{
  "id": 253294,
  "title": "Is the cumulative feature effective?",
  "url": "/competitions/mlb-player-digital-engagement-forecasting/discussion/253294",
  "author_name": "teyosan1229",
  "post_date": "2021-07-15T21:48:35.482000",
  "votes": 4,
  "comment_count": 5,
  "views": 0,
  "content": "<p>hello kaggler</p>\n<p>Is the cumulative feature effective?</p>\n<p>for example shohei ohtani</p>\n<table>\n<thead>\n<tr>\n<th>gameDate</th>\n<th>num of homerun</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>2021-04-02</td>\n<td>1</td>\n</tr>\n<tr>\n<td>2021-04-30</td>\n<td>7</td>\n</tr>\n<tr>\n<td>2021-07-09</td>\n<td>33</td>\n</tr>\n</tbody>\n</table>\n<p>33 home-runs means he's the home-run king.</p>\n<p>I'd like to know your thoughts on cumulative features in general.</p>\n<p>I'll try it.</p>\n<p>thank you</p>",
  "messages": [
    {
      "id": 1390244,
      "postDate": "2021-07-16T13:33:14.947Z",
      "content": "<p>I add some cumulative features in my model (such like hr, avg, era…) and it increase probably 0.3%~0.6% in LB and CV</p>",
      "rawMarkdown": "I add some cumulative features in my model (such like hr, avg, era...) and it increase probably 0.3%~0.6% in LB and CV",
      "votes": 3,
      "replies": [
        {
          "id": 1391712,
          "postDate": "2021-07-18T01:39:54.280Z",
          "content": "<p>Thanks for the interesting comments<br>\nIt makes me want to try it more and more!</p>",
          "rawMarkdown": "Thanks for the interesting comments\nIt makes me want to try it more and more!",
          "votes": 1
        }
      ]
    },
    {
      "id": 1389622,
      "postDate": "2021-07-15T21:48:35.483Z",
      "content": "<p>hello kaggler</p>\n<p>Is the cumulative feature effective?</p>\n<p>for example shohei ohtani</p>\n<table>\n<thead>\n<tr>\n<th>gameDate</th>\n<th>num of homerun</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>2021-04-02</td>\n<td>1</td>\n</tr>\n<tr>\n<td>2021-04-30</td>\n<td>7</td>\n</tr>\n<tr>\n<td>2021-07-09</td>\n<td>33</td>\n</tr>\n</tbody>\n</table>\n<p>33 home-runs means he's the home-run king.</p>\n<p>I'd like to know your thoughts on cumulative features in general.</p>\n<p>I'll try it.</p>\n<p>thank you</p>",
      "rawMarkdown": "hello kaggler\n\nIs the cumulative feature effective?\n\nfor example shohei ohtani\n\n| gameDate | num of homerun|\n| --- | --- |\n| 2021-04-02 | 1 |\n| 2021-04-30 | 7 |\n| 2021-07-09 | 33 |\n\n33 home-runs means he's the home-run king.\n\nI'd like to know your thoughts on cumulative features in general.\n\nI'll try it.\n\nthank you",
      "votes": 4
    },
    {
      "id": 1390099,
      "postDate": "2021-07-16T11:01:29.240Z",
      "content": "<p>Not much, but if you provide update during testing it may work.</p>",
      "rawMarkdown": "Not much, but if you provide update during testing it may work.",
      "votes": 2,
      "replies": [
        {
          "id": 1391711,
          "postDate": "2021-07-18T01:38:32.157Z",
          "content": "<p>Thank you for your comment.<br>\nLike you said, responding to test data is a must.</p>",
          "rawMarkdown": "Thank you for your comment.\nLike you said, responding to test data is a must."
        }
      ]
    },
    {
      "id": 1390237,
      "postDate": "2021-07-16T13:20:36.570Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 1390244,
      "author_name": "joshspchang",
      "author_url": "",
      "post_date": "2021-07-16T13:33:14.947000",
      "content": "<p>I add some cumulative features in my model (such like hr, avg, era…) and it increase probably 0.3%~0.6% in LB and CV</p>",
      "votes": 3,
      "replies": [
        {
          "id": 1391712,
          "author_name": "teyosan1229",
          "author_url": "",
          "post_date": "2021-07-18T01:39:54.280000",
          "content": "<p>Thanks for the interesting comments<br>\nIt makes me want to try it more and more!</p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 1390099,
      "author_name": "lego_king",
      "author_url": "",
      "post_date": "2021-07-16T11:01:29.240000",
      "content": "<p>Not much, but if you provide update during testing it may work.</p>",
      "votes": 2,
      "replies": [
        {
          "id": 1391711,
          "author_name": "teyosan1229",
          "author_url": "",
          "post_date": "2021-07-18T01:38:32.157000",
          "content": "<p>Thank you for your comment.<br>\nLike you said, responding to test data is a must.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1390237,
      "author_name": "",
      "author_url": "",
      "post_date": "2021-07-16T13:20:36.570000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1390244": "I add some cumulative features in my model (such like hr, avg, era...) and it increase probably 0.3%~0.6% in LB and CV",
    "1389622": "hello kaggler\n\nIs the cumulative feature effective?\n\nfor example shohei ohtani\n\n| gameDate | num of homerun|\n| --- | --- |\n| 2021-04-02 | 1 |\n| 2021-04-30 | 7 |\n| 2021-07-09 | 33 |\n\n33 home-runs means he's the home-run king.\n\nI'd like to know your thoughts on cumulative features in general.\n\nI'll try it.\n\nthank you",
    "1390099": "Not much, but if you provide update during testing it may work.",
    "1390237": ""
  }
}