose complains on boot when not installed.
Wouldn't it be nicer if it is specified on requirements.txt?
I installed this node both manually and automatically using the "install missing custom nodes" feature. But it still shows the node as missing. And shows up in the missing custom node list. I've tried a fresh install of ComfyUI and still same issue. Any idea how to resolve?
Is it possible to get similar functionality in the new "batch_seed_mode" in the face_detailer (from the impact pack) to create consistent faces? Right now the face detailer is generating significantly different faces in a batch even if the underlying images in the batch are very similar to each other.
Greatly appreciate the work!
hi. i get the following error when using the LoRA Loader Block weight:
Error occurred when executing LoraLoaderBlockWeight //Inspire:
Error while deserializing header: HeaderTooLarge
File "/home/user/ai/ComfyUI/execution.py", line 152, in recursive_execute
output_data, output_ui = get output data(obj, input data all)
File "/home/user/al/ComfyUI/execution.py", line 82, in get output_data
return_ values = map node_ over list (obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "home/user/al/ComfyUl/execution.py", line 75, in map node over list
results.append (getattr (obj, func) (**slice_dict (input_data_all, i)))
File " /home/user/ai/ComfyUI/custom nodes/ComfyUI-Inspire-Pack/lora block weight.py", line 250, in doit
lora = comfy.utils. load torch file (lora path, safe load=True)
File "/home/user/ai/ComfyUI/comfy/utils.py", line 11, in load_torch_file
sd = safetensors. torch. load_file (ckpt, device=device.type)
File "/home/user/ai/stable-diffusion-webui/venv/lib/python3.10/site-packages/safetensors/torch.py",line259,in
load file
with safe open(filename, framework="pt", device=device) as f:
The current prompt reading requires a file per prompt, which is painful when I have a set of 1000 prompts already in a CSV file. Can we have CSV as another option here, and perhaps a use of something like comply for prompt weighting and negative use? Getting all of the prompts into one file is the primary pain point I am trying to resolve.
Would it be possible to implement the start/end steps as included in the lates ip-adapter nodes for comfyui?
Hello,
First, thank you for your incredible work!
I'm encountering an issue with the facedetailer adding color to my monochrome images. Is there a possibility of adding a "Color Preprocessor Provider (SEGS)" feature?
Your consideration of this request would be greatly appreciated.
Thank you,
not an issue (sorry, not sure where to post this as it's more a question ) :
how do we check variation seed?
i saw your tutorial on variation seed and the other one for explorting variation seeds.... but what if we want to find the exact seed and variation seed that was used prior? if i run a variation seed with the float, it'll run through images as expected... but it's generating multiple images each time. is there a way to run JUST that seed/variation?
i'm guessing through seed explorer and doing the calculations to reach it? is there something built into inspire pack as a node that achieves this ?
love all the work you do, btw. really awesome stuff. and your web tutorials! i saw folks mention the lack of voice, but with how much you do and churn these out... it totally makes sense you are getting out what you can and it's greatly appreciated. insanely detailed and thorough!
Hitting a wall on a standard X/Y plot. Would appreciate some help trying to get past whatever this is.
Ckpt:
[1] sd_xl_base_1.0
ERROR:root:!!! Exception during processing !!!
ERROR:root:Traceback (most recent call last):
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 153, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 83, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 76, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\lora_block_weight.py", line 260, in doit
model_lora, clip_lora, populated_vector = LoraLoaderBlockWeight.load_lora_for_models(model, clip, lora, strength_model, strength_clip, inverse, seed, A, B, block_vector)
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\lora_block_weight.py", line 151, in load_lora_for_models
raise ValueError(f"[LoraLoaderBlockWeight] invalid block_vector '{block_vector}'")
ValueError: [LoraLoaderBlockWeight] invalid block_vector 'A,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
B2'
Prompt executed in 24.10 seconds
Can you create a text import function along the lines of Image from Directory that allows us to load multiple content from text files in a folder?
I used the xyplot for lora block weight and trying to get the similar result like "effective block analyzer" in sd-webui but never worked. I guess the xyplot mode: diff or diff+heatmap is for this function. When I switched to diff mode and changed A and B in different value, it just give me the same result as simple mode. No comparison images and diff images, no error message either. Do I need to do more to get the diff out? thank you.
I found that the code 'from libs import utils' was added in the files inspire/regional_nodes.py and inspire/segs_support.py. How can I install 'libs'?
