{
  "id": 6764,
  "title": "Final model (agent) submissions",
  "url": "/competitions/flight2-main/discussion/6764",
  "author_name": "",
  "post_date": "2014-01-07T11:24:47.727Z",
  "votes": null,
  "comment_count": 12,
  "views": 6889,
  "content": "<p>Hello,</p>\n<p>can you please provide what should be submitted on 18Jan and how ? to only submission i see in the web page is of .csv files.</p>",
  "messages": [
    {
      "id": "37076",
      "postDate": "01/07/2014 11:24:47",
      "content": "<p>Hello,</p>\n<p>can you please provide what should be submitted on 18Jan and how ? to only submission i see in the web page is of .csv files.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37101",
      "postDate": "01/07/2014 23:50:07",
      "content": "<p>I'll be adding this to the submission page:</p>\n<p>[quote=Nick Sardo;1234]</p>\n<p>When you submit an Entry, you also will be able to upload your model to the Website. Your final model must contain all code and parameter settings necessary to evaluate your models on new data, and include a README file with instructions on how to do so. The purpose of this is to ensure this Competition is fair and no answers were looked up for the Final Evaluation Set period. This model submission is required to be eligible to win a prize. This final model will be used solely for the purpose of verifying eligibility for prizes. You may upload either the archive itself or a SHA-256 hash of the archive.<br>Kaggle will check the final submissions from the Entrants at the top of the Final Evaluation Leaderboard to make sure that Entrant&#8217;s final submission was created using that Entrant&#8217;s final model submission. If a potential winner did not use the final model codebase to create the final submission Entry, that potential winner will be disqualified and Kaggle will evaluate the final submission of the next Entrant on the Leaderboard as a potential winner.</p>\n<p>For an entry to be eligible to win, any open source or third party code or other materials used in connection with that Entry must be limited to code that GE will be entitled to commercially use and exploit, and any open-source-related or third party licensor requirements and restrictions must be disclosed as part of the model submission.</p>\n<p>Please do not upload Kaggle-provided data files, but mention which ones are used and how to link them to your agent in the readme.</p>\n<p>Regarding non-Kaggle data (Previous leaderboard GRIB files): Zip this data and paste its hash into a text file called 'Data Hash.txt'. Place this text file into your model archive. If your model is in-the-money, we'll contact you to retrieve the non-Kaggle data and validate the hash.</p>\n<p>&nbsp;</p>\n<p>[/quote]</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37115",
      "postDate": "01/08/2014 07:10:10",
      "content": "<p>I'm confused: how are we supposed to create the hash of the zip file containing grib data <span>before</span> such data is even available (I'm talking about grib files for the final leaderboard)? Or are we supposed to update the model archive with the hash, together with the final submission in February?</p>\n<p>Thank you.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37117",
      "postDate": "01/08/2014 07:39:14",
      "content": "<p>[quote=Gabriele Seppi;37115]</p>\n<p>I'm confused: how are we supposed to create the hash of the zip file containing grib data before such data is even available (I'm talking about grib files for the final leaderboard)? Or are we supposed to update the model archive with the hash, together with the final submission in February?</p>\n<p>Thank you.</p>\n<p>[/quote]</p>\n<p>Indeed, I'm only referring to grib files from previous leaderboards, which people may be using in their code. &nbsp;We'll download all grib files for the next leaderboard, so a hash of those won't be necessary. &nbsp;Thanks</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37121",
      "postDate": "01/08/2014 09:12:35",
      "content": "<p>All clear now. Thanks =)</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37159",
      "postDate": "01/09/2014 10:06:26",
      "content": "<p>Flight Quest 1 allowed a hash of the source code to be uploaded instead of the source code itself. I really liked that, since&nbsp;it allowed participants to keep their code to themselves unless it turned out to be in the money. While the commercial value of my own experiments may be questionable, I would appreciate&nbsp;this to&nbsp;be an option also for Flight Quest 2. Who knows, somebody may have a nugget or two in there, even if their overall solution is not in the money.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37169",
      "postDate": "01/09/2014 18:20:37",
      "content": "<p>[quote=TecS;37159]</p>\n<p>Flight Quest 1 allowed a hash of the source code to be uploaded instead of the source code itself. I really liked that, since&nbsp;it allowed participants to keep their code to themselves unless it turned out to be in the money. While the commercial value of my own experiments may be questionable, I would appreciate&nbsp;this to&nbsp;be an option also for Flight Quest 2. Who knows, somebody may have a nugget or two in there, even if their overall solution is not in the money.</p>\n<p>[/quote]</p>\n<p>That's fine too. &nbsp;See updated model submission instructions.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37171",
