{
  "id": 167359,
  "title": "What is the best score you can achieve using only tabular data?",
  "url": "/competitions/siim-isic-melanoma-classification/discussion/167359",
  "author_name": "lixiangyang",
  "post_date": "2020-07-16T07:12:23.701000",
  "votes": 4,
  "comment_count": 5,
  "views": 0,
  "content": "<p>In the later stage, should we pay more attention to table data or picture data to improve scores? What is the highest score you can get with tabular data?</p>",
  "messages": [
    {
      "id": 931905,
      "postDate": "2020-07-16T14:47:20.453Z",
      "content": "<p>Take attention to Image and blend with tabular data</p>",
      "rawMarkdown": "Take attention to Image and blend with tabular data",
      "votes": 1
    },
    {
      "id": 931420,
      "postDate": "2020-07-16T07:12:23.700Z",
      "content": "<p>In the later stage, should we pay more attention to table data or picture data to improve scores? What is the highest score you can get with tabular data?</p>",
      "rawMarkdown": "In the later stage, should we pay more attention to table data or picture data to improve scores? What is the highest score you can get with tabular data?",
      "votes": 2
    },
    {
      "id": 931502,
      "postDate": "2020-07-16T08:33:50.097Z",
      "rawMarkdown": "",
      "isDeleted": true,
      "replies": [
        {
          "id": 931570,
          "postDate": "2020-07-16T09:18:05.093Z",
          "content": "<p>You can explore last year winners' <a href=\"https://arxiv.org/pdf/1910.03910.pdf\">paper</a> and find out that Meta Data makes some reasonable influence with smaller models, when with bigger ones it can give only marginal boost</p>",
          "rawMarkdown": "You can explore last year winners' [paper](https://arxiv.org/pdf/1910.03910.pdf) and find out that Meta Data makes some reasonable influence with smaller models, when with bigger ones it can give only marginal boost",
          "votes": 2
        },
        {
          "id": 931575,
          "postDate": "2020-07-16T09:20:11.047Z",
          "rawMarkdown": "",
          "isDeleted": true
        }
      ]
    },
    {
      "id": 931466,
      "postDate": "2020-07-16T08:01:59.773Z",
      "rawMarkdown": "",
      "votes": -1,
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 931905,
      "author_name": "Manh Lab",
      "author_url": "",
      "post_date": "2020-07-16T14:47:20.453000",
      "content": "<p>Take attention to Image and blend with tabular data</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 931502,
      "author_name": "",
      "author_url": "",
      "post_date": "2020-07-16T08:33:50.097000",
      "content": "",
      "votes": 0,
      "replies": [
        {
          "id": 931570,
          "author_name": "Volodymyr",
          "author_url": "",
          "post_date": "2020-07-16T09:18:05.093000",
          "content": "<p>You can explore last year winners' <a href=\"https://arxiv.org/pdf/1910.03910.pdf\">paper</a> and find out that Meta Data makes some reasonable influence with smaller models, when with bigger ones it can give only marginal boost</p>",
          "votes": 2,
          "replies": []
        },
        {
          "id": 931575,
          "author_name": "",
          "author_url": "",
          "post_date": "2020-07-16T09:20:11.047000",
          "content": "",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 931466,
      "author_name": "",
      "author_url": "",
      "post_date": "2020-07-16T08:01:59.773000",
      "content": "",
      "votes": -1,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "931905": "Take attention to Image and blend with tabular data",
    "931420": "In the later stage, should we pay more attention to table data or picture data to improve scores? What is the highest score you can get with tabular data?",
    "931502": "",
    "931466": ""
  }
}