RuntimeError: cuDNN error: CUDNN_STATUS_NOT_SUPPORTED. This error may appear if you passed in a non-contiguous input.

Mukosame / Zooming-Slow-Mo-CVPR-2020

Fast and Accurate One-Stage Space-Time Video Super-Resolution (accepted in CVPR 2020)

GNU General Public License v3.0

915 stars 164 forks source link

RuntimeError: cuDNN error: CUDNN_STATUS_NOT_SUPPORTED. This error may appear if you passed in a non-contiguous input. #9

Open hiredd opened 4 years ago

hiredd commented 4 years ago

test using zsm_my_video.sh get error ！

Mukosame commented 4 years ago

Hi, thanks for bringing this up. Can you provide more details about your input and the running environment?

hiredd commented 4 years ago

Hi, thanks for bringing this up. Can you provide more details about your input and the running environment?

python video_to_zsm.py --video video.MP4 --model ../experiments/pretrainedmodels/xiang2020zooming.pth --fps 50 --output out.mp4 environment: pytorch1.2 CUDA 10.0

EAGLE50 commented 4 years ago

me too! firstly, I have the issue #8 ,then I changed the parameter of N_out=3. Unfortunately,I have this error now.

tianxiaguixin002 commented 3 years ago

when I change N_out=3, it can run normallly, input size = 480*856, my GPU is V100, code costs 19.5GB GPU memory.