      "postDate": "01/09/2014 18:25:40",
      "content": "<p>I have a doubt. What if my model requires, either as a config file or through input parameters, certain values, such as the days to compute and their cutoff times? Clearly those values are going to change from the current leaderboard to the final one.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37173",
      "postDate": "01/09/2014 18:56:47",
      "content": "<p>I and probably most others have the same issue. In theory it can be solved by doing a directory scan to detect the data files and extract everything from there. For now however I have an input file for the dates too. I guess if you describe in your readme what needs to be in there (and possibly provide an example of the file for the test data) it should be fine. In my opinion this is part of the description of how to plug the data files into your model. If I'm wrong about this I would really like to know. :-)</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37177",
      "postDate": "01/09/2014 19:35:11",
      "content": "<p>Actually I have other parameters that will change for the final leaderboard, regardless of directory scans. But yes, I had your same thought: the readme is hopefully there to explain also what could change (and possibly in which way).</p>\n<p>An official statement, though, would be much appreciated and would probably make everyone sleep tranquil =)</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37209",
      "postDate": "01/10/2014 10:28:30",
      "content": "<p>I have one more question: the model we upload must be able to reproduce our selected submission? I'm asking this in case we have some late modifications and we're unable to produce a solution file to submit by the deadline...</p>\n<p>Thanks you.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37293",
      "postDate": "01/11/2014 19:18:47",
      "content": "<p>Hi,</p>\n<p>Has anyone used python for code submission? if yes, which package (numpy/pandas etc.)&nbsp;offers better functionality&nbsp;and is easier for handling big datasets?</p>\n<p>X</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "37537",
      "postDate": "01/17/2014 20:50:50",
      "content": "<p>Regarding the hash to generate on grib files: do we have to do this process if we used the huge zip file kindly provided by TecS (https://www.gequest.com/c/flight2-main/download/Weather%20Forecast%20GRIB2%20Files%20(Sept%2011-24)%20-%20Provided%20by%20TecS.zip)?</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 37101,
      "author_name": "nicksardo",
      "author_url": "",
      "post_date": "01/07/2014 23:50:07",
      "content": "<p>I'll be adding this to the submission page:</p>\n<p>[quote=Nick Sardo;1234]</p>\n<p>When you submit an Entry, you also will be able to upload your model to the Website. Your final model must contain all code and parameter settings necessary to evaluate your models on new data, and include a README file with instructions on how to do so. The purpose of this is to ensure this Competition is fair and no answers were looked up for the Final Evaluation Set period. This model submission is required to be eligible to win a prize. This final model will be used solely for the purpose of verifying eligibility for prizes. You may upload either the archive itself or a SHA-256 hash of the archive.<br>Kaggle will check the final submissions from the Entrants at the top of the Final Evaluation Leaderboard to make sure that Entrant&#8217;s final submission was created using that Entrant&#8217;s final model submission. If a potential winner did not use the final model codebase to create the final submission Entry, that potential winner will be disqualified and Kaggle will evaluate the final submission of the next Entrant on the Leaderboard as a potential winner.</p>\n<p>For an entry to be eligible to win, any open source or third party code or other materials used in connection with that Entry must be limited to code that GE will be entitled to commercially use and exploit, and any open-source-related or third party licensor requirements and restrictions must be disclosed as part of the model submission.</p>\n<p>Please do not upload Kaggle-provided data files, but mention which ones are used and how to link them to your agent in the readme.</p>\n<p>Regarding non-Kaggle data (Previous leaderboard GRIB files): Zip this data and paste its hash into a text file called 'Data Hash.txt'. Place this text file into your model archive. If your model is in-the-money, we'll contact you to retrieve the non-Kaggle data and validate the hash.</p>\n<p>&nbsp;</p>\n<p>[/quote]</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37115,
      "author_name": "gabrieleseppi",
      "author_url": "",
      "post_date": "01/08/2014 07:10:10",
      "content": "<p>I'm confused: how are we supposed to create the hash of the zip file containing grib data <span>before</span> such data is even available (I'm talking about grib files for the final leaderboard)? Or are we supposed to update the model archive with the hash, together with the final submission in February?</p>\n<p>Thank you.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37117,
      "author_name": "nicksardo",
      "author_url": "",
      "post_date": "01/08/2014 07:39:14",
