{
  "fetched_at": "2026-10-19T07:51:13Z",
  "source_url": "https://www.kaggle.com/competitions/dog-vs-cat-classification/data",
  "pages": [
    {
      "name": "data-description",
      "content": "# Dataset\n\n* Train: including 25.000 images which are separated in the two folders dog and cat relevant to each class. We need to use that to train the model.\n* Test: including 8.000 hidden-label images to evaluate and rank leader board.\n* submission.csv: sample submission files which is the result of the model on the test dataset. That include two columns: \n * ID: are images name (such as 000001.jpg)\n * label: the model's prediction",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "# Introduction\nThe internet boom leads to appear of many amazing AI applications that can replace the human in the recognition task. In this competition, you are supplied with the dog and cat images and try to train a computer vision model that can recognize a new image belonging to what class.\n\nThe competition is also a good chance for you to practice and compare the different SOTA backbones of Computer Vision such as:\n\n* Transformer: ViT, DeiT, CoAtNet, SwinTransformer, MobileViT\n* CNN: EfficientNet, ConvNeXt, ResNet, ResNext, MobileNet, AlexNet, InceptionNet, DenseNet,...\n* Other architectures: MLP-Mixer, PatchConvNet, Neural Architecture Search,... ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "# Evaluation\nAccuracy is a metric for that competition. 30 percent of the test dataset are used as a public score showing in the leader board and the remaining 70 percent are kept as private score. The final result = 0.3 * public score + 0.7 * private score",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "# Timeline\n1 month from the start time\n\n# Rule\n- You can submit a maximum of 5 times per day during the competition time.\n- You can only use the pre-trained models from ImageNet dataset. \n- Do not accept to use of the external dataset\n- Team size can be up to 5 members\n- If two people have the same score, the smaller model (based on the number of parameters) will be considered a higher rank. The ensemble model will be considered the total parameters as the sum of all member models.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}