{
  "id": 71046,
  "title": "Do you really Believe 0.0000 RMSE ",
  "url": "/competitions/ga-customer-revenue-prediction/discussion/71046",
  "author_name": "",
  "post_date": "2018-11-09T14:35:22.815408900Z",
  "votes": -3,
  "comment_count": 7,
  "views": 0,
  "content": "<p>As a matter of fact its not possible to build model which Exactly fits with test data. in general only one possible case to make to 0.0000 is calculate the RMSE against actual data so that it will generate 0.0000 score.\nanother possibility is there is Bug in the scoring algorithm,</p>\n\n<p>as we are seeing since one month there are lots of issues around the competition its really frustrating, hope kaggle team thoroughly check all the issues before starting the competition   </p>",
  "messages": [
    {
      "id": "418248",
      "postDate": "11/09/2018 14:35:22",
      "content": "<p>As a matter of fact its not possible to build model which Exactly fits with test data. in general only one possible case to make to 0.0000 is calculate the RMSE against actual data so that it will generate 0.0000 score.\nanother possibility is there is Bug in the scoring algorithm,</p>\n\n<p>as we are seeing since one month there are lots of issues around the competition its really frustrating, hope kaggle team thoroughly check all the issues before starting the competition   </p>",
      "rawMarkdown": "As a matter of fact its not possible to build model which Exactly fits with test data. in general only one possible case to make to 0.0000 is calculate the RMSE against actual data so that it will generate 0.0000 score.\nanother possibility is there is Bug in the scoring algorithm,\n\nas we are seeing since one month there are lots of issues around the competition its really frustrating, hope kaggle team thoroughly check all the issues before starting the competition",
      "votes": null
    },
    {
      "id": "418252",
      "postDate": "11/09/2018 14:38:57",
      "content": "<p>Hello Thodime,</p>\n\n<p>It's not an \"issue\". The test_v2.csv = LeaderBoard.. </p>\n\n<p>You will need to have a good CV strategy as we won't be able to \"rely\" on the LeaderBoard :)</p>",
      "rawMarkdown": "Hello Thodime,\n\nIt's not an \"issue\". The test_v2.csv = LeaderBoard.. \n\nYou will need to have a good CV strategy as we won't be able to \"rely\" on the LeaderBoard :)",
      "votes": null
    },
    {
      "id": "418254",
      "postDate": "11/09/2018 14:40:05",
      "content": "<p>Looks like there's another problem with the testing data, here's what the guy that scored 0.0000 wrote a bit earlier:\n<a href=\"https://www.kaggle.com/c/ga-customer-revenue-prediction/discussion/71025\">https://www.kaggle.com/c/ga-customer-revenue-prediction/discussion/71025</a>  </p>",
      "rawMarkdown": "Looks like there's another problem with the testing data, here's what the guy that scored 0.0000 wrote a bit earlier:\nhttps://www.kaggle.com/c/ga-customer-revenue-prediction/discussion/71025",
      "votes": null
    },
    {
      "id": "418270",
      "postDate": "11/09/2018 15:16:14",
      "content": "<p>This is a jorke</p>",
      "rawMarkdown": "This is a jorke",
      "votes": null
    },
    {
      "id": "418344",
      "postDate": "11/09/2018 17:39:04",
      "content": "<p>Have no fear of perfection - you'll never reach it. </p>\n\n<p>Salvador Dali</p>",
      "rawMarkdown": "Have no fear of perfection - you'll never reach it. \n\nSalvador Dali",
      "votes": null
    },
    {
      "id": "418511",
      "postDate": "11/10/2018 02:05:22",
      "content": "<p>what happened？ I can't submit！</p>",
      "rawMarkdown": "what happened？ I can't submit！",
      "votes": null
    },
    {
      "id": "418728",
      "postDate": "11/10/2018 13:54:07",
      "content": "<p>After the leak detection by Jerome Vallet, this contest will have two phases. In the first phase the test set contains the target intentionally (it would be like a validation set). The challenge of the second phase is to predict the target in a future time frame.</p>",
      "rawMarkdown": "After the leak detection by Jerome Vallet, this contest will have two phases. In the first phase the test set contains the target intentionally (it would be like a validation set). The challenge of the second phase is to predict the target in a future time frame.",
      "votes": null
    },
    {
      "id": "418774",
      "postDate": "11/10/2018 15:37:50",
      "content": "<p>what happened？ I can't submit！</p>",
      "rawMarkdown": "what happened？ I can't submit！",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 418252,
      "author_name": "xtelima",
      "author_url": "",
      "post_date": "11/09/2018 14:38:57",
      "content": "<p>Hello Thodime,</p>\n\n<p>It's not an \"issue\". The test_v2.csv = LeaderBoard.. </p>\n\n<p>You will need to have a good CV strategy as we won't be able to \"rely\" on the LeaderBoard :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 418254,
      "author_name": "justuser",
      "author_url": "",
      "post_date": "11/09/2018 14:40:05",
      "content": "<p>Looks like there's another problem with the testing data, here's what the guy that scored 0.0000 wrote a bit earlier:\n<a href=\"https://www.kaggle.com/c/ga-customer-revenue-prediction/discussion/71025\">https://www.kaggle.com/c/ga-customer-revenue-prediction/discussion/71025</a>  </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 418270,
      "author_name": "wanxi1990",
      "author_url": "",
      "post_date": "11/09/2018 15:16:14",
      "content": "<p>This is a jorke</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 418344,
      "author_name": "a45632",
      "author_url": "",
      "post_date": "11/09/2018 17:39:04",
      "content": "<p>Have no fear of perfection - you'll never reach it. </p>\n\n<p>Salvador Dali</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 418511,
      "author_name": "tbwhsb88",
      "author_url": "",
      "post_date": "11/10/2018 02:05:22",
      "content": "<p>what happened？ I can't submit！</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 418728,
      "author_name": "paulorzp",
      "author_url": "",
      "post_date": "11/10/2018 13:54:07",
      "content": "<p>After the leak detection by Jerome Vallet, this contest will have two phases. In the first phase the test set contains the target intentionally (it would be like a validation set). The challenge of the second phase is to predict the target in a future time frame.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 418774,
      "author_name": "dinglubjut25079",
      "author_url": "",
      "post_date": "11/10/2018 15:37:50",
      "content": "<p>what happened？ I can't submit！</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "418248": "As a matter of fact its not possible to build model which Exactly fits with test data. in general only one possible case to make to 0.0000 is calculate the RMSE against actual data so that it will generate 0.0000 score.\nanother possibility is there is Bug in the scoring algorithm,\n\nas we are seeing since one month there are lots of issues around the competition its really frustrating, hope kaggle team thoroughly check all the issues before starting the competition",
    "418252": "Hello Thodime,\n\nIt's not an \"issue\". The test_v2.csv = LeaderBoard.. \n\nYou will need to have a good CV strategy as we won't be able to \"rely\" on the LeaderBoard :)",
    "418254": "Looks like there's another problem with the testing data, here's what the guy that scored 0.0000 wrote a bit earlier:\nhttps://www.kaggle.com/c/ga-customer-revenue-prediction/discussion/71025",
    "418270": "This is a jorke",
    "418344": "Have no fear of perfection - you'll never reach it. \n\nSalvador Dali",
    "418511": "what happened？ I can't submit！",
    "418728": "After the leak detection by Jerome Vallet, this contest will have two phases. In the first phase the test set contains the target intentionally (it would be like a validation set). The challenge of the second phase is to predict the target in a future time frame.",
    "418774": "what happened？ I can't submit！"
  },
  "source": "meta"
}