      "content": "<p>[quote=Gabriele Seppi;37115]</p>\n<p>I'm confused: how are we supposed to create the hash of the zip file containing grib data before such data is even available (I'm talking about grib files for the final leaderboard)? Or are we supposed to update the model archive with the hash, together with the final submission in February?</p>\n<p>Thank you.</p>\n<p>[/quote]</p>\n<p>Indeed, I'm only referring to grib files from previous leaderboards, which people may be using in their code. &nbsp;We'll download all grib files for the next leaderboard, so a hash of those won't be necessary. &nbsp;Thanks</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37121,
      "author_name": "gabrieleseppi",
      "author_url": "",
      "post_date": "01/08/2014 09:12:35",
      "content": "<p>All clear now. Thanks =)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37159,
      "author_name": "hyperborean",
      "author_url": "",
      "post_date": "01/09/2014 10:06:26",
      "content": "<p>Flight Quest 1 allowed a hash of the source code to be uploaded instead of the source code itself. I really liked that, since&nbsp;it allowed participants to keep their code to themselves unless it turned out to be in the money. While the commercial value of my own experiments may be questionable, I would appreciate&nbsp;this to&nbsp;be an option also for Flight Quest 2. Who knows, somebody may have a nugget or two in there, even if their overall solution is not in the money.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37169,
      "author_name": "nicksardo",
      "author_url": "",
      "post_date": "01/09/2014 18:20:37",
      "content": "<p>[quote=TecS;37159]</p>\n<p>Flight Quest 1 allowed a hash of the source code to be uploaded instead of the source code itself. I really liked that, since&nbsp;it allowed participants to keep their code to themselves unless it turned out to be in the money. While the commercial value of my own experiments may be questionable, I would appreciate&nbsp;this to&nbsp;be an option also for Flight Quest 2. Who knows, somebody may have a nugget or two in there, even if their overall solution is not in the money.</p>\n<p>[/quote]</p>\n<p>That's fine too. &nbsp;See updated model submission instructions.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37171,
      "author_name": "gabrieleseppi",
      "author_url": "",
      "post_date": "01/09/2014 18:25:40",
      "content": "<p>I have a doubt. What if my model requires, either as a config file or through input parameters, certain values, such as the days to compute and their cutoff times? Clearly those values are going to change from the current leaderboard to the final one.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37173,
      "author_name": "wmestrom",
      "author_url": "",
      "post_date": "01/09/2014 18:56:47",
      "content": "<p>I and probably most others have the same issue. In theory it can be solved by doing a directory scan to detect the data files and extract everything from there. For now however I have an input file for the dates too. I guess if you describe in your readme what needs to be in there (and possibly provide an example of the file for the test data) it should be fine. In my opinion this is part of the description of how to plug the data files into your model. If I'm wrong about this I would really like to know. :-)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37177,
      "author_name": "gabrieleseppi",
      "author_url": "",
      "post_date": "01/09/2014 19:35:11",
      "content": "<p>Actually I have other parameters that will change for the final leaderboard, regardless of directory scans. But yes, I had your same thought: the readme is hopefully there to explain also what could change (and possibly in which way).</p>\n<p>An official statement, though, would be much appreciated and would probably make everyone sleep tranquil =)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37209,
      "author_name": "gabrieleseppi",
      "author_url": "",
      "post_date": "01/10/2014 10:28:30",
      "content": "<p>I have one more question: the model we upload must be able to reproduce our selected submission? I'm asking this in case we have some late modifications and we're unable to produce a solution file to submit by the deadline...</p>\n<p>Thanks you.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37293,
      "author_name": "pallavi125220",
      "author_url": "",
      "post_date": "01/11/2014 19:18:47",
      "content": "<p>Hi,</p>\n<p>Has anyone used python for code submission? if yes, which package (numpy/pandas etc.)&nbsp;offers better functionality&nbsp;and is easier for handling big datasets?</p>\n<p>X</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 37537,
      "author_name": "gabrieleseppi",
      "author_url": "",
      "post_date": "01/17/2014 20:50:50",
      "content": "<p>Regarding the hash to generate on grib files: do we have to do this process if we used the huge zip file kindly provided by TecS (https://www.gequest.com/c/flight2-main/download/Weather%20Forecast%20GRIB2%20Files%20(Sept%2011-24)%20-%20Provided%20by%20TecS.zip)?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "37076": "",
    "37101": "",
    "37115": "",
    "37117": "",
    "37121": "",
    "37159": "",
    "37169": "",
    "37171": "",
    "37173": "",
    "37177": "",
    "37209": "",
    "37293": "",
    "37537": ""
  },
  "source": "meta"
}