comfyui中的k采样器报错Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 54] to have 4

最新推荐文章于 2025-03-05 14:46:35 发布
最新推荐文章于 2025-03-05 14:46:35 发布
阅读量2.8k 收藏 10
点赞数 19
文章标签： windows 服务器 microsoft
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/M1226401728/article/details/142941742
版权

求助大佬！！！我运行comfyui的时候，k采样器变紫，同时报错显示：

Error occurred when executing KSampler: Given groups=1, weight of size [320, 4, 3, 3], expected input[2, 8, 96, 54] to have 4 channels, but got 8 channels instead File "E:\ComfyUI-aki-v1.3\execution.py", line 316, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "E:\ComfyUI-aki-v1.3\execution.py", line 191, in get_output_data return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "E:\ComfyUI-aki-v1.3\execution.py", line 168, in _map_node_over_list process_inputs(input_dict, i) File "E:\ComfyUI-aki-v1.3\execution.py", line 157, in process_inputs results.append(getattr(obj, func)(**inputs)) File "E:\ComfyUI-aki-v1.3\nodes.py", line 1429, in sample return common_ksampler(model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise) File "E:\ComfyUI-aki-v1.3\nodes.py", line 1396, in common_ksampler samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-Impact-Pack\modules\impact\sample_error_enhancer.py", line 22, in informative_sample raise e File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-Impact-Pack\modules\impact\sample_error_enhancer.py", line 9, in informative_sample return original_sample(*args, **kwargs) # This code helps interpret error messages that occur within exceptions but does not have any impact on other operations. File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-AnimateDiff-Evolved\animatediff\sampling.py", line 420, in motion_sample return orig_comfy_sample(model, noise, *args, **kwargs) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\sampling.py", line 116, in acn_sample return orig_comfy_sample(model, *args, **kwargs) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\utils.py", line 117, in uncond_multiplier_check_cn_sample return orig_comfy_sample(model, *args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\sample.py", line 43, in sample samples = sampler.sample(noise, positive, negative, cfg=cfg, latent_image=latent_image, start_step=start_step, last_step=last_step, force_full_denoise=force_full_denoise, denoise_mask=noise_mask, sigmas=sigmas, callback=callback, disable_pbar=disable_pbar, seed=seed) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 104, in KSampler_sample return orig_fn(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-TiledDiffusion\utils.py", line 51, in KSampler_sample return orig_fn(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 829, in sample return sample(self.model, noise, positive, negative, cfg, self.device, sampler, sigmas, self.model_options, latent_image=latent_image, denoise_mask=denoise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 122, in sample return orig_fn(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 729, in sample return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 716, in sample output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 695, in inner_sample samples = sampler.sample(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 87, in KSAMPLER_sample return orig_fn(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-TiledDiffusion\utils.py", line 34, in KSAMPLER_sample return orig_fn(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 600, in sample samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options) File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context return func(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\k_diffusion\sampling.py", line 144, in sample_euler denoised = model(x, sigma_hat * s_in, **extra_args) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 299, in __call__ out = self.inner_model(x, sigma, model_options=model_options, seed=seed) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 682, in __call__ return self.predict_noise(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 685, in predict_noise return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI_smZNodes\smZNodes.py", line 162, in sampling_function out = orig_fn(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 279, in sampling_function out = calc_cond_batch(model, conds, x, timestep, model_options) File "E:\ComfyUI-aki-v1.3\comfy\samplers.py", line 226, in calc_cond_batch output = model_options['model_function_wrapper'](model.apply_model, {"input": input_x, "timestep": timestep_, "c": c, "cond_or_uncond": cond_or_uncond}).chunk(batch_chunks) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-IC-Light-Native\ic_light_nodes.py", line 116, in wrapper_func return existing_wrapper(unet_apply, params=apply_c_concat(params)) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-IC-Light-Native\ic_light_nodes.py", line 108, in unet_dummy_apply return unet_apply(x=params["input"], t=params["timestep"], **params["c"]) File "E:\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\utils.py", line 69, in apply_model_uncond_cleanup_wrapper return orig_apply_model(self, *args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\model_base.py", line 142, in apply_model model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds).float() File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl return forward_call(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\custom_nodes\FreeU_Advanced\nodes.py", line 176, in __temp__forward h = forward_timestep_embed(module, h, emb, context, transformer_options, time_context=time_context, num_video_frames=num_video_frames, image_only_indicator=image_only_indicator) File "E:\ComfyUI-aki-v1.3\comfy\ldm\modules\diffusionmodules\openaimodel.py", line 50, in forward_timestep_embed x = layer(x) File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl return forward_call(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\comfy\ops.py", line 97, in forward return super().forward(*args, **kwargs) File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\nn\modules\conv.py", line 460, in forward return self._conv_forward(input, self.weight, self.bias) File "E:\ComfyUI-aki-v1.3\python\lib\site-packages\torch\nn\modules\conv.py", line 456, in _conv_forward return F.conv2d(input, weight, bias, self.stride,

comfyui 更新后运行，遇到K采样器报错：AttributeError: ‘UNetModel object has no attribute ‘default image only indic
代码简单说 Vue、JAVA、PHP、Node.js 熟练运用，接口、架构、性能全搞定。
10-21 1671
解决的办法是:删掉freeU节点，到秋叶启动器中扩展里面，卸载这个插件，如果不是启动器安装的直接手动自己去custom nodes文件夹下删掉也是一样的 freeu 插件。
ComfyUI常规报错及处理办法
热门推荐
细节控-AI先行者的博客
12-15 1万+
ComfyUI的插件管理，和应用端，没有webUI成熟。玩comfyui就要适应各种不断的报错信息。
【已解决】RuntimeError: Given groups=1, weight of size [512, 1024, 3, 3], expected input[1, 640, 8, 8]
weixin_44718503的博客
11-19 1797
权重维度：[512, 1024, 3, 3]，意味着有512个输出通道（或卷积核），每个卷积核期望有1024个输入通道，卷积核的尺寸是3x3。卷积层期望的输入通道数是1024，但实际提供的只有640。这表明在模型的前一层和当前层之间，通道数丢失了。输入张量维度：[1, 640, 8, 8]，表示输入张量有1个批次，640个通道，特征图尺寸是8x8。这个错误发生在卷积层接收到的输入张量与其权重期望的通道数不一致时。主要是受到了大佬这篇博文的启发。
ComfyUI报错/WanVideoSampler：torch._scaled_mm is only supported on CUDA devices with compute capability
qq_42763564的博客
02-28 902
ComfyUI报错/WanVideoSampler：torch._scaled_mm is only supported on CUDA devices with compute capability >= 9.0 or 8.9, or ROCm MI300+
(241条消息)Python中报错“TypeError:‘list‘objectisnotcallable”的解决方法
11-10
资源(241条消息)Python中报错“TypeError:‘list‘objectisnotcallable”的解决方法知识分享
RuntimeError: Given groups=1, weight of size 32 3 3 3, expected input[1, 4, 416, 416] to have 3 chan
weixin_37718439的博客
02-05 1万+
·····运行eriklindernoren/PyTorch-YOLOv3，出现错误RuntimeError: Given groups=1, weight of size 32 3 3 3, expected input[1, 4, 416, 416] to have 3 channels, but got 4 channels instead ·····解决方法： ·····把img = tr...
