GTSRB-CNN

• Train Loss 0.0958, Train Accuracy 0.9702, Val Loss 0.3132, Val Accuracy 0.9222

Dataset

• GTSRB: You could download it either from the official GTSRB website, or Kaggle.

  https://benchmark.ini.rub.de/

  https://www.kaggle.com/datasets/meowmeowmeowmeowmeow/gtsrb-german-traffic-sign

• If you're using google colab, then download the dataset as a zipped file and upload it to your google drive.

• If you're using Kaggle, then you don't need to download the dataset.

  Here's a link to the notebook on Kaggle: https://www.kaggle.com/code/remyhos/gtsrb

Future work

• I used the traditional CNNs, however, I think using ViT or GRU-based classifier will improve the performance.

Contributing

Contributions are welcome! If you find any bugs or have any suggestions, please start an issue or contact the author.

If you would like to contribute code to this project, please follow these steps:

1. Fork this repository.
2. Create a new branch with a descriptive name (git checkout -b my_new_feature).
3. Write your code and tests, ensuring that they pass.
4. Push your changes to your fork (git push origin my_new_feature).
5. Open a pull request and describe your changes.

Thank you for your contributions!

Credit

• This was the original code: https://www.kaggle.com/code/georgesebastianson/alternative-gtsrb