{
  "id": 78752,
  "title": "Smaller Historical Transactions DataSet",
  "url": "/competitions/elo-merchant-category-recommendation/discussion/78752",
  "author_name": "",
  "post_date": "2019-01-27T15:16:10.135776900Z",
  "votes": 5,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hello everyone,</p>\n\n<p>As you might have noticed, the historical transactions dataset is quite big when loaded into memory without any type casting. </p>\n\n<p>Thus, I decided to create two contributions: </p>\n\n<ol>\n<li>An EDA kernel where I explain how to get the dataset to a smaller\nfootprint. You can check it here: <a href=\"https://www.kaggle.com/yassinealouini/how-small-could-it-get\">https://www.kaggle.com/yassinealouini/how-small-could-it-get</a> </li>\n<li>A script where I have only put the optimization pipeline. You can find it here: <a href=\"https://www.kaggle.com/yassinealouini/small-historical-transactions-parquet\">https://www.kaggle.com/yassinealouini/small-historical-transactions-parquet</a></li>\n</ol>\n\n<p>Let me know your thoughts in the comments below. Thanks in advance. :) </p>",
  "messages": [
    {
      "id": "462059",
      "postDate": "01/27/2019 15:16:10",
      "content": "<p>Hello everyone,</p>\n\n<p>As you might have noticed, the historical transactions dataset is quite big when loaded into memory without any type casting. </p>\n\n<p>Thus, I decided to create two contributions: </p>\n\n<ol>\n<li>An EDA kernel where I explain how to get the dataset to a smaller\nfootprint. You can check it here: <a href=\"https://www.kaggle.com/yassinealouini/how-small-could-it-get\">https://www.kaggle.com/yassinealouini/how-small-could-it-get</a> </li>\n<li>A script where I have only put the optimization pipeline. You can find it here: <a href=\"https://www.kaggle.com/yassinealouini/small-historical-transactions-parquet\">https://www.kaggle.com/yassinealouini/small-historical-transactions-parquet</a></li>\n</ol>\n\n<p>Let me know your thoughts in the comments below. Thanks in advance. :) </p>",
      "rawMarkdown": "Hello everyone,\n\nAs you might have noticed, the historical transactions dataset is quite big when loaded into memory without any type casting. \n\nThus, I decided to create two contributions: \n\n 1. An EDA kernel where I explain how to get the dataset to a smaller\n   footprint. You can check it here: https://www.kaggle.com/yassinealouini/how-small-could-it-get \n 2. A script where I have only put the optimization pipeline. You can find it here: https://www.kaggle.com/yassinealouini/small-historical-transactions-parquet\n\nLet me know your thoughts in the comments below. Thanks in advance. :)",
      "votes": null
    },
    {
      "id": "466330",
      "postDate": "02/05/2019 06:18:47",
      "content": "<p>Hey, \nNice Work!</p>",
      "rawMarkdown": "Hey, \nNice Work!",
      "votes": null
    },
    {
      "id": "466429",
      "postDate": "02/05/2019 11:09:31",
      "content": "<p>Thanks. 🙂 </p>",
      "rawMarkdown": "Thanks. 🙂",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 466330,
      "author_name": "ibtesama",
      "author_url": "",
      "post_date": "02/05/2019 06:18:47",
      "content": "<p>Hey, \nNice Work!</p>",
      "votes": null,
      "replies": [
        {
          "id": 466429,
          "author_name": "yassinealouini",
          "author_url": "",
          "post_date": "02/05/2019 11:09:31",
          "content": "<p>Thanks. 🙂 </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "462059": "Hello everyone,\n\nAs you might have noticed, the historical transactions dataset is quite big when loaded into memory without any type casting. \n\nThus, I decided to create two contributions: \n\n 1. An EDA kernel where I explain how to get the dataset to a smaller\n   footprint. You can check it here: https://www.kaggle.com/yassinealouini/how-small-could-it-get \n 2. A script where I have only put the optimization pipeline. You can find it here: https://www.kaggle.com/yassinealouini/small-historical-transactions-parquet\n\nLet me know your thoughts in the comments below. Thanks in advance. :)",
    "466330": "Hey, \nNice Work!",
    "466429": "Thanks. 🙂"
  },
  "source": "meta"
}