ComfyUI报错：AttributeError: ‘UNetModel‘ object has no attribute ‘default_image_only_indicator‘
聚梦小课堂
03-03 1万+
​本来想使用SVD生成个视频来着，结果更新完comfyui，运行之前的一个工作流的时候，运行到采样节点KSampler的时候，遇到报错了。后台的提示是：AttributeError: 'UNetModel' object has no attribute 'default_image_only_indicator'；然后简单查了下资料，发现comfyui的官网也有好多人遇到这个问题，解决的办法是... ​
comfyui中报错，出现一下问题，请大神帮忙
hhcvbc的博客
03-04 4527
comfyui 报错问题
【Comfyui】Error occurred when executing ADE UseEvolvedSampling: ModelPatcher’ object has no attribute
sert287的博客
08-15 885
ModelPatcher’ object has no attribute 'current device。这个报错主要原因是：更新了ComfyUI,没有更新AnimateDiff Envolved节点导致的。2.点击“安装节点”，搜索“AnimateDiff Envolved”；解决方法：更新AnimateDiff Envolved节点。注意：如果在线更新不好用的话，可以手动下载文件更新。1.点击ComfyUI界面右下角“管理器”4.更新完成后，重启ComfyUI。
一篇文章解决ComfyUI常见的故障报错！
2401_84760527的博客
09-20 9349
学习和使用ComfyUI最痛苦的是什么？就是这满屏的红色方框和和[报错信息] “报错信息”)，处理完一批又一批，很多人玩了一两个流程就搞不下去了，很多初学者因此就放弃了。有道是：配置流程大半天，跑通出图三分钟。ComfyUI的快感可能就是这三分钟。体验[comfyUI]给你带来的高效和快感，希望这篇文章能帮助解决常见的技术问题，让你畅快淋漓的使用ComfyUI。
【comfyui教程】comfyui攻略：故障报错应对指南！
2401_84760527的博客
11-12 7363
ComfyUI一旦正确配置和安装，确实能提供高效而强大的工作环境，使得处理各种任务变得更加简单。用户能够体验到它带来的便利性和灵活性，特别是在处理复杂的流程和自定义需求时。然而，ComfyUI的一个显著挑战在于它需要定期进行版本更新和组件升级。这在国内用户的使用体验中尤其突出，因为网络环境和访问限制使得获取最新的更新和组件成为了一个不小的挑战。用户可能需要寻找特殊的下载途径或使用代理服务来维持其软件的最新状态，这不仅增加了使用的复杂性，也可能导致在关键时刻遇到技术障碍。
ComfyUI 图片RMBG后出现通道不匹配问题
尘思录
05-16 2637
在ComfyUI中，将图片通过controlnet后送入sampler，流程能正常运行。原因为张量输入数据的张量通过不匹配。采用以下三个节点之一。
``` import torch import torch.nn as nn from skimage.segmentation import active_contour import numpy as np from torchvision.ops import FeaturePyramidNetwork from torchvision.models import resnet50 import os os.environ['KMP_DUPLICATE_LIB_OK']='TRUE' class ImageSegmentationModel(nn.Module): def __init__(self): super(ImageSegmentationModel,self).__init__() self.conv_layers = nn.Sequential( nn.Conv2d(1,1,kernel_size=3,stride=2), nn.MaxPool2d(kernel_size=3,stride=2), nn.ReLU(), nn.Conv2d(1, 1, kernel_size=3, stride=2), nn.MaxPool2d(kernel_size=3, stride=2), nn.ReLU(),nn.Conv2d(1,1,kernel_size=3,stride=2), nn.MaxPool2d(kernel_size=3,stride=2), nn.ReLU(), nn.Conv2d(1,1,kernel_size=3,stride=2) ) #使用resnet作为特征提取器 self.resnet = resnet50(pretrained=True) #移除resnet最后一层全连接层 self.resnet = nn.Sequential(*list(self.resnet.children())[:-2]) #修改，调整fpn输入通道 self.fpn = FeaturePyramidNetwork([256,512,1024,2048],256) def preprocess(self,x): #将输入图像尺寸调整为511×511，使用双线性插值法 x = torch.nn.functional.interpolate(x,size=(511,511),mode='bilinear',align_corners=False) #将输入图像转换为灰度图像，通过对通道维度求均值 x = torch.mean(x,dim=1,keepdim=True) #将张量转换为numpy数组，以便使用skimage库进行处理 x_np = x.detach().cpu().numpy() segmented = [] #对每个样本进行chan—vese分割 for i in range(x_np.shape[0]): img = x_np[i,0] #初始化活动轮廓的点 init =np.array([[img.shape[1]-1,0],[img.shape[1]-1,img.shape[0]-1],[0,img.shape[0]-1,],[0,0]]) #使用c-v函数进行分割 snake = active_contour(img,init,alpha=0.015,beta=10,gamma=0.001) #初始化分割结果矩阵 seg = np.zeros_like(img) from skimage.draw import