{
  "id": 365290,
  "title": "Sample Submission Questions",
  "url": "/competitions/otto-recommender-system/discussion/365290",
  "author_name": "",
  "post_date": "2022-11-10T15:49:06.637027100Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Can someone explain why the structure of the sample submission. I dont understand what the unique value is and why its always the same. Also why is the event given even though we have to predict it.<br>\nThanks in advance!</p>",
  "messages": [
    {
      "id": "2024560",
      "postDate": "11/10/2022 15:49:06",
      "content": "<p>Can someone explain why the structure of the sample submission. I dont understand what the unique value is and why its always the same. Also why is the event given even though we have to predict it.<br>\nThanks in advance!</p>",
      "rawMarkdown": "Can someone explain why the structure of the sample submission. I dont understand what the unique value is and why its always the same. Also why is the event given even though we have to predict it.\nThanks in advance!",
      "votes": null
    },
    {
      "id": "2024663",
      "postDate": "11/10/2022 17:06:29",
      "content": "<p>We are making <code>1,671,803</code> times <code>3</code> lists of predictions. There are <code>1,671,803</code> sessions. In this competition, a \"session\" means \"user\". So there are <code>1,671,803</code> users. The test period is one week long. For each user, we need to predict what they will <code>click</code>, <code>cart</code> and <code>order</code> during the remainder of the test period (after the information we are provided about each user). The order of the 20 numbers within our lists does not matter.</p>\n<p>So for user \"12899779\", we add 3 rows to <code>submission.csv</code>. Where the second column contains a list of our 20 predictions. The metric is <code>recall@20</code>. </p>\n<pre><code>column 1                    column 2\n12899779_clicks  [51, 22, 73, 84, 95, ...]\n12899779_carts   [8, 23, 12, 41, 52, ...]\n12899779_orders  [11, 21, 37, 40, 60, ...]\n</code></pre>",
      "rawMarkdown": "We are making `1,671,803` times `3` lists of predictions. There are `1,671,803` sessions. In this competition, a \"session\" means \"user\". So there are `1,671,803` users. The test period is one week long. For each user, we need to predict what they will `click`, `cart` and `order` during the remainder of the test period (after the information we are provided about each user). The order of the 20 numbers within our lists does not matter.\n\nSo for user \"12899779\", we add 3 rows to `submission.csv`. Where the second column contains a list of our 20 predictions. The metric is `recall@20`. \n\n    column 1                    column 2\n    12899779_clicks  [51, 22, 73, 84, 95, ...]\n    12899779_carts   [8, 23, 12, 41, 52, ...]\n    12899779_orders  [11, 21, 37, 40, 60, ...]",
      "votes": null
    },
    {
      "id": "2045663",
      "postDate": "11/27/2022 14:52:34",
      "content": "<p>What is the meaning or interpretation of the numbers in the list after 12899779_clicks?</p>",
      "rawMarkdown": "What is the meaning or interpretation of the numbers in the list after 12899779_clicks?",
      "votes": null
    },
    {
      "id": "2045666",
      "postDate": "11/27/2022 14:55:35",
      "content": "<p>They are item id numbers. In this competition, there are 1.8 million unique items.</p>",
      "rawMarkdown": "They are item id numbers. In this competition, there are 1.8 million unique items.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2024663,
      "author_name": "cdeotte",
      "author_url": "",
      "post_date": "11/10/2022 17:06:29",
      "content": "<p>We are making <code>1,671,803</code> times <code>3</code> lists of predictions. There are <code>1,671,803</code> sessions. In this competition, a \"session\" means \"user\". So there are <code>1,671,803</code> users. The test period is one week long. For each user, we need to predict what they will <code>click</code>, <code>cart</code> and <code>order</code> during the remainder of the test period (after the information we are provided about each user). The order of the 20 numbers within our lists does not matter.</p>\n<p>So for user \"12899779\", we add 3 rows to <code>submission.csv</code>. Where the second column contains a list of our 20 predictions. The metric is <code>recall@20</code>. </p>\n<pre><code>column 1                    column 2\n12899779_clicks  [51, 22, 73, 84, 95, ...]\n12899779_carts   [8, 23, 12, 41, 52, ...]\n12899779_orders  [11, 21, 37, 40, 60, ...]\n</code></pre>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2045663,
      "author_name": "galbatorix",
      "author_url": "",
      "post_date": "11/27/2022 14:52:34",
      "content": "<p>What is the meaning or interpretation of the numbers in the list after 12899779_clicks?</p>",
      "votes": null,
      "replies": [
        {
          "id": 2045666,
          "author_name": "cdeotte",
          "author_url": "",
          "post_date": "11/27/2022 14:55:35",
          "content": "<p>They are item id numbers. In this competition, there are 1.8 million unique items.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2024560": "Can someone explain why the structure of the sample submission. I dont understand what the unique value is and why its always the same. Also why is the event given even though we have to predict it.\nThanks in advance!",
    "2024663": "We are making `1,671,803` times `3` lists of predictions. There are `1,671,803` sessions. In this competition, a \"session\" means \"user\". So there are `1,671,803` users. The test period is one week long. For each user, we need to predict what they will `click`, `cart` and `order` during the remainder of the test period (after the information we are provided about each user). The order of the 20 numbers within our lists does not matter.\n\nSo for user \"12899779\", we add 3 rows to `submission.csv`. Where the second column contains a list of our 20 predictions. The metric is `recall@20`. \n\n    column 1                    column 2\n    12899779_clicks  [51, 22, 73, 84, 95, ...]\n    12899779_carts   [8, 23, 12, 41, 52, ...]\n    12899779_orders  [11, 21, 37, 40, 60, ...]",
    "2045663": "What is the meaning or interpretation of the numbers in the list after 12899779_clicks?",
    "2045666": "They are item id numbers. In this competition, there are 1.8 million unique items."
  },
  "source": "meta"
}