search

xinntao / Real-ESRGAN

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
BSD 3-Clause "New" or "Revised" License
28.45k stars 3.57k forks source link

Error "compute_indices_weights_nearest" not implemented for 'Half' #756

Closed Dewey-Wang closed 8 months ago

Dewey-Wang commented 8 months ago

I followed all the installation instruction and try to apply the model on the inputs\0030.jpg. This is what I got.

(intern) C:\Users\Ding-Yang\Real-ESRGAN>python inference_realesrgan.py -n RealESRGAN_x4plus -i inputs\0030.jpg --outscale 3.5 --face_enhance

C:\Users\Ding-Yang\anaconda3\envs\intern\Lib\site-packages\torchvision\models_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead. warnings.warn( C:\Users\Ding-Yang\anaconda3\envs\intern\Lib\site-packages\torchvision\models_utils.py:223: UserWarning: Arguments other than a weight enum or None for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing weights=None. warnings.warn(msg) Downloading: "https://github.com/xinntao/facexlib/releases/download/v0.1.0/detection_Resnet50_Final.pth" to C:\Users\Ding-Yang\Real-ESRGAN\gfpgan\weights\detection_Resnet50_Final.pth

100%|████████████████████████████████████████████████████████████████████████████████| 104M/104M [00:00<00:00, 113MB/s] Downloading: "https://github.com/xinntao/facexlib/releases/download/v0.2.2/parsing_parsenet.pth" to C:\Users\Ding-Yang\Real-ESRGAN\gfpgan\weights\parsing_parsenet.pth

100%|██████████████████████████████████████████████████████████████████████████████| 81.4M/81.4M [00:00<00:00, 112MB/s] Downloading: "https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.3.pth" to C:\Users\Ding-Yang\anaconda3\envs\intern\Lib\site-packages\gfpgan\weights\GFPGANv1.3.pth

100%|████████████████████████████████████████████████████████████████████████████████| 332M/332M [00:02<00:00, 117MB/s] Testing 0 0030 Error "compute_indices_weights_nearest" not implemented for 'Half' If you encounter CUDA out of memory, try to set --tile with a smaller number.

Dewey-Wang commented 8 months ago

The solution:

  1. https://github.com/xinntao/Real-ESRGAN/issues/586#issuecomment-1475380006 reintsall the torch, torchvision, and torchaudio. pip install torchaudio==2.0.1

  2. https://github.com/xinntao/Real-ESRGAN/issues/555#issuecomment-1705470424