polygon #绘制多边形区域（有无绘制必要？） rr, cc = polygon(snake[:,1],snake[:,0],seg.shape) #将多边形区域标记为1 seg[rr, cc] = 1 segmented.append(seg) #将分割结果转换为numpy数组 segmented = np.array(segmented) #将数组转换为张量，并添加通道维度 segmented = torch.from_numpy(segmented).unsqueeze(1).float().to(x.device) return segmented def forward(self,x): y = torch.nn.functional.interpolate(x,size=(511,511),mode='bilinear',align_corners=False) #对输入图像进行预处理 x = self.preprocess(x) #通过卷积层序列进行特征提取 conv_output = self.conv_layers(x) #通过resnet50提取特征 resnet_features = self.resnet(y) #构建fpn输入特征字典 fpn_input = { 'feat1': resnet_features[0], 'feat2': resnet_features[1], 'feat3': resnet_features[2], 'feat4': resnet_features[3] } #通过fpn网络 fpn_output = self.fpn(fpn_input) #最终卷积操作 fpn_output_upsampled = torch.nn.functional.interpolate(fpn_output['feat1'], size=(511, 511), mode='bilinear', align_corners=False) final_output = nn.functional.conv2d(fpn_output_upsampled,conv_output,stride=1,paddong=0) return final_output```运行上述代码时发生报错：RuntimeError: Given groups=1, weight of size [256, 1024, 1, 1], expected input[1, 2048, 16, 16] to have 1024 channels, but got 2048 channels instead。该怎么解决？
最新发布
03-15
因此，当用户使用`*list(resnet.children())[:-2]`时，可能保留了到layer4的部分，但这样整个Sequential的输出就是layer4的输出，即2048通道的特征图，而无法获得layer1到layer3的输出。 因此，这里存在一个根本性...
动手学深度学习实现DAY-3
weixin_40054643的博客
02-25 4079
节选自“ElitesAI·动手学深度学习PyTorch版”. （Task06：批量归一化和残差网络；凸优化；梯度下降（1天）） Task09：目标检测基础；图像风格迁移；图像分类案例1（1天） Task10：图像分类案例2；GAN；DCGAN（1天） 批量归一化（BatchNormalization） 对输入的标准化（浅层模型） 处理后的任意一个特征在数据集中所有样本上的均值为0、标准差为1...
Caffe2 - (二十) Detectron 之 config.py 文件参数
长风破浪会有时，直挂云帆济沧海
03-22 7218
Caffe2 - (二十) Detectron 之 config.py 文件 给出了 Detectron 的默认参数，其位于 lib/core/config.py. 类似于 Faster R-CNN 中对应的形式.一般不更改该文件参数，可以根据设置 yaml 文件并利用 merge_cfg_from_file (yaml_file) 来加载自定义参数
基于循环特征位移聚合器的车道线检测（RESA: Recurrent Feature-Shift Aggregator for Lane Detection）
Wincher's blog
03-28 5794
2021年的车道线检测新方法。 官方公开视频、论文、源码： https://www.bilibili.com/video/BV1664y1o7wg https://arxiv.org/abs/2008.13719 https://github.com/ZJULearning/resa （该视频对现有车道线检测方法进行了分类、归纳总结，很完善。建议食用） 基于深度学习的车道线检测方法可分为如下几类： 实例分隔。例：LaneNet 语义分隔。例：SCNN 网格化。 多项式。 基于Anchor。 本方法是
RuntimeError: Given groups=1, weight of size [32, 32, 1, 1], expected input[1, 64, 28, 28] to have 3
m0_45924886的博客
05-07 5471
报错：RuntimeError: Given groups=1, weight of size [32, 32, 1, 1], expected input[1, 64, 28, 28] to have 32 channels, but got 64 channels instead。报此类错误是前后卷积层的通道不匹配，比如这里期望输入的通道是32，而上一层卷积传入的通道却是64，所以更改后面的输入通道即可。
returned non-zero exit status
weixin_45808526的博客
03-05 126
pip环境混乱。
keras 模型中自定义上采样函数，加载模型报错: ValueError: Unknown layer: BilinearUpsampling
qq_21466543的博客
10-28 2313
模型自定义了BilinearUpsampling层如下代码，模型训练能正常调用BilinearUpsampling，模型保存用ModelCheckpoint，测试时加载模型报错：ValueError: Unknown layer: BilinearUpsampling class BilinearUpsampling(Layer): def __init__(self, upsamp...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值