{
  "id": 180561,
  "title": "May we get approximate latitude and longitude of the test sites?",
  "url": "/competitions/birdsong-recognition/discussion/180561",
  "author_name": "",
  "post_date": "2020-09-05T14:32:25.079469Z",
  "votes": 2,
  "comment_count": 8,
  "views": 0,
  "content": "<p>I was reviewing the training data and it provides location based on latitude and longitude.</p>\n<p>From a production standpoint, it seems like you will always know where you are taking recordings and can therefore put that in as an input to the model. Therefore, I was wondering if we could get approximate latitude and longitude's for the three testing sites to use as inputs to our model during testing as well?</p>",
  "messages": [
    {
      "id": "999287",
      "postDate": "09/05/2020 14:32:25",
      "content": "<p>I was reviewing the training data and it provides location based on latitude and longitude.</p>\n<p>From a production standpoint, it seems like you will always know where you are taking recordings and can therefore put that in as an input to the model. Therefore, I was wondering if we could get approximate latitude and longitude's for the three testing sites to use as inputs to our model during testing as well?</p>",
      "rawMarkdown": "I was reviewing the training data and it provides location based on latitude and longitude.\n\nFrom a production standpoint, it seems like you will always know where you are taking recordings and can therefore put that in as an input to the model. Therefore, I was wondering if we could get approximate latitude and longitude's for the three testing sites to use as inputs to our model during testing as well?",
      "votes": null
    },
    {
      "id": "999289",
      "postDate": "09/05/2020 14:34:36",
      "content": "<p><a href=\"https://www.kaggle.com/stefankahl\" target=\"_blank\">@stefankahl</a> ^^</p>",
      "rawMarkdown": "stefankahl ^^",
      "votes": null
    },
    {
      "id": "999412",
      "postDate": "09/05/2020 16:13:41",
      "content": "<p>That would make sense but please don't release such information 10 days before the end of the competition … </p>",
      "rawMarkdown": "That would make sense but please don't release such information 10 days before the end of the competition ...",
      "votes": null
    },
    {
      "id": "999448",
      "postDate": "09/05/2020 16:46:57",
      "content": "<p>I see your perspective. . . and I agree. It doesn't really seem fair, and I would find the prospect pretty demoralizing if I were in gold right now.</p>\n<p>I do wish someone else had thought about this and pointed it out a couple of months back though.</p>\n<p>Anyways, cheers.</p>",
      "rawMarkdown": "I see your perspective. . . and I agree. It doesn't really seem fair, and I would find the prospect pretty demoralizing if I were in gold right now.\n\nI do wish someone else had thought about this and pointed it out a couple of months back though.\n\nAnyways, cheers.",
      "votes": null
    },
    {
      "id": "999702",
      "postDate": "09/05/2020 23:24:15",
      "content": "<p>If host wanted us to use meta data for prediction then they would have provided meta data in test.  It is clear they want us to produce the best possible audio only models.</p>\n<p>My 2 cents.</p>",
      "rawMarkdown": "If host wanted us to use meta data for prediction then they would have provided meta data in test.  It is clear they want us to produce the best possible audio only models.\n\nMy 2 cents.",
      "votes": null
    },
    {
      "id": "1000027",
      "postDate": "09/06/2020 08:02:41",
      "content": "<p>This was discussed a bit here in Tom Denton's reply (short answer is No)<br>\n<a href=\"https://www.kaggle.com/c/birdsong-recognition/discussion/159172#905796\" target=\"_blank\">https://www.kaggle.com/c/birdsong-recognition/discussion/159172#905796</a></p>",
      "rawMarkdown": "This was discussed a bit here in Tom Denton's reply (short answer is No)\nhttps://www.kaggle.com/c/birdsong-recognition/discussion/159172#905796",
      "votes": null
    },
    {
      "id": "1000688",
      "postDate": "09/06/2020 17:48:45",
      "content": "<p>(Was just coming in to link back to that discussion. Thanks!)</p>",
      "rawMarkdown": "(Was just coming in to link back to that discussion. Thanks!)",
      "votes": null
    },
    {
      "id": "1001128",
      "postDate": "09/07/2020 04:47:06",
      "content": "<p>No worries.  At least can be good for something in this competition!!</p>",
