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

Evaluator.py map issue with dataset #252

Open
D10752002 opened this issue Jul 7, 2022 · 5 comments
Open

Evaluator.py map issue with dataset #252

D10752002 opened this issue Jul 7, 2022 · 5 comments

Comments

@D10752002
Copy link

I have been able to train the model successfully using a custom dataset, but unable to calculate map with my custom dataset.
evaluator.py code has map calculation using a dataset with names as 0.jpg , 1.jpg , 2, ... so on. the custom dataset doesn't have such naming. Is there some way to get around this issue?

@Rvv1296
Copy link

Rvv1296 commented Jul 7, 2022

Hello @D10752002 ,
I am trying to replicate this repo.
Can you please give me steps how to train my model.
I am unable to complete the training step for VOC data set as well as for custom data set.
Thank you

@D10752002
Copy link
Author

Hi, I'm modifying this model to train a custom dataset, and I'm using colab, instead of importing multiple files, I had to define all those in different cells, and since I'm running it in colab, I'm not using parsers, and I had to modify many lines of the code for my model. So, I'm not sure where you might be facing errors in the code. The training code was working fine except for the map evaluator part since I'm using a custom dataset. Sorry if I could not be of more help.

@Rvv1296
Copy link

Rvv1296 commented Jul 12, 2022

Hello @D10752002,
For custom data set have you set data type as a Customer in cfg file ?

@D10752002
Copy link
Author

@Rvv1296 yes I have set it as customer

@argusswift
Copy link
Owner

@Rvv1296 yes I have set it as customer

well done

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

3 participants