search

errno-mmd / mmdmatic

MMD Motion Auto-Trace Installer on Conda
MIT License
27 stars 4 forks source link

RuntimeError:CUDA out of memory #17

Closed CGMikeG closed 3 years ago

CGMikeG commented 3 years ago

File "M:\Programs\mmdmatic\tool\mannequinchallenge-vmd-ver1.03.02\models\hourglass.py", line 54, in forward ret.append(conv(x)) File "C:\Users\Venice.conda\envs\mmdmat\lib\site-packages\torch\nn\modules\module.py", line 547, in call result = self.forward(*input, *kwargs) File "C:\Users\Venice.conda\envs\mmdmat\lib\site-packages\torch\nn\modules\container.py", line 92, in forward input = module(input) File "C:\Users\Venice.conda\envs\mmdmat\lib\site-packages\torch\nn\modules\module.py", line 547, in call result = self.forward(input, **kwargs) File "C:\Users\Venice.conda\envs\mmdmat\lib\site-packages\torch\nn\modules\conv.py", line 343, in forward return self.conv2d_forward(input, self.weight) File "C:\Users\Venice.conda\envs\mmdmat\lib\site-packages\torch\nn\modules\conv.py", line 340, in conv2d_forward self.padding, self.dilation, self.groups) RuntimeError: CUDA out of memory. Tried to allocate 36.00 MiB (GPU 0; 2.00 GiB total capacity; 1.15 GiB already allocated; 32.91 MiB free; 56.71 MiB cached)

CGMikeG commented 3 years ago

I have already tried reducing the frame count to just 30 but for some reason I will got the same problem.

errno-mmd commented 3 years ago

NVIDIA GPU (with 4GB VRAM or more) is needed.

Please refer Supported platforms section in README. https://github.com/errno-mmd/mmdmatic/blob/master/README.en.md#supported-platforms