{
  "fetched_at": "2026-10-22T12:49:41Z",
  "source_url": "https://www.kaggle.com/competitions/artist-identification/data",
  "pages": [
    {
      "name": "data-description",
      "content": "## Files\n\n*   **artist_dataset.zip** - Contains the train and test directories\n*   **train.csv** - Contains the paths to each training image as well as their classes\n*   **sample_submission.csv** - a sample submission file in the correct format",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Your goal is to design a model capable of taking an image of any size and classify it by the artist who created it. The competition will be open to submissions until Tuesday, October 22nd at 5:00 AM. After the competition ends, the three highest ranked teams in accuracy will be invited to present their work. \n\nThe goal of this competition is to evaluate the way the models go about classifying these paintings. Moreover, the cash prizes will be distributed to the team that gives the best presentation, demonstrating they know why the model gave the results that it did. However, accuracy will also be considered when determining the winning team. ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Submissions will be in the form of a .csv file. The first column should contain a list of all the image names in the test set. The remaining columns should contain your model's top five predicted classes with the most confident in the left and the least confident in the right. The columns should be named exactly what they are in the sample_submission.csv headers. \n\n## Example Submission File:\n    Image Name,Class1,Class2,Class3,Class4,Class5\n    \"ccc83541.jpg\",giovanni-boldini,leonardo-da-vinci,caravaggio,fyodor-vasilyev,george-morland\n    \"5adf0be9.jpg\",andrea-del-sarto,giorgio-vasari,thomas-gainsborough,louise-elisabeth-vigee-le-brun,ivan-vladimirov\n    \"28d1c9fd.jpg\",ivan-kramskoy,paolo-uccello,alfred-stevens,andrea-del-sarto,william-hogarth\n\nYour submissions will be evaluated using a custom scoring scheme that balances accuracy with fair representation of all artists. For each painting in the test set, your model's prediction will receive a score based on its accuracy: 1 point for correct prediction (top 1), 1/2 point for top 2, 1/3 point for top 3, 1/4 point for top 4, 1/5 point for top 5, and 0 points if not in the top 5. An average score will be calculated for each artist across their paintings. The final score will be the average of all these artist-specific scores, ensuring that each artist contributes equally to the total score, regardless of how many of their works are in the test set. This method incentivizes models to perform well across all artists, not just the most common ones.\n\nApproximately half of the training instances will be made private. The leaderboard will only evaluate on public data during the competition. The final rankings will be determined based on performance on the full dataset, including both public and private instances. This makes things a little more interesting as you won't be sure who truly is winning.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This hackathon is run through the BYU AI Association. Any participants wanting to compete for prizes must be willing to give a brief presentation on their work on October 23rd at 7:00. The top three teams will provide a repository validating the authenticity of their presentations.\n\nTeam sizes should be 2-4 with some exceptions allowed to assure that everyone who wants to participate can participate.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "Welcome to the Artist Classification Challenge! In this competition, you'll dive into the world of early European art, using machine learning to identify the creators behind various paintings. But there's a twist – we're not just looking for accuracy. We want you to derive insights on what features are useful for distinguishing artists works from each other, ultimately being able to use your model to determine more about an artist's style compared to that of other artists. ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Dataset",
      "content": "The dataset used in this competition is derived from the work described in the following paper:\n\nTan, W. R., Chan, C. S., Aguirre, H., & Tanaka, K. (2019). Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork. IEEE Transactions on Image Processing, 28(1), 394-409. https://doi.org/10.1109/TIP.2018.2866698\n\nWe do not claim this dataset as our own. Our usage is purely for research purposes only.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}