{
  "id": 58346,
  "title": "How use activate_train and activate_test?",
  "url": "/competitions/avito-demand-prediction/discussion/58346",
  "author_name": "",
  "post_date": "2018-06-06T12:10:15.000361700Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>By using this data.\nI think that grasp the tendency to be displayed.</p>\n\n<p>How do you guys think?</p>",
  "messages": [
    {
      "id": "339151",
      "postDate": "06/06/2018 12:10:15",
      "content": "<p>By using this data.\nI think that grasp the tendency to be displayed.</p>\n\n<p>How do you guys think?</p>",
      "rawMarkdown": "By using this data.\nI think that grasp the tendency to be displayed.\n\nHow do you guys think?",
      "votes": null
    },
    {
      "id": "339233",
      "postDate": "06/06/2018 14:55:48",
      "content": "<p>You can create groupby features and merge them on user_id.</p>\n\n<p>E.g. you can count the amount of ads per user.\nSee this kernel for an example: <a href=\"https://www.kaggle.com/bminixhofer/aggregated-features-lightgbm\">https://www.kaggle.com/bminixhofer/aggregated-features-lightgbm</a></p>",
      "rawMarkdown": "You can create groupby features and merge them on user_id.\n\nE.g. you can count the amount of ads per user.\nSee this kernel for an example: https://www.kaggle.com/bminixhofer/aggregated-features-lightgbm",
      "votes": null
    },
    {
      "id": "339471",
      "postDate": "06/07/2018 02:26:40",
      "content": "<p>This is very good kernel!\nThank you for telling me!</p>",
      "rawMarkdown": "This is very good kernel!\nThank you for telling me!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 339233,
      "author_name": "kevinbonnes",
      "author_url": "",
      "post_date": "06/06/2018 14:55:48",
      "content": "<p>You can create groupby features and merge them on user_id.</p>\n\n<p>E.g. you can count the amount of ads per user.\nSee this kernel for an example: <a href=\"https://www.kaggle.com/bminixhofer/aggregated-features-lightgbm\">https://www.kaggle.com/bminixhofer/aggregated-features-lightgbm</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 339471,
          "author_name": "seiya1998",
          "author_url": "",
          "post_date": "06/07/2018 02:26:40",
          "content": "<p>This is very good kernel!\nThank you for telling me!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "339151": "By using this data.\nI think that grasp the tendency to be displayed.\n\nHow do you guys think?",
    "339233": "You can create groupby features and merge them on user_id.\n\nE.g. you can count the amount of ads per user.\nSee this kernel for an example: https://www.kaggle.com/bminixhofer/aggregated-features-lightgbm",
    "339471": "This is very good kernel!\nThank you for telling me!"
  },
  "source": "meta"
}