      "rawMarkdown": "No worries.  At least can be good for something in this competition!!",
      "votes": null
    },
    {
      "id": "1001586",
      "postDate": "09/07/2020 12:26:12",
      "content": "<p>Oooo nice thanks for the share!</p>",
      "rawMarkdown": "Oooo nice thanks for the share!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 999289,
      "author_name": "j7hoenix",
      "author_url": "",
      "post_date": "09/05/2020 14:34:36",
      "content": "<p><a href=\"https://www.kaggle.com/stefankahl\" target=\"_blank\">@stefankahl</a> ^^</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 999412,
      "author_name": "theoviel",
      "author_url": "",
      "post_date": "09/05/2020 16:13:41",
      "content": "<p>That would make sense but please don't release such information 10 days before the end of the competition … </p>",
      "votes": null,
      "replies": [
        {
          "id": 999448,
          "author_name": "j7hoenix",
          "author_url": "",
          "post_date": "09/05/2020 16:46:57",
          "content": "<p>I see your perspective. . . and I agree. It doesn't really seem fair, and I would find the prospect pretty demoralizing if I were in gold right now.</p>\n<p>I do wish someone else had thought about this and pointed it out a couple of months back though.</p>\n<p>Anyways, cheers.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 999702,
          "author_name": "cpmpml",
          "author_url": "",
          "post_date": "09/05/2020 23:24:15",
          "content": "<p>If host wanted us to use meta data for prediction then they would have provided meta data in test.  It is clear they want us to produce the best possible audio only models.</p>\n<p>My 2 cents.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1000027,
      "author_name": "something4kag",
      "author_url": "",
      "post_date": "09/06/2020 08:02:41",
      "content": "<p>This was discussed a bit here in Tom Denton's reply (short answer is No)<br>\n<a href=\"https://www.kaggle.com/c/birdsong-recognition/discussion/159172#905796\" target=\"_blank\">https://www.kaggle.com/c/birdsong-recognition/discussion/159172#905796</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 1000688,
          "author_name": "tomdenton",
          "author_url": "",
          "post_date": "09/06/2020 17:48:45",
          "content": "<p>(Was just coming in to link back to that discussion. Thanks!)</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1001128,
          "author_name": "something4kag",
          "author_url": "",
          "post_date": "09/07/2020 04:47:06",
          "content": "<p>No worries.  At least can be good for something in this competition!!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1001586,
          "author_name": "j7hoenix",
          "author_url": "",
          "post_date": "09/07/2020 12:26:12",
          "content": "<p>Oooo nice thanks for the share!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "999287": "I was reviewing the training data and it provides location based on latitude and longitude.\n\nFrom a production standpoint, it seems like you will always know where you are taking recordings and can therefore put that in as an input to the model. Therefore, I was wondering if we could get approximate latitude and longitude's for the three testing sites to use as inputs to our model during testing as well?",
    "999289": "stefankahl ^^",
    "999412": "That would make sense but please don't release such information 10 days before the end of the competition ...",
    "999448": "I see your perspective. . . and I agree. It doesn't really seem fair, and I would find the prospect pretty demoralizing if I were in gold right now.\n\nI do wish someone else had thought about this and pointed it out a couple of months back though.\n\nAnyways, cheers.",
    "999702": "If host wanted us to use meta data for prediction then they would have provided meta data in test.  It is clear they want us to produce the best possible audio only models.\n\nMy 2 cents.",
    "1000027": "This was discussed a bit here in Tom Denton's reply (short answer is No)\nhttps://www.kaggle.com/c/birdsong-recognition/discussion/159172#905796",
    "1000688": "(Was just coming in to link back to that discussion. Thanks!)",
    "1001128": "No worries.  At least can be good for something in this competition!!",
    "1001586": "Oooo nice thanks for the share!"
  },
  "source": "meta"
}