{
  "id": 73990,
  "title": "Unable to merge datasets",
  "url": "/competitions/elo-merchant-category-recommendation/discussion/73990",
  "author_name": "",
  "post_date": "2018-12-07T10:09:41.877477400Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I think we have to merge historical and merchants data in chunks... Anyone facing problem while merging these two datasets in kernel?</p>",
  "messages": [
    {
      "id": "435007",
      "postDate": "12/07/2018 10:09:41",
      "content": "<p>I think we have to merge historical and merchants data in chunks... Anyone facing problem while merging these two datasets in kernel?</p>",
      "rawMarkdown": "I think we have to merge historical and merchants data in chunks... Anyone facing problem while merging these two datasets in kernel?",
      "votes": null
    },
    {
      "id": "435049",
      "postDate": "12/07/2018 11:48:02",
      "content": "<p>No buddy, everything works fine.</p>",
      "rawMarkdown": "No buddy, everything works fine.",
      "votes": null
    },
    {
      "id": "435099",
      "postDate": "12/07/2018 13:49:55",
      "content": "<p>Try to find out what are the most memory consuming columns by using <code>df.memory_usage(deep=True)</code>. You could try to convert date type, some of <code>object</code> type columns can be converted to <code>category</code></p>",
      "rawMarkdown": "Try to find out what are the most memory consuming columns by using `df.memory_usage(deep=True)`. You could try to convert date type, some of `object` type columns can be converted to `category`",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 435049,
      "author_name": "ashirwadsangwan",
      "author_url": "",
      "post_date": "12/07/2018 11:48:02",
      "content": "<p>No buddy, everything works fine.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 435099,
      "author_name": "lumaxy",
      "author_url": "",
      "post_date": "12/07/2018 13:49:55",
      "content": "<p>Try to find out what are the most memory consuming columns by using <code>df.memory_usage(deep=True)</code>. You could try to convert date type, some of <code>object</code> type columns can be converted to <code>category</code></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "435007": "I think we have to merge historical and merchants data in chunks... Anyone facing problem while merging these two datasets in kernel?",
    "435049": "No buddy, everything works fine.",
    "435099": "Try to find out what are the most memory consuming columns by using `df.memory_usage(deep=True)`. You could try to convert date type, some of `object` type columns can be converted to `category`"
  },
  "source": "meta"
}