Please help! The Value Sender node does not work after the new update.
I noticed a new methot to input a value, but only reroute would connect.
Wich kind of input should i put into?
It´s maybe a bug?
[WARN] ComfyUI-Manager: Your ComfyUI version is outdated. Please update to the latest version.
Traceback (most recent call last):
File "/home/h3c/Documents/ComfyUI/nodes.py", line 1647, in load_custom_node
module_spec.loader.exec_module(module)
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "/home/h3c/Documents/ComfyUI/custom_nodes/ComfyUI-Inspire-Pack/init.py", line 22, in
imported_module = importlib.import_module(".{}".format(module_name), name)
File "/usr/lib/python3.10/importlib/init.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1050, in _gcd_import
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in _load_unlocked
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "/home/h3c/Documents/ComfyUI/custom_nodes/ComfyUI-Inspire-Pack/lora_block_weight.py", line 3, in
import comfy.lora
ModuleNotFoundError: No module named 'comfy.lora'
Cannot import /home/h3c/Documents/ComfyUI/custom_nodes/ComfyUI-Inspire-Pack module for custom nodes: No module named 'comfy.lora'
[WARN] ComfyUI-Impact-Pack: Your ComfyUI version is outdated. Please update to the latest version.
(pysssss:CustomScripts) [warning] it looks like you're running an old version of ComfyUI that requires manual setup of web files, it is recommended you update your installation.
Registered sys.path: ['/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/init.py', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/custom_midas_repo', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/custom_pycocotools', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/custom_mmpkg', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/custom_oneformer', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/custom_detectron2', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src/controlnet_aux', '/home/h3c/Documents/ComfyUI/custom_nodes/comfyui_controlnet_aux/src', '/home/h3c/Documents/ComfyUI/comfy', '/home/h3c/.local/lib/python3.10/site-packages/git/ext/gitdb', '/home/h3c/Documents/ComfyUI', '/usr/lib/python310.zip', '/usr/lib/python3.10', '/usr/lib/python3.10/lib-dynload', '/home/h3c/.local/lib/python3.10/site-packages', '/usr/local/lib/python3.10/dist-packages', '/usr/lib/python3/dist-packages', '/tmp/tmpc2nncunf', '../..', '/home/h3c/Documents/ComfyUI/custom_nodes/ComfyUI-Impact-Pack/modules', '/home/h3c/Documents/ComfyUI/custom_nodes/ComfyUI-Impact-Pack/subpack']
The mediapipe face detector of SEGs was working perfectly for me as of the 10th of September. Then both you and Fannovel16 updated you packs, now I can't get it to detect the face part correctly no matter what I do. I tried to revert to an earlier commit, but unfortunately, evidently I'm not smart enough to make that work. Can you look into weather this can be made to work as it was when you first made it? Thanks!
What is the reason despite installing the
Inspire-Pack)
Hello,
thank you for your work!
I want to know that the number of lora block weights in inspire pack is different from webui extension. What is the difference between the two?
Thank you.
Error occurred when executing ApplyRegionalIPAdapters //Inspire:
IPAdapterApply.apply_ipadapter() got an unexpected keyword argument 'weight_type'
Just watched your video on hands, but am hitting an error.
!!! Exception during processing !!!
Traceback (most recent call last):
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\impact_pack.py", line 1703, in doit
DetailerForEach.do_detail(image, segs, model, clip, vae, guide_size, guide_size_for, max_size, seed, steps, cfg,
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\impact_pack.py", line 609, in do_detail
enhanced_pil = core.enhance_detail(cropped_image, model, clip, vae, guide_size, guide_size_for_bbox, max_size,
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\core.py", line 227, in enhance_detail
positive = control_net_wrapper.apply(positive, upscaled_image)
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\core.py", line 1268, in apply
image = self.preprocessor.apply(image)
File "D:\stable_diffusion\ComfyUI\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\segs_support.py", line 37, in apply
return obj.estimate_pose(image, detect_hand, detect_body, detect_face)[0]
KeyError: 0
Thought I would try and use this node to read embeds for IPadapter. Not sure where to put these, so I wanted to try this. However, I can't seem to get the path to work correctly in windows. I get this error:
Error occurred when executing LoadImagesFromDir //Inspire:
object of type 'NoneType' has no len()
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\stable_diffusion\ComfyUI\ComfyUI\execution.py", line 96, in get_output_data
output_is_list = [False] * len(results[0])
wildcardEncode (inspire pack) seed can't work for input
He is either random or fixed before conversion, and does not change according to the node that flows in.
by the way Global seed (inspire pack) seems not work good at control_before_generate,.
when starting comfyui I get this after all nodes are loaded:
Starting server
To see the GUI go to: http://127.0.0.1:8188
FETCH DATA from: D:\Ai_art\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\extension-node-map.json
[ERROR] An error occurred while retrieving information for the 'WildcardEncode //Inspire' node.
