{
  "id": 491325,
  "title": "What does it mean [end_time] in the submission?",
  "url": "/competitions/birdclef-2024/discussion/491325",
  "author_name": "",
  "post_date": "2024-04-05T12:43:12.602174Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Could someone clarify how to form row_id from model predictions and test filenames <strong>soundscape_xxxxxx.ogg</strong> ?</p>\n<p>In the description we have:</p>\n<blockquote>\n  <p>For each row_id, you should predict the probability that a given bird species was present. There is one column per bird species, so you will need to provide 182 predictions per row. Each row covers a five-second window of audio.</p>\n  <p>row_id: A slug of soundscape_[soundscape_id]_[end_time] for the prediction.</p>\n</blockquote>\n<p>and </p>\n<blockquote>\n  <p>test_soundscapes/ When you submit a notebook, the test_soundscapes directory will be populated with approximately 1,100 recordings to be used for scoring. They are 4 minutes long and in ogg audio format. The file names are randomized but have the general form of soundscape_xxxxxx.ogg</p>\n</blockquote>\n<p>Should split each 4 min long test file to 48 pieces (5 second each) and add a segment id to row_id? </p>\n<p>Like for 1 soundscape_xxxxxx.ogg file we will have 48 rows in the submission  (soundscape_xxxxxx_5, soundscape_xxxxxx_10, soundscape_xxxxxx_15, etc.) </p>",
  "messages": [
    {
      "id": "2736856",
      "postDate": "04/05/2024 12:43:12",
      "content": "<p>Could someone clarify how to form row_id from model predictions and test filenames <strong>soundscape_xxxxxx.ogg</strong> ?</p>\n<p>In the description we have:</p>\n<blockquote>\n  <p>For each row_id, you should predict the probability that a given bird species was present. There is one column per bird species, so you will need to provide 182 predictions per row. Each row covers a five-second window of audio.</p>\n  <p>row_id: A slug of soundscape_[soundscape_id]_[end_time] for the prediction.</p>\n</blockquote>\n<p>and </p>\n<blockquote>\n  <p>test_soundscapes/ When you submit a notebook, the test_soundscapes directory will be populated with approximately 1,100 recordings to be used for scoring. They are 4 minutes long and in ogg audio format. The file names are randomized but have the general form of soundscape_xxxxxx.ogg</p>\n</blockquote>\n<p>Should split each 4 min long test file to 48 pieces (5 second each) and add a segment id to row_id? </p>\n<p>Like for 1 soundscape_xxxxxx.ogg file we will have 48 rows in the submission  (soundscape_xxxxxx_5, soundscape_xxxxxx_10, soundscape_xxxxxx_15, etc.) </p>",
      "rawMarkdown": "Could someone clarify how to form row_id from model predictions and test filenames **soundscape_xxxxxx.ogg** ?\n\nIn the description we have:\n\n>For each row_id, you should predict the probability that a given bird species was present. There is one column per bird species, so you will need to provide 182 predictions per row. Each row covers a five-second window of audio.\n\n>row_id: A slug of soundscape_[soundscape_id]_[end_time] for the prediction.\n\nand \n\n>test_soundscapes/ When you submit a notebook, the test_soundscapes directory will be populated with approximately 1,100 recordings to be used for scoring. They are 4 minutes long and in ogg audio format. The file names are randomized but have the general form of soundscape_xxxxxx.ogg\n\n\n\nShould split each 4 min long test file to 48 pieces (5 second each) and add a segment id to row_id? \n\nLike for 1 soundscape_xxxxxx.ogg file we will have 48 rows in the submission  (soundscape_xxxxxx_5, soundscape_xxxxxx_10, soundscape_xxxxxx_15, etc.)",
      "votes": null
    },
    {
      "id": "2736954",
      "postDate": "04/05/2024 13:44:48",
      "content": "<p>Yes, split a test file into 5-second chunks and add the end time of a chunk to the <code>row_id</code></p>",
      "rawMarkdown": "Yes, split a test file into 5-second chunks and add the end time of a chunk to the `row_id`",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2736954,
      "author_name": "hichambellafkir",
      "author_url": "",
      "post_date": "04/05/2024 13:44:48",
      "content": "<p>Yes, split a test file into 5-second chunks and add the end time of a chunk to the <code>row_id</code></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2736856": "Could someone clarify how to form row_id from model predictions and test filenames **soundscape_xxxxxx.ogg** ?\n\nIn the description we have:\n\n>For each row_id, you should predict the probability that a given bird species was present. There is one column per bird species, so you will need to provide 182 predictions per row. Each row covers a five-second window of audio.\n\n>row_id: A slug of soundscape_[soundscape_id]_[end_time] for the prediction.\n\nand \n\n>test_soundscapes/ When you submit a notebook, the test_soundscapes directory will be populated with approximately 1,100 recordings to be used for scoring. They are 4 minutes long and in ogg audio format. The file names are randomized but have the general form of soundscape_xxxxxx.ogg\n\n\n\nShould split each 4 min long test file to 48 pieces (5 second each) and add a segment id to row_id? \n\nLike for 1 soundscape_xxxxxx.ogg file we will have 48 rows in the submission  (soundscape_xxxxxx_5, soundscape_xxxxxx_10, soundscape_xxxxxx_15, etc.)",
    "2736954": "Yes, split a test file into 5-second chunks and add the end time of a chunk to the `row_id`"
  },
  "source": "meta"
}