{
  "fetched_at": "2026-10-19T08:35:58Z",
  "source_url": "https://www.kaggle.com/competitions/acc-german-traffic-sign-classification/data",
  "pages": [
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "<p>German traffic sign dataset contains about 39000 training images and 12000 testing images of 43 classes with class id ranging from 0 to 43. Participant joining this competition will try to predict a class id for each testing image in test/ folder and submit only 1 csv file called: submission.csv.</p>\n\n<h2>File descriptions</h2>\n<ul>\n<li><strong>train/</strong> - the folder containing 43 sub-folders of same category images</li>\n<li><strong>test/</strong> - the folder containing testing images</li>\n<li><strong>sample_submission.csv</strong> - a sample submission file in the correct format</li>\n</ul>\n\n<h2>Data fields of submission file (submission.csv)</h2>\n<ul>\n<li><strong>Filename</strong> - (string) name of testing image</li>\n<li><strong>ClassId</strong> - (int) categorical id</li>\n</ul>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p>This is an in-class competition of FPT University Autonomous Car Club. We hope to provide to student a playground to apply what they have learned in practice. About this classification problem, we will try to predict categorical id of German traffic sign images. To get your score in leader board, you just create a kernel, train models on it and output a csv file called: submission.csv - the only file that is accepted to compute the score</p>\n\n<h2>About Autonomous Car Club</h2>\n<p>Our mission is to apply newest technologies such as Machine learning, Deep learning, IOT ion the work of building an autonomous system for moving vehicles. Visiting github of autocar club: </p>\n<pre>https://github.com/autocarclub/ACC</pre>\n<p>Or have a quick glance at us on Facebook: </p>\n<pre>https://www.facebook.com/Autonomous-Car-Club-1538400206286964/</pre>\n\n<h2>Acknowledgements</h2>\n<p>This dataset is free to download from <a href=\"http://benchmark.ini.rub.de/?section=gtsrb&amp;subsection=news\">this site</a>. The test set is also taken from there but we change the image name in order that participants can trick by getting ground truth provided from that site</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<h2>Metrics</h2>\n<p>The evaluation metric for this competition is Categorical Accuracy, which is given by the equation:</p>\n<p>\\[ accuracy = \\frac{truePredictions}{totalPredictions}  \\] </p>\n\n<h2>Submission Format</h2>\n<p><strong>For every author in the dataset</strong>, submission files should contain two columns: Filename and ClassId. ClassId should be an integer representing categorical id of corresponding image</p>\n<p>The file should contain a header and have the following format:</p>\n<pre>Filename,ClassId<br>543467.2353644.ppm,1<br></pre>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}