{
  "id": 203076,
  "title": "What is the mean of prior_question_elapsed_time == 0?",
  "url": "/competitions/riiid-test-answer-prediction/discussion/203076",
  "author_name": "",
  "post_date": "2020-12-13T15:41:35.747661300Z",
  "votes": 5,
  "comment_count": 4,
  "views": 0,
  "content": "<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4184342%2F2bfa2cd213722affafa3c7f7a00d66f7%2F1.png?generation=1607874029997709&amp;alt=media\" alt=\"\"></p>\n<p>What is the mean of prior_question_elapsed_time == 0? <br>\nBug?</p>",
  "messages": [
    {
      "id": "1111276",
      "postDate": "12/13/2020 15:41:35",
      "content": "<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4184342%2F2bfa2cd213722affafa3c7f7a00d66f7%2F1.png?generation=1607874029997709&amp;alt=media\" alt=\"\"></p>\n<p>What is the mean of prior_question_elapsed_time == 0? <br>\nBug?</p>",
      "rawMarkdown": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4184342%2F2bfa2cd213722affafa3c7f7a00d66f7%2F1.png?generation=1607874029997709&alt=media)\n\nWhat is the mean of prior_question_elapsed_time == 0? \nBug?",
      "votes": null
    },
    {
      "id": "1111474",
      "postDate": "12/13/2020 18:21:15",
      "content": "<p>I guess it is possible that a user sometimes has to click through questions to resume where he/she were.<br>\nUsing <code>tmp = data_df[data_df['prior_question_elapsed_time']==0.0]</code> and <code>tmp['answered_correctly'].sum()/len(tmp)</code> yields 0.63 which is slightly less than the average (only 10m rows read).</p>",
      "rawMarkdown": "I guess it is possible that a user sometimes has to click through questions to resume where he/she were.\nUsing `tmp = data_df[data_df['prior_question_elapsed_time']==0.0]` and `tmp['answered_correctly'].sum()/len(tmp)` yields 0.63 which is slightly less than the average (only 10m rows read).",
      "votes": null
    },
    {
      "id": "1113165",
      "postDate": "12/15/2020 08:49:53",
      "content": "<blockquote>\n  <p>it is possible that a user sometimes has to click through questions to resume where he/she were.</p>\n</blockquote>\n<p>I wonder why this would be needed- doesn't the software take them directly to the relevant place ?</p>",
      "rawMarkdown": ">it is possible that a user sometimes has to click through questions to resume where he/she were.\n\nI wonder why this would be needed- doesn't the software take them directly to the relevant place ?",
      "votes": null
    },
    {
      "id": "1113569",
      "postDate": "12/15/2020 14:58:13",
      "content": "<p>Not sure and I think more EDA is needed.</p>",
      "rawMarkdown": "Not sure and I think more EDA is needed.",
      "votes": null
    },
    {
      "id": "1115800",
      "postDate": "12/16/2020 15:28:07",
      "content": "<p>maybe it have nan in value</p>",
      "rawMarkdown": "maybe it have nan in value",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1111474,
      "author_name": "scaomath",
      "author_url": "",
      "post_date": "12/13/2020 18:21:15",
      "content": "<p>I guess it is possible that a user sometimes has to click through questions to resume where he/she were.<br>\nUsing <code>tmp = data_df[data_df['prior_question_elapsed_time']==0.0]</code> and <code>tmp['answered_correctly'].sum()/len(tmp)</code> yields 0.63 which is slightly less than the average (only 10m rows read).</p>",
      "votes": null,
      "replies": [
        {
          "id": 1113165,
          "author_name": "watzisname",
          "author_url": "",
          "post_date": "12/15/2020 08:49:53",
          "content": "<blockquote>\n  <p>it is possible that a user sometimes has to click through questions to resume where he/she were.</p>\n</blockquote>\n<p>I wonder why this would be needed- doesn't the software take them directly to the relevant place ?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1113569,
          "author_name": "scaomath",
          "author_url": "",
          "post_date": "12/15/2020 14:58:13",
          "content": "<p>Not sure and I think more EDA is needed.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1115800,
      "author_name": "fasternews",
      "author_url": "",
      "post_date": "12/16/2020 15:28:07",
      "content": "<p>maybe it have nan in value</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1111276": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4184342%2F2bfa2cd213722affafa3c7f7a00d66f7%2F1.png?generation=1607874029997709&alt=media)\n\nWhat is the mean of prior_question_elapsed_time == 0? \nBug?",
    "1111474": "I guess it is possible that a user sometimes has to click through questions to resume where he/she were.\nUsing `tmp = data_df[data_df['prior_question_elapsed_time']==0.0]` and `tmp['answered_correctly'].sum()/len(tmp)` yields 0.63 which is slightly less than the average (only 10m rows read).",
    "1113165": ">it is possible that a user sometimes has to click through questions to resume where he/she were.\n\nI wonder why this would be needed- doesn't the software take them directly to the relevant place ?",
    "1113569": "Not sure and I think more EDA is needed.",
    "1115800": "maybe it have nan in value"
  },
  "source": "meta"
}