{
  "id": 131517,
  "title": "Is it so easy to leak or overfit?",
  "url": "/competitions/google-cloud-ncaa-march-madness-2020-division-1-mens-tournament/discussion/131517",
  "author_name": "",
  "post_date": "2020-02-20T07:46:12.192720Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I just use some public kernel features mixed with mine and large the max_depth, the cv didn't improve as I expected, but the public lb improved into the range of deleting... <br>\nI think there is no leakage in my codes, so is it so easy to overfit? <br>\nPS: I didn't add ground truth to my submission. <br>\nsor for my poor expression =)  </p>",
  "messages": [
    {
      "id": "751448",
      "postDate": "02/20/2020 07:46:12",
      "content": "<p>I just use some public kernel features mixed with mine and large the max_depth, the cv didn't improve as I expected, but the public lb improved into the range of deleting... <br>\nI think there is no leakage in my codes, so is it so easy to overfit? <br>\nPS: I didn't add ground truth to my submission. <br>\nsor for my poor expression =)  </p>",
      "rawMarkdown": "I just use some public kernel features mixed with mine and large the max_depth, the cv didn't improve as I expected, but the public lb improved into the range of deleting...  \nI think there is no leakage in my codes, so is it so easy to overfit?  \nPS: I didn't add ground truth to my submission.  \nsor for my poor expression =)",
      "votes": null
    },
    {
      "id": "751499",
      "postDate": "02/20/2020 08:38:52",
      "content": "<p>Public LB is calculated based on 2015 - 2019 tourney data. But many public kernels use 1985 - 2019 tourney as training data. So if you add features or use larger max_depth or num_leaves, models will learn exact answer easier.</p>",
      "rawMarkdown": "Public LB is calculated based on 2015 - 2019 tourney data. But many public kernels use 1985 - 2019 tourney as training data. So if you add features or use larger max_depth or num_leaves, models will learn exact answer easier.",
      "votes": null
    },
    {
      "id": "751528",
      "postDate": "02/20/2020 08:57:01",
      "content": "<p>Thank you! I was too careless to realise it.</p>",
      "rawMarkdown": "Thank you! I was too careless to realise it.",
      "votes": null
    },
    {
      "id": "751770",
      "postDate": "02/20/2020 13:52:28",
      "content": "",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 751499,
      "author_name": "naka2ka",
      "author_url": "",
      "post_date": "02/20/2020 08:38:52",
      "content": "<p>Public LB is calculated based on 2015 - 2019 tourney data. But many public kernels use 1985 - 2019 tourney as training data. So if you add features or use larger max_depth or num_leaves, models will learn exact answer easier.</p>",
      "votes": null,
      "replies": [
        {
          "id": 751528,
          "author_name": "cnzengshiyuan",
          "author_url": "",
          "post_date": "02/20/2020 08:57:01",
          "content": "<p>Thank you! I was too careless to realise it.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 751770,
          "author_name": "yushg123",
          "author_url": "",
          "post_date": "02/20/2020 13:52:28",
          "content": "",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "751448": "I just use some public kernel features mixed with mine and large the max_depth, the cv didn't improve as I expected, but the public lb improved into the range of deleting...  \nI think there is no leakage in my codes, so is it so easy to overfit?  \nPS: I didn't add ground truth to my submission.  \nsor for my poor expression =)",
    "751499": "Public LB is calculated based on 2015 - 2019 tourney data. But many public kernels use 1985 - 2019 tourney as training data. So if you add features or use larger max_depth or num_leaves, models will learn exact answer easier.",
    "751528": "Thank you! I was too careless to realise it.",
    "751770": ""
  },
  "source": "meta"
}