{
  "id": 181473,
  "title": "Where can I get enough 'nocall' data?",
  "url": "/competitions/birdsong-recognition/discussion/181473",
  "author_name": "",
  "post_date": "2020-09-09T03:09:22.412598200Z",
  "votes": 2,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I saw many people built binary clf to predict if there is 'nocall' and get AUC 0.9+.  </p>\n<p>So, does anybody explain how to do this and where I can get 'nocall' data. On example_test_audio, there are only limited labels of 'nocall'. Is it enough to build this classification?</p>\n<p>Thanks for any help!</p>",
  "messages": [
    {
      "id": "1003488",
      "postDate": "09/09/2020 03:09:22",
      "content": "<p>I saw many people built binary clf to predict if there is 'nocall' and get AUC 0.9+.  </p>\n<p>So, does anybody explain how to do this and where I can get 'nocall' data. On example_test_audio, there are only limited labels of 'nocall'. Is it enough to build this classification?</p>\n<p>Thanks for any help!</p>",
      "rawMarkdown": "I saw many people built binary clf to predict if there is 'nocall' and get AUC 0.9+.  \n\nSo, does anybody explain how to do this and where I can get 'nocall' data. On example_test_audio, there are only limited labels of 'nocall'. Is it enough to build this classification?\n\nThanks for any help!",
      "votes": null
    },
    {
      "id": "1003780",
      "postDate": "09/09/2020 08:37:38",
      "content": "<p>This is a key question indeed.  Maybe look at external data thread for inspiration.</p>",
      "rawMarkdown": "This is a key question indeed.  Maybe look at external data thread for inspiration.",
      "votes": null
    },
    {
      "id": "1004879",
      "postDate": "09/10/2020 05:31:35",
      "content": "<p>Thx, I will check them.</p>",
      "rawMarkdown": "Thx, I will check them.",
      "votes": null
    },
    {
      "id": "1008155",
      "postDate": "09/12/2020 19:29:22",
      "content": "<p>Can you not just use any other audio available on Kaggle or do you look for specific audio?</p>",
      "rawMarkdown": "Can you not just use any other audio available on Kaggle or do you look for specific audio?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1003780,
      "author_name": "cpmpml",
      "author_url": "",
      "post_date": "09/09/2020 08:37:38",
      "content": "<p>This is a key question indeed.  Maybe look at external data thread for inspiration.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1004879,
          "author_name": "woshifym",
          "author_url": "",
          "post_date": "09/10/2020 05:31:35",
          "content": "<p>Thx, I will check them.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1008155,
      "author_name": "aliabdin1",
      "author_url": "",
      "post_date": "09/12/2020 19:29:22",
      "content": "<p>Can you not just use any other audio available on Kaggle or do you look for specific audio?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1003488": "I saw many people built binary clf to predict if there is 'nocall' and get AUC 0.9+.  \n\nSo, does anybody explain how to do this and where I can get 'nocall' data. On example_test_audio, there are only limited labels of 'nocall'. Is it enough to build this classification?\n\nThanks for any help!",
    "1003780": "This is a key question indeed.  Maybe look at external data thread for inspiration.",
    "1004879": "Thx, I will check them.",
    "1008155": "Can you not just use any other audio available on Kaggle or do you look for specific audio?"
  },
  "source": "meta"
}