{
  "id": 4295,
  "title": "Submission format",
  "url": "/competitions/challenges-in-representation-learning-the-black-box-learning-challenge/discussion/4295",
  "author_name": "",
  "post_date": "2013-04-13T06:08:45.677Z",
  "votes": null,
  "comment_count": 8,
  "views": 2295,
  "content": "<p>Note that your submissions should be of the type &quot;1.0&quot; instead of just &quot;1&quot;</p>\r\n<p>Thanks!</p>\r\n<p>Dumitru</p>",
  "messages": [
    {
      "id": "22733",
      "postDate": "04/13/2013 06:08:45",
      "content": "<p>Note that your submissions should be of the type &quot;1.0&quot; instead of just &quot;1&quot;</p>\r\n<p>Thanks!</p>\r\n<p>Dumitru</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "22734",
      "postDate": "04/13/2013 06:33:59",
      "content": "<p>[quote]</p>\r\n<h3># of Predictions</h3>\r\n<p>We expect the solution file to have 10,000 predictions. Header rows are optional. Please see the sample submission file on the\r\n<a href=\"http://www.kaggle.com/c/challenges-in-representation-learning-the-black-box-learning-challenge/data\">\r\ndata page</a> for an example of a valid submission.</p>\r\n<p>[/quote]</p>\r\n<p>I think you forgot to add the mentioned sample submission file.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "22864",
      "postDate": "04/15/2013 22:50:41",
      "content": "<p>Yeah not seeing any sample submission file...</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "22865",
      "postDate": "04/15/2013 22:53:15",
      "content": "<p>sure, will add one in just a bit!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "22866",
      "postDate": "04/15/2013 22:59:42",
      "content": "<p>Done! It's now downloadable on the <a href=\"https://www.kaggle.com/c/challenges-in-representation-learning-the-black-box-learning-challenge/data\">\r\ndata</a> page. I've also made it a reference benchmark (it's just a random sample of integers, so it's not a particularly good model).</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "22870",
      "postDate": "04/16/2013 00:06:19",
      "content": "<p>Note that there's also a sample submission script for those of you using python:</p>\r\n<p>https://github.com/lisa-lab/pylearn2/blob/master/pylearn2/scripts/icml_2013_wrepl/black_box/make_submission.py</p>\r\n<p>It will make a correctly formatted submission file for you, given any pylearn2 model saved in a pickle file.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "23220",
      "postDate": "04/20/2013 03:07:58",
      "content": "<p>Does someone know how to save a file in the correct format, using R?</p>\r\n<p>My current method is to open my predictions .csv in excel, change the format to #.# then save it as .txt</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "23223",
      "postDate": "04/20/2013 03:54:38",
      "content": "<p>[quote=Benoit Plante;23220]</p>\r\n<p>Does someone knows how to save a file in the correct format, using R?</p>\r\n<p>My current method is to open the .csv in excel, change the format to #.# then save it as .txt</p>\r\n<p>[/quote]</p>\r\n<p>&nbsp;</p>\r\n<pre> write.csv(data.frame(label = formatC(pred, digits=1, format=&quot;f&quot;)),<br>          file = &quot;predictions.csv&quot;, row.names = F, quote = F)</pre>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "23261",
      "postDate": "04/21/2013 03:32:47",
      "content": "<p>[quote=Leustagos;23223]</p>\r\n<pre> write.csv(data.frame(label = formatC(pred, digits=1, format=&quot;f&quot;)),<br>          file = &quot;predictions.csv&quot;, row.names = F, quote = F)</pre>\r\n<p>[/quote]</p>\r\n<p>Thanks it worked!</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 22734,
      "author_name": "kimsimonsen",
      "author_url": "",
      "post_date": "04/13/2013 06:33:59",
      "content": "<p>[quote]</p>\r\n<h3># of Predictions</h3>\r\n<p>We expect the solution file to have 10,000 predictions. Header rows are optional. Please see the sample submission file on the\r\n<a href=\"http://www.kaggle.com/c/challenges-in-representation-learning-the-black-box-learning-challenge/data\">\r\ndata page</a> for an example of a valid submission.</p>\r\n<p>[/quote]</p>\r\n<p>I think you forgot to add the mentioned sample submission file.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 22864,
      "author_name": "theovanrooy",
      "author_url": "",
      "post_date": "04/15/2013 22:50:41",
      "content": "<p>Yeah not seeing any sample submission file...</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 22865,
      "author_name": "dumitru0",
      "author_url": "",
      "post_date": "04/15/2013 22:53:15",
      "content": "<p>sure, will add one in just a bit!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 22866,
      "author_name": "dumitru0",
      "author_url": "",
      "post_date": "04/15/2013 22:59:42",
      "content": "<p>Done! It's now downloadable on the <a href=\"https://www.kaggle.com/c/challenges-in-representation-learning-the-black-box-learning-challenge/data\">\r\ndata</a> page. I've also made it a reference benchmark (it's just a random sample of integers, so it's not a particularly good model).</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 22870,
      "author_name": "iangoodfellow",
      "author_url": "",
      "post_date": "04/16/2013 00:06:19",
      "content": "<p>Note that there's also a sample submission script for those of you using python:</p>\r\n<p>https://github.com/lisa-lab/pylearn2/blob/master/pylearn2/scripts/icml_2013_wrepl/black_box/make_submission.py</p>\r\n<p>It will make a correctly formatted submission file for you, given any pylearn2 model saved in a pickle file.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 23220,
      "author_name": "benoitplante",
      "author_url": "",
      "post_date": "04/20/2013 03:07:58",
      "content": "<p>Does someone know how to save a file in the correct format, using R?</p>\r\n<p>My current method is to open my predictions .csv in excel, change the format to #.# then save it as .txt</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 23223,
      "author_name": "leustagos",
      "author_url": "",
      "post_date": "04/20/2013 03:54:38",
      "content": "<p>[quote=Benoit Plante;23220]</p>\r\n<p>Does someone knows how to save a file in the correct format, using R?</p>\r\n<p>My current method is to open the .csv in excel, change the format to #.# then save it as .txt</p>\r\n<p>[/quote]</p>\r\n<p>&nbsp;</p>\r\n<pre> write.csv(data.frame(label = formatC(pred, digits=1, format=&quot;f&quot;)),<br>          file = &quot;predictions.csv&quot;, row.names = F, quote = F)</pre>",
      "votes": null,
      "replies": []
    },
    {
      "id": 23261,
      "author_name": "benoitplante",
      "author_url": "",
      "post_date": "04/21/2013 03:32:47",
      "content": "<p>[quote=Leustagos;23223]</p>\r\n<pre> write.csv(data.frame(label = formatC(pred, digits=1, format=&quot;f&quot;)),<br>          file = &quot;predictions.csv&quot;, row.names = F, quote = F)</pre>\r\n<p>[/quote]</p>\r\n<p>Thanks it worked!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "22733": "",
    "22734": "",
    "22864": "",
    "22865": "",
    "22866": "",
    "22870": "",
    "23220": "",
    "23223": "",
    "23261": ""
  },
  "source": "meta"
}