Traceback (most recent call last):
File "D:\Ai_art\ComfyUI_windows_portable\ComfyUI\server.py", line 417, in get_object_info
out[x] = node_info(x)
File "D:\Ai_art\ComfyUI_windows_portable\ComfyUI\server.py", line 395, in node_info
info['input'] = obj_class.INPUT_TYPES()
File "D:\Ai_art\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\prompt_support.py", line 318, in INPUT_TYPES
wildcards = nodes.NODE_CLASS_MAPPINGS['ImpactWildcardEncode'].get_wildcard_list()
AttributeError: type object 'ImpactWildcardEncode' has no attribute 'get_wildcard_list'
if I change sampler_name to dpmpp_2m_sde, then the video sequence only contains a few frames, e.g. 3-4 out of 20. And the decoded latent is also just noise that you would see at step 4.
Canny works great, but when switching to the OpenPose Preprocessors Provider (SEGS), an error occurs:
the example json:
How to blend preprocessor provider (segs) 's output or convert image to segs_preprocessor
Sometimes, the result of the preprocessor is not as good as we want it to be,
so I try to improve the final result by blend them,
Question:
1 How how to blend segs_preprocessor
2 How to convert image to segs_preprocessor
3 How to preview segs_preprocessor?
Just letting you (and others) know that this node conflicts with the Nested Nodes. I don't know enough about coding to know if this issue is fixable or not.
Plz filter without the <lora;**> before output the conditioning in WildCard Encode,
I find remove and filter all the <lora;**> before output the conditioning can get the better Semantic understanding and better graphics effect
Error occurred when executing ApplyRegionalIPAdapters //Inspire:
Error(s) in loading state_dict for ImageProjModel:
size mismatch for proj.weight: copying a param with shape torch.Size([8192, 1280]) from checkpoint, the shape in current model is torch.Size([8192, 1024]).
Not sure if this is an actual bug but is it possible Inspire Pack Image Loader places heavier resource requirements on Comfyui relative to the vanilla image loader?
A couple of times in the last week or so when I've loaded images via Inspire Pack image loader Comfyui starts to respond slowly and eventually becomes non-responsive. When I check the logs I notice large amounts of encoded data. Initially i thought it was the size of images I'd placed in CLIPspace. However, last night I seemed to fix the issue by deleting the Inspire Image Loader and replacing it with vanilla (same images though). After reverting to Vanilla Loader I also went into log, and cleared it in case Comfyui is allocating background resources to maintaining the log.
Plz add A111 mixed prompt [Cockapoo_ear:Chihuahua:0.3] and other Features like smzNode
https://github.com/shiimizu/ComfyUI_smZNodes
Alternating words
Weight normalization
Usage of BREAK and AND keywords
Optional embedding: identifier
this is different between wildcardEncode ,smzNode, bnk advanced clip text encode
`Directory 'F:\ai-ref\Openpose\poses cannot be found.'
