search

OpenGVLab / DragGAN

Unofficial Implementation of DragGAN - "Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold" （DragGAN 全功能实现，在线Demo，本地部署试用，代码、模型已全部开源，支持Windows, macOS, Linux）
4.99k stars 489 forks source link

Mac 电脑上异常： AssertionError: Torch not compiled with CUDA enabled #72

Open chwshuang opened 1 year ago

chwshuang commented 1 year ago

drag_gan.py 171 行 api.py 171 行 ··· File "/Users/abc/DragGAN/drag_gan.py", line 171, in neighbor points.append(torch.tensor([i, j]).float().cuda()) File "/Users/abc/miniconda3/envs/draggan2/lib/python3.7/site-packages/torch/cuda/init.py", line 221, in _lazy_init raise AssertionError("Torch not compiled with CUDA enabled") AssertionError: Torch not compiled with CUDA enabled ···

没有CUAD的机器上，代码需要修改为： 167 行 ·device = F.device· 171行 ··· points.append(torch.tensor([i, j]).float().cuda()) 上面的代码修改为下面代码： points.append(torch.tensor([i, j]).float().to(device)) ···

mifeng925 commented 1 year ago

1.打开 draggan.py 搜索 cuda 替换成 cpu 2.使用 python -m draggan.web --device cpu 运行