{
  "fetched_at": "2026-10-24T16:55:26Z",
  "source_url": "https://www.kaggle.com/competitions/graphadon-2/data",
  "pages": [
    {
      "name": "data-description",
      "content": "You have two files saved as torch data file. Each file contains contains a dataset, aka a list of graphs described by their structures and features. \n\nThe train set comes along the y values to be able to learn your model. The test set has not longer the y values, and you have to predict it by fair means. \n\n## Files\n\n*   **train_dataset.pt** - the training set with y values\n*   **test_dataset.pt** - the test set, only the graphs\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Dataset description",
      "content": "The dataset is quite similar to the MUTAG introduced in the practical session, but with more data. Take care to first check the dimensionality of your node features to adapt your GNN.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": " - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "Similarly to the first GNN you implemented on MUTAG, learn a GNN model on the train dataset, predict on the test set and submit your results !  ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Your submission will be evaluated according to the accuracy of your performance. Submit a file using the `generate_kaggle_file` function. It should produce a valid text file in a csv format to be submitted.\n\nTake care ! To avoid training on the test set, you only have **10 possible submissions** !\n\n## Submission File\nFor each ID in the test set, you must predict a probability for the TARGET variable. The file should contain a header and have the following format:\n\n    ID,TARGET\n    2,0\n    5,0\n    6,0\n    etc.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}