{
  "id": 19398,
  "title": "Submitting predictions for second stage.",
  "url": "/competitions/second-annual-data-science-bowl/discussion/19398",
  "author_name": "",
  "post_date": "2016-03-08T19:00:30.927Z",
  "votes": 2,
  "comment_count": 2,
  "views": 422,
  "content": "<p>Hello,</p>\n\n<p>My team (and a few others) are participating in this competition for credit in an academic module. We did not submit our model after reading this reply from a Kaggle admin to another participant: \n&quot;The leaderboard will be reset for the second stage, so at minimum you will need to resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model). That said, you should definitely submit your model and make at attempt at the prizes. New data + potential overfitting + luck can move you a lot more than eight places!&quot; \nAs the message says &quot;the minimum you will need to do is resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model).&quot;, however there does not appear to be anywhere to submit predictions anymore. </p>\n\n<p>What should participants like us who have not uploaded a model do to make predictions on the test data. </p>\n\n<p>Thank you</p>",
  "messages": [
    {
      "id": "110843",
      "postDate": "03/08/2016 19:00:30",
      "content": "<p>Hello,</p>\n\n<p>My team (and a few others) are participating in this competition for credit in an academic module. We did not submit our model after reading this reply from a Kaggle admin to another participant: \n&quot;The leaderboard will be reset for the second stage, so at minimum you will need to resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model). That said, you should definitely submit your model and make at attempt at the prizes. New data + potential overfitting + luck can move you a lot more than eight places!&quot; \nAs the message says &quot;the minimum you will need to do is resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model).&quot;, however there does not appear to be anywhere to submit predictions anymore. </p>\n\n<p>What should participants like us who have not uploaded a model do to make predictions on the test data. </p>\n\n<p>Thank you</p>",
      "rawMarkdown": "Hello,\r\n\r\nMy team (and a few others) are participating in this competition for credit in an academic module. We did not submit our model after reading this reply from a Kaggle admin to another participant: \r\n\"The leaderboard will be reset for the second stage, so at minimum you will need to resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model). That said, you should definitely submit your model and make at attempt at the prizes. New data + potential overfitting + luck can move you a lot more than eight places!\" \r\nAs the message says \"the minimum you will need to do is resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model).\", however there does not appear to be anywhere to submit predictions anymore. \r\n\r\nWhat should participants like us who have not uploaded a model do to make predictions on the test data. \r\n\r\nThank you",
      "votes": null
    },
    {
      "id": "110845",
      "postDate": "03/08/2016 19:18:52",
      "content": "<p>I think submissions are only temporarily suspended. See <a href=\"https://www.kaggle.com/c/second-annual-data-science-bowl/forums/t/19382/test-set-released\">https://www.kaggle.com/c/second-annual-data-science-bowl/forums/t/19382/test-set-released</a></p>",
      "rawMarkdown": "I think submissions are only temporarily suspended. See https://www.kaggle.com/c/second-annual-data-science-bowl/forums/t/19382/test-set-released",
      "votes": null
    },
    {
      "id": "110846",
      "postDate": "03/08/2016 19:25:17",
      "content": "<p>Thank you very much for the quick reply. </p>",
      "rawMarkdown": "Thank you very much for the quick reply.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 110845,
      "author_name": "hyperborean",
      "author_url": "",
      "post_date": "03/08/2016 19:18:52",
      "content": "<p>I think submissions are only temporarily suspended. See <a href=\"https://www.kaggle.com/c/second-annual-data-science-bowl/forums/t/19382/test-set-released\">https://www.kaggle.com/c/second-annual-data-science-bowl/forums/t/19382/test-set-released</a></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 110846,
      "author_name": "jamesfearn",
      "author_url": "",
      "post_date": "03/08/2016 19:25:17",
      "content": "<p>Thank you very much for the quick reply. </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "110843": "Hello,\r\n\r\nMy team (and a few others) are participating in this competition for credit in an academic module. We did not submit our model after reading this reply from a Kaggle admin to another participant: \r\n\"The leaderboard will be reset for the second stage, so at minimum you will need to resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model). That said, you should definitely submit your model and make at attempt at the prizes. New data + potential overfitting + luck can move you a lot more than eight places!\" \r\nAs the message says \"the minimum you will need to do is resubmit predictions on the test set (you can keep a prize-ineligible spot on the leaderboard without uploading a model).\", however there does not appear to be anywhere to submit predictions anymore. \r\n\r\nWhat should participants like us who have not uploaded a model do to make predictions on the test data. \r\n\r\nThank you",
    "110845": "I think submissions are only temporarily suspended. See https://www.kaggle.com/c/second-annual-data-science-bowl/forums/t/19382/test-set-released",
    "110846": "Thank you very much for the quick reply."
  },
  "source": "meta"
}