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

Running the evaluation code #19

Open
Hmz30265 opened this issue Aug 17, 2023 · 4 comments
Open

Running the evaluation code #19

Hmz30265 opened this issue Aug 17, 2023 · 4 comments

Comments

@Hmz30265
Copy link

Hello authors, thank you for the contribution.
I am currently making a simple simulation of the evaluation code so I could understand how the evaluation process is working. Right now, I have created couple dummy points and I can get IoUs and IoU matches. When I tried to run this code :

def compute_ap_and_acc(pred_matches, gt_matches):

It gives me this error message:

errorshAPO

I don't know how you do you explain this problem. Could you help me with that? For your information, my two parameters look like this:

errorshapo2

Please let me know if you could help me with that. I am looking forward to hearing back from you. Thank you!

@Hmz30265
Copy link
Author

Also, I am trying to run the colab you made, which will be super helpful. However, I encountered some version issues.
error
Is there anyway to fix those?

@zubair-irshad
Copy link
Owner

Thanks, @MattMiaozhuangHe for your interest in our work. I would suggest you go through this issue in our accompanying repo, CenterSnap where evaluation is discussed in detail. I would try to take a pre-trained checkpoint and produce outputs and feed that into the function to get a better sense of evaluation. Dummy variables might not work and could have their own problems like class mismatch etc.

Regarding the other issue, I would manually install torch 1.7.1+cu101 from here in your notebook and install the rest of the dependencies using requirements.txt You might have to comment out that line which installs pytorch in the requirements.txt and point the notebook to install requirements.txt from the local repo.

Hope it helps!

@Hmz30265 Hmz30265 reopened this Aug 21, 2023
@Hmz30265
Copy link
Author

Thank you for answering my questions @zubair-irshad , I have successfully download torch1.7.1+cu101 and that error message does not show again. However, I believe a bigger problem is that the Colab only support Python 3.10.

Python version

And it can't be fallback to earlier version of Python, which is in the required in this notebook:

ERROR: Ignored the following versions that require a different python version: 1.6.2 Requires-Python >=3.7,<3.10; 1.6.3 Requires-Python >=3.7,<3.10; 1.7.0 Requires-Python >=3.7,<3.10; 1.7.1 Requires-Python >=3.7,<3.10

Is there anyway I can fix this problem? Looking forward to hearing back from you!

@zubair-irshad
Copy link
Owner

Apologies for a delayed response as we were busy on our end with other code releases/paper deadlines. Were you able to solve this issue. Does adding python==3.9 in the requirements.txt make it work for you?

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