MrForExample
Python version error
Hello. The python version doesn't work for me, although the miniconda version worked fine. What's wrong? My PC: RTX 2060 SUPER 8GB, 32GB RAM, CUDA 12.1 + TORCH 2.2.0
Error occurred when executing [Comfy3D] Convert 3DGS to Mesh with NeRF and Marching Cubes:
Allocation on device 0 would exceed allowed memory. (out of memory) Currently allocated : 4.63 GiB Requested : 506.12 GiB Device limit : 8.00 GiB Free (according to CUDA): 1.48 GiB PyTorch limit (set by user-supplied memory fraction) : 17179869184.00 GiB
File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\ComfyUI\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\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:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\ComfyUI\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\ComfyUI\custom_nodes\ComfyUI-3D-Pack\nodes.py", line 1337, in convert_gs_ply converter.fit_nerf() File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\ComfyUI\custom_nodes\ComfyUI-3D-Pack\algorithms\nerf_marching_cubes_converter.py", line 174, in fit_nerf image_gt, alpha_gt = self.render_gs(pose) ^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\ComfyUI\custom_nodes\ComfyUI-3D-Pack\algorithms\nerf_marching_cubes_converter.py", line 80, in render_gs out = self.gs_renderer.render(self.gaussians.unsqueeze(0), cam_view.unsqueeze(0), cam_view_proj.unsqueeze(0), cam_pos.unsqueeze(0)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\ComfyUI\custom_nodes\ComfyUI-3D-Pack\lgm\core\gs.py", line 76, in render rendered_image, radii, rendered_depth, rendered_alpha = rasterizer( ^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1511, in wrapped_call_impl return self.call_impl(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\torch\nn\modules\module.py", line 1520, in call_impl return forward_call(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\diff_gaussian_rasterization__init.py", line 213, in forward return rasterize_gaussians( ^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\diff_gaussian_rasterization__init_.py", line 32, in rasterize_gaussians return RasterizeGaussians.apply( ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\torch\autograd\function.py", line 553, in apply return super().apply(*args, **kwargs) # type: ignore[misc] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\OpenSource\NeuralNetworks\ImageVideoGeneration\ComfyWebUI\python_embeded\Lib\site-packages\diff_gaussian_rasterization__init_.py", line 92, in forward num_rendered, color, depth, alpha, radii, geomBuffer, binningBuffer, imgBuffer = _C.rasterize_gaussians(*args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
8gb of vram might be too low. Which workflow did you use? And how many images?