File "f:\AI\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "f:\AI\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "f:\AI\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "F:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\inspire\image_util.py", line 31, in load_images
raise FileNotFoundError(f"Directory '{directory} cannot be found.'")`
Moving the folder to the ComfyUI directory and writing the path in the format you wrote in #17 does not help
Wildcards do not load in the "Wilcard Encode (inspire)" of the Inspire Pack when I select "populate"
but they work on the "ImpactWilcardEncode" of the impact pack
this is what I get from the cmd:
got prompt
[ERROR] An error occurred during the on_prompt_handler processing
Traceback (most recent call last):
File "D:\SD\ComfyUI_windows_portable\ComfyUI\server.py", line 633, in trigger_on_prompt
json_data = handler(json_data)
^^^^^^^^^^^^^^^^^^
File "D:\SD\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\inspire\inspire_server.py", line 203, in onprompt
populate_wildcards(json_data)
File "D:\SD\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\inspire\inspire_server.py", line 181, in populate_wildcards
inputs['populated_text'] = wildcard_process(text=inputs['wildcard_text'], seed=int(inputs['seed']))
^^^^^^^^^^^^^^^^^^^
TypeError: int() argument must be a string, a bytes-like object or a real number, not 'list'
CLIP:
Requested to load BaseModel
Hi, I am running an XY plot workflow via API, although the workflow contains no Inspire Pack nodes, I get the following message when the script starts (the script does continue and complete without problems though)
got prompt
[ERROR] An error occurred during the on_prompt_handler processing
Traceback (most recent call last):
File "/home/jh/ai/ComfyUI/server.py", line 625, in trigger_on_prompt
json_data = handler(json_data)
File "/home/jh/ai/ComfyUI/custom_nodes/ComfyUI-Inspire-Pack/inspire_server.py", line 122, in onprompt
is_changed = prompt_seed_update(json_data)
File "/home/jh/ai/ComfyUI/custom_nodes/ComfyUI-Inspire-Pack/inspire_server.py", line 63, in prompt_seed_update
seed_widget_map = json_data['extra_data']['extra_pnginfo']['workflow']['seed_widgets']
KeyError: 'extra_data'
Hi I wanted to try SEGS for inpainting workflows, but it looks like there's no preprocessor for inpainting controlnet, is it possible to add? Thank you!
When some nodes are muted and/or bypassed, GlobalSeed does not replace the seed in any of the nodes.
Here is an example workflow:
In some cases the workflow won't execute at all, but I cannot reproduce that anymore rn...
fix this ,its easy and it is logical error on your side
`def load_images(self, directory: str, image_load_cap: int = 0, start_index: int = 0):
if not os.path.isdir(directory):
raise FileNotFoundError(f"Directory '{directory} cannot be found.'")
dir_files = os.listdir(directory)
if len(dir_files) == 0:
raise FileNotFoundError(f"No files in directory '{directory}'.")
# Filter files by extension
valid_extensions = ['.jpg', '.jpeg', '.png']
dir_files = [f for f in dir_files if any(f.lower().endswith(ext) for ext in valid_extensions)]
dir_files = sorted(dir_files)
dir_files = [os.path.join(directory, x) for x in dir_files]
# start at start_index
dir_files = dir_files[start_index:]`
This is your good tutor to use your "xy plot lora block weight " to get perfect fixed LBW
https://www.youtube.com/watch?v=X9v0xQrInn8&t=252s
https://githubfast.com/ltdrdata/ComfyUI-extension-tutorials/blob/Main/ComfyUI-Inspire-Pack/tutorial/LoraBlockWeight.md
How about add "xy plot model block weight"? need 2 model input: 1model and 1x and 1y outputs:
Is it expected for the results of KSampler (Inspire) (in CPU mode) to differ from KSampler:
I followed the youtube window example (https://youtu.be/X9v0xQrInn8). What I found was the Block Weight Node doesn’t seem to be loading the Lora and I ended up with the same 3 images recurring no matter which blocks we being activated in either scenario NONE, ALL, MIDD or FULL TEST 17.
E.g.:
Model: Henmix_Real_v40
Lora: edgKM
Running windows 10.
The module at the custom_nodes/ComfyUI-Impact-Pack/impact_subpack path appears to be incomplete.
Recommended to delete the path and restart ComfyUI.
If the issue persists, please report it to https://github.com/ltdrdata/ComfyUI-Impact-Pack/issues.
I use the GlobalSeed //Inspire node a lot and I love it!
Would it be possible to have the same for sampler_name/scheduler?
My use case is that I have a larger workflow with many samplers (~14) and sometimes want to test it with different samplers. Extracting as primitive works, but makes the workflow messier, especially since primitives have (had?) race condition bugs with reroutes.
GlobalSeed solves this beautifully for the seed, it would be great to have the same for the sampler/scheduler. Like with the seed, converting sampler_name/scheduler to input should prevent the replacement.
Thanks!
When doing the testing, should the first block always be 1? Is that representing the base?
B1:A,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
B2:0,A,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
B3:0,0,A,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
B4:0,0,0,A,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
plz delete resources\lbw-preset.txt in github, every update will replace my edited lbw-preset.txt
(just give a readme let user create it himself ) or if you have another good idead
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.