{
  "fetched_at": "2026-10-24T16:48:11Z",
  "source_url": "https://www.kaggle.com/competitions/fashion-mnist-itba-2019/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": "",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Fashion-MNIST is a dataset that is meant to be a direct replacement of the overused MNIST dataset. It is first introduced by Han Xiao and Kashif Rasul and Roland Vollgraf in their paper Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms.\n\nEach data point in the Fashion-MNIST dataset is a 28x28 image flattened out as a vector of length 784 (just like MNIST). If you perform reshape to each vector and visualize the 28x28 image, you will see that the images are one of the following 10 items\n\nT-shirt/top  \nTrouser  \nPullover  \nDress  \nCoat  \nSandal  \nShirt  \nSneaker  \nBag  \nAnkle boot  \n\nThe above 10 items are labelled with integers from 0 to 9 respectively. \nThe objective of the task is to come up with an algorithm to assign a label to a given image (a given 784 length vector) that matches with the true label.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "As a classification task, the evaluation of the submissions will be based on the percentage of correctly classified samples.  \n\nTo make a submission, you need to write out your predicted labels for each on the data point in the test_image matrix as a CSV file, in the following format:   \nId,Category  \n0,2  \n1,3  \n2,9  \nNote that the first row is the header. The first column is the index of the data point, which should be 0-4999 since we have 5000 test samples. The second column is the predicted label.  ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}