{
  "fetched_at": "2026-10-20T06:50:35Z",
  "source_url": "https://www.kaggle.com/competitions/funml-sp24-hw7-problem-5/data",
  "pages": [
    {
      "name": "data-description",
      "content": "You can access the data through the assignment notebook",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Problem 5: CURE-TSR Classification Competition on Kaggle (20 pts, Bonus)\n\nIn this problem, you need to choose and design your own training strategy to achieve better top-1 accuracy in a Kaggle competition.\n\n- You should start with one of pretrained alexnet or vgg16 or resnet18.\n\n- You should only use the train data for training your network. You can use any loss function, optimizer, regularization, data augumentation strategy or modify model structure to help youself get better test results.\n\n- You should explain your strategy and desgin in detail.\n\n- Use the cell provided below to submit test result as a .csv file in Kaggle.\n\n- In Kaggle, 50% of your result will be used to calculate the top-1 accuracy for the score on the leaderboard. The rest of 50% will be used to calculate your actual score after the deadline. This means the score on the leaderboard only gives you a sense of how you perform, and it will not be your final score.\n\n- You have 10 times to submit result in Kaggle every day. Please use it wisely.\n\n- 10 pts will be according to your strategy design and the rest of 10 pts depends on your ranking.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "-",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "-",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}