Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MTI-NET NYUD task,some files lost #7

Closed
TianhaoFu opened this issue Dec 10, 2020 · 3 comments
Closed

MTI-NET NYUD task,some files lost #7

TianhaoFu opened this issue Dec 10, 2020 · 3 comments

Comments

@TianhaoFu
Copy link

TianhaoFu commented Dec 10, 2020

Hi,I'm running your code,but i find some files is not included in your respository:

FileNotFoundError: [Errno 2] No such file or directory: './results/NYUD/hrnet_w18/single_task/semseg/re
sults/NYUD_test_semseg.json'

could you please provide these files?
thanks a lot!

@TianhaoFu TianhaoFu changed the title MIT-NET NTUD ,some files lost MIT-NET NYUD task,some files lost Dec 10, 2020
@TianhaoFu TianhaoFu changed the title MIT-NET NYUD task,some files lost MTI-NET NYUD task,some files lost Dec 10, 2020
@SimonVandenhende
Copy link
Owner

Could you specify what script you are trying to run exactly?

@TianhaoFu
Copy link
Author

thanks for your reply. I'm using python main.py --config_env configs/env.yml --config_exp configs/nyud/hrnet18/mti_net.yml, and i came across that problem.

@SimonVandenhende
Copy link
Owner

The multi-task learning performance is calculated using Equation 10 from https://arxiv.org/pdf/2004.13379.pdf
Note that this requires the performance of a set of single-tasking baselines,
therefore in order to determine the multi-task learning performance for MTI-Net, it requires to train the single-tasking models first.
You could still run the code, if you decide to only report segmentation and depth, and ignore the multi-tasking performance for MTI-Net.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants