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

How to use custom datasets to test the performance of a model? #58

Open
weekiiy opened this issue Apr 1, 2024 · 4 comments
Open

How to use custom datasets to test the performance of a model? #58

weekiiy opened this issue Apr 1, 2024 · 4 comments

Comments

@weekiiy
Copy link

weekiiy commented Apr 1, 2024

How to use custom datasets to test the performance of a model? I did not see the test module in the code.I want to do trajectory prediction on my own dataset to test the performance of the model, but I couldn't find the testing module in the code. How to use your own dataset for testing and prediction?

@GentleSmile
Copy link
Member

The testing module loads the dataset here. You need to modify the dataset to your own format.

@weekiiy
Copy link
Author

weekiiy commented Apr 8, 2024

Thank you for your reply. If I want to modify the map information that matches the training set and the validation set, how to do so and how to modify it.

@GentleSmile
Copy link
Member

Map information is converted into vectors during training and inference. You need to modify map processing to support your own map format.

@weekiiy
Copy link
Author

weekiiy commented Apr 24, 2024

Thank you for your reply, which effectively solved my above problems. I have the following questions

  1. If I want to apply this code for trajectory prediction in a certain area, there are already vehicle trajectory data sets in this area, but the editing of the map becomes difficult. If it's a 20m square of empty space, how do we build this simple map and apply it to our algorithm?
  2. What if I only want to use my own map data during testing?
  3. What if I want to change the density of the target point?
    This model is really excellent, but I did encounter the above problems in application, thank you for your patience to answer

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