Training error report

[hhhddddddd]

Great job!
I tried to test the model effect on a custom data set, and now I have some problems.

The instructions I carried out are as follows:
python run_video.py \ -i /home/xuankai/code/Car.mp4 \ -o output/ \ -p paper \ --res 0 \ --fps 0 \ --restore-res True
I'm not sure if there is something wrong with my data format. Looking forward to your reply!

[markkua]

hi, the paper preset uses strictly dilation rates {1, 10, 25}. Your video has only 17 frames, which is too short for this setting. Please try fast preset, which will automatically reduce the dilations, or manually specify other dilation rates.

[hhhddddddd]

Thank you very much for your reply!
When I use A to control the output resolution, the output depth video will be noisy. How can I solve this problem?

The instructions I carried out are as follows:
python run_video.py \ -i /home/xuankai/code/Car.mp4 \ -o output \ -p fast \ --verbose \ --restore-res True \

[markkua]

What does your input video look like? Does it also fail with --restore-res False?

[hhhddddddd]

Thank you very much for your reply!
The resolution of the video I inputted is 400*940. It is running correctly with --restore-res False. But I want to keep the output resolution unchanged.

[markkua]

Could you share your input video? I can't reproduce this issue with my data.

[hhhddddddd]

https://drive.google.com/file/d/1fIBoY3RKw_peuLiXHaLDjAAuUkCZOaR1/view?usp=drive_link