(lhm_env) (base) PS D:\PythonWork\LHM> python ./app.py
GPU ID: 0, Total Memory: 24564.0 MB, Used Memory: 1592.0 MB, Available Memory: 22.43 GB
GPU ID: 0, Total Memory: 24564.0 MB, Used Memory: 1592.0 MB, Available Memory: 22.43 GB
as your model does not large than 18GB, we will use LHM-MINI instead.
[896]
Loading model...
A matching Triton is not available, some optimizations will not be enabled
Traceback (most recent call last):
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\xformers_init_.py", line 55, in _is_triton_available
from xformers.triton.softmax import softmax as triton_softmax # noqa
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\xformers\triton\softmax.py", line 11, in
import triton
ModuleNotFoundError: No module named 'triton'
D:\PythonWork\LHM./engine/pose_estimation\blocks/../dinov2\dinov2\layers\swiglu_ffn.py:43: UserWarning: xFormers is available (SwiGLU)
warnings.warn("xFormers is available (SwiGLU)")
D:\PythonWork\LHM./engine/pose_estimation\blocks/../dinov2\dinov2\layers\attention.py:27: UserWarning: xFormers is available (Attention)
warnings.warn("xFormers is available (Attention)")
D:\PythonWork\LHM./engine/pose_estimation\blocks/../dinov2\dinov2\layers\block.py:33: UserWarning: xFormers is available (Block)
warnings.warn("xFormers is available (Block)")
person center is head
Model loaded
Fetching 4 files: 100%|██████████████████████████████████████████████████████████████████████████| 4/4 [00:00<?, ?it/s]
D:\PythonWork\LHM\LHM\models\encoders\dinov2\layers\swiglu_ffn.py:43: UserWarning: xFormers is available (SwiGLU)
warnings.warn("xFormers is available (SwiGLU)")
D:\PythonWork\LHM\LHM\models\encoders\dinov2\layers\attention.py:27: UserWarning: xFormers is available (Attention)
warnings.warn("xFormers is available (Attention)")
D:\PythonWork\LHM\LHM\models\encoders\dinov2\layers\block.py:39: UserWarning: xFormers is available (Block)
warnings.warn("xFormers is available (Block)")
==========skip_decoder:True
Traceback (most recent call last):
File "D:\PythonWork\LHM\app.py", line 793, in
launch_gradio_app()
File "D:\PythonWork\LHM\app.py", line 785, in launch_gradio_app
lhm = _build_model(cfg)
File "D:\PythonWork\LHM\app.py", line 282, in _build_model
model = hf_model_cls.from_pretrained(cfg.model_name)
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\huggingface_hub\utils_validators.py", line 114, in _inner_fn
return fn(*args, **kwargs)
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\huggingface_hub\hub_mixin.py", line 511, in from_pretrained
instance = cls._from_pretrained(
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\huggingface_hub\hub_mixin.py", line 727, in _from_pretrained
model = cls(**model_kwargs)
File "D:\PythonWork\LHM\LHM\utils\hf_hub.py", line 23, in init
super().init(**config)
File "D:\PythonWork\LHM\LHM\models\modeling_human_lrm.py", line 528, in init
super(ModelHumanLRMSapdinoBodyHeadSD3_5, self).init(**kwargs)
File "D:\PythonWork\LHM\LHM\models\modeling_human_lrm.py", line 128, in init
self.renderer = GS3DRenderer(
File "D:\PythonWork\LHM\LHM\models\rendering\gs_renderer.py", line 792, in init
self.smplx_model = SMPLXVoxelMeshModel(
File "D:\PythonWork\LHM\LHM\models\rendering\smpl_x_voxel_dense_sampling.py", line 511, in init
self.smplx_init()
File "D:\PythonWork\LHM\LHM\models\rendering\smpl_x_voxel_dense_sampling.py", line 773, in smplx_init
nn_vertex_idxs = knn_points(
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\pytorch3d\ops\knn.py", line 189, in knn_points
p1_dists, p1_idx = _knn_points.apply(
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\torch\autograd\function.py", line 598, in apply
return super().apply(*args, **kwargs) # type: ignore[misc]
File "D:\PythonWork\LHM\lhm_env\lib\site-packages\pytorch3d\ops\knn.py", line 74, in forward
idx, dists = _C.knn_points_idx(p1, p2, lengths1, lengths2, norm, K, version)
RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
我运行的install_cu121.bat(https://github.com/aigc3d/LHM/blob/main/install_cu121.bat)
里面是这样的
@echo off
echo Installing Torch 2.3.0...
pip install torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/cu121
pip install -U xformers==0.0.26.post1 --index-url https://download.pytorch.org/whl/cu121
echo Installing dependencies...
pip install -r requirements.txt
echo Uninstalling basicsr to avoid conflicts...
pip uninstall -y basicsr
pip install git+https://github.com/XPixelGroup/BasicSR
echo Installing pytorch3d...
pip install "git+https://github.com/facebookresearch/pytorch3d.git"
echo Installing sam2...
pip install git+https://github.com/hitsz-zuoqi/sam2/
echo Installing diff-gaussian-rasterization...
pip install git+https://github.com/ashawkey/diff-gaussian-rasterization/
echo Installing simple-knn...
pip install git+https://github.com/camenduru/simple-knn/
echo Installation completed!
pause
我不知道是什么编译的，就运行的这个bat
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
请问有什么解决方法吗？或者有什么我可以提供的。显卡是4090，cuda12.1，pytorch2.3,python 3.10。卡这里好久了
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
请问有什么解决方法吗？或者有什么我可以提供的。显卡是4090，cuda12.1，pytorch2.3,python 3.10。卡这里好久了
你打开x64 terminal，看下这个readme 咋做的 https://github.com/aigc3d/LHM/blob/feat/comfyui/Windows11_install.md
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
请问有什么解决方法吗？或者有什么我可以提供的。显卡是4090，cuda12.1，pytorch2.3,python 3.10。卡这里好久了
你打开x64 terminal，看下这个readme 咋做的 https://github.com/aigc3d/LHM/blob/feat/comfyui/Windows11_install.md
windows的安装确实比较mind fuck，按照教程先装，有问题的话再贴到这个issue下面
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
请问有什么解决方法吗？或者有什么我可以提供的。显卡是4090，cuda12.1，pytorch2.3,python 3.10。卡这里好久了
你打开x64 terminal，看下这个readme 咋做的 https://github.com/aigc3d/LHM/blob/feat/comfyui/Windows11_install.md
windows的安装确实比较mind fuck，按照教程先装，有问题的话再贴到这个issue下面
一般情况提供的bat文件不能直接安装好，每个人的windows 配置都不一样
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
请问有什么解决方法吗？或者有什么我可以提供的。显卡是4090，cuda12.1，pytorch2.3,python 3.10。卡这里好久了
你打开x64 terminal，看下这个readme 咋做的 https://github.com/aigc3d/LHM/blob/feat/comfyui/Windows11_install.md
windows的安装确实比较mind fuck，按照教程先装，有问题的话再贴到这个issue下面
一般情况提供的bat文件不能直接安装好，每个人的windows 配置都不一样
好的，非常感谢。我也会把我运行bat的执行结果发这里。
你的pytorch3d包是咋编译的铁子，是不是用cpu编译的？
请问有什么解决方法吗？或者有什么我可以提供的。显卡是4090，cuda12.1，pytorch2.3,python 3.10。卡这里好久了
我也有同样的问题，我的显卡是v100，显存40G，cuda 11.8 pytorch3d我是按照教程从元码编译安装的。
aigc3d
