{
  "id": 20590,
  "title": "does the training set cover all users in test and holdout?",
  "url": "/competitions/expedia-hotel-recommendations/discussion/20590",
  "author_name": "",
  "post_date": "2016-05-01T11:29:49.507Z",
  "votes": null,
  "comment_count": 2,
  "views": 506,
  "content": "<p>I noticed that every user that appeared in the test set was also in the training set. Is this also the case with the holdout data set for the private leader board? This would make sense, as it would mean that we're only predicting bookings for users who have some history.</p>",
  "messages": [
    {
      "id": "117882",
      "postDate": "05/01/2016 11:29:49",
      "content": "<p>I noticed that every user that appeared in the test set was also in the training set. Is this also the case with the holdout data set for the private leader board? This would make sense, as it would mean that we're only predicting bookings for users who have some history.</p>",
      "rawMarkdown": "I noticed that every user that appeared in the test set was also in the training set. Is this also the case with the holdout data set for the private leader board? This would make sense, as it would mean that we're only predicting bookings for users who have some history.",
      "votes": null
    },
    {
      "id": "117885",
      "postDate": "05/01/2016 11:48:02",
      "content": "<p><code>test.csv</code> contains data for both public and private leaderboard, we just don't know which row is which =). So if your finding is correct than it holds for both public and private LB.</p>",
      "rawMarkdown": "`test.csv` contains data for both public and private leaderboard, we just don't know which row is which =). So if your finding is correct than it holds for both public and private LB.",
      "votes": null
    },
    {
      "id": "117891",
      "postDate": "05/01/2016 13:59:46",
      "content": "<p>I am new to kaggle. Could I know the difference between the public and private leaderboard? In the sense that do we only need to submit predictions for the test data? </p>\n\n<p>I would also like to know if all users in test appears in train is correct, as I found differing result. </p>",
      "rawMarkdown": "I am new to kaggle. Could I know the difference between the public and private leaderboard? In the sense that do we only need to submit predictions for the test data? \r\n\r\nI would also like to know if all users in test appears in train is correct, as I found differing result.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 117885,
      "author_name": "dvasyukova",
      "author_url": "",
      "post_date": "05/01/2016 11:48:02",
      "content": "<p><code>test.csv</code> contains data for both public and private leaderboard, we just don't know which row is which =). So if your finding is correct than it holds for both public and private LB.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 117891,
      "author_name": "kirukrish",
      "author_url": "",
      "post_date": "05/01/2016 13:59:46",
      "content": "<p>I am new to kaggle. Could I know the difference between the public and private leaderboard? In the sense that do we only need to submit predictions for the test data? </p>\n\n<p>I would also like to know if all users in test appears in train is correct, as I found differing result. </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "117882": "I noticed that every user that appeared in the test set was also in the training set. Is this also the case with the holdout data set for the private leader board? This would make sense, as it would mean that we're only predicting bookings for users who have some history.",
    "117885": "`test.csv` contains data for both public and private leaderboard, we just don't know which row is which =). So if your finding is correct than it holds for both public and private LB.",
    "117891": "I am new to kaggle. Could I know the difference between the public and private leaderboard? In the sense that do we only need to submit predictions for the test data? \r\n\r\nI would also like to know if all users in test appears in train is correct, as I found differing result."
  },
  "source": "meta"
}