{
  "id": 174226,
  "title": "What data do we use for submission?",
  "url": "/competitions/birdsong-recognition/discussion/174226",
  "author_name": "",
  "post_date": "2020-08-12T18:12:27.987969600Z",
  "votes": 2,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I'm a bit confused by the <code>Data</code> description page. I see the training data, which is well organized and understandable. </p>\n<p>But then it says that the test data is hidden. The only <code>test</code> files I see are two labeled \"example\". So where is the data that I perform inference on for my submission? </p>\n<p>This is my first \"real\" competition after doing a couple of the smaller beginner's comps. Perhaps this is my noobness speaking, but the explanation of the data is very convoluted and hard to understand. </p>",
  "messages": [
    {
      "id": "968140",
      "postDate": "08/12/2020 18:12:27",
      "content": "<p>I'm a bit confused by the <code>Data</code> description page. I see the training data, which is well organized and understandable. </p>\n<p>But then it says that the test data is hidden. The only <code>test</code> files I see are two labeled \"example\". So where is the data that I perform inference on for my submission? </p>\n<p>This is my first \"real\" competition after doing a couple of the smaller beginner's comps. Perhaps this is my noobness speaking, but the explanation of the data is very convoluted and hard to understand. </p>",
      "rawMarkdown": "I'm a bit confused by the `Data` description page. I see the training data, which is well organized and understandable. \n\nBut then it says that the test data is hidden. The only `test` files I see are two labeled \"example\". So where is the data that I perform inference on for my submission? \n\nThis is my first \"real\" competition after doing a couple of the smaller beginner's comps. Perhaps this is my noobness speaking, but the explanation of the data is very convoluted and hard to understand.",
      "votes": null
    },
    {
      "id": "968616",
      "postDate": "08/13/2020 06:25:56",
      "content": "<p>There have been a few posts in discussion about how to handle test audio and submission and the confusion on it, if you want to search for them.  Most people are just using  a dummy set of data to make sure their notebook generates a file.  <br>\nThis inference notebook can give you the idea of how to deal with generating a simple submission file so you can do submit once your inference notebook finishes (in output the Submit button will be next to submission.csv).  Submit will then run on the hidden test set -<br>\n<a href=\"https://www.kaggle.com/ttahara/inference-birdsong-baseline-resnest50-fast\" target=\"_blank\">https://www.kaggle.com/ttahara/inference-birdsong-baseline-resnest50-fast</a><br>\nsee code under section  \"read data\" <br>\nif test<em>audio does not exist use ../birdcall-check/test</em>audio - this is a dataset using audio clips from the start of train.  Use add data in your notebook to include it - <br>\n<a href=\"https://www.kaggle.com/shonenkov/birdcall-check\" target=\"_blank\">https://www.kaggle.com/shonenkov/birdcall-check</a><br>\nshonenkov has a notebook that created it if you want to look at that, copy and create a dataset of your own if you  prefer.</p>",
      "rawMarkdown": "There have been a few posts in discussion about how to handle test audio and submission and the confusion on it, if you want to search for them.  Most people are just using  a dummy set of data to make sure their notebook generates a file.  \n\nThis inference notebook can give you the idea of how to deal with generating a simple submission file so you can do submit once your inference notebook finishes (in output the Submit button will be next to submission.csv).  Submit will then run on the hidden test set -\n\nhttps://www.kaggle.com/ttahara/inference-birdsong-baseline-resnest50-fast\nsee code under section  \"read data\" \nif test_audio does not exist use ../birdcall-check/test_audio - this is a dataset using audio clips from the start of train.  Use add data in your notebook to include it - \nhttps://www.kaggle.com/shonenkov/birdcall-check\n\nshonenkov has a notebook that created it if you want to look at that, copy and create a dataset of your own if you  prefer.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 968616,
      "author_name": "something4kag",
      "author_url": "",
      "post_date": "08/13/2020 06:25:56",
      "content": "<p>There have been a few posts in discussion about how to handle test audio and submission and the confusion on it, if you want to search for them.  Most people are just using  a dummy set of data to make sure their notebook generates a file.  <br>\nThis inference notebook can give you the idea of how to deal with generating a simple submission file so you can do submit once your inference notebook finishes (in output the Submit button will be next to submission.csv).  Submit will then run on the hidden test set -<br>\n<a href=\"https://www.kaggle.com/ttahara/inference-birdsong-baseline-resnest50-fast\" target=\"_blank\">https://www.kaggle.com/ttahara/inference-birdsong-baseline-resnest50-fast</a><br>\nsee code under section  \"read data\" <br>\nif test<em>audio does not exist use ../birdcall-check/test</em>audio - this is a dataset using audio clips from the start of train.  Use add data in your notebook to include it - <br>\n<a href=\"https://www.kaggle.com/shonenkov/birdcall-check\" target=\"_blank\">https://www.kaggle.com/shonenkov/birdcall-check</a><br>\nshonenkov has a notebook that created it if you want to look at that, copy and create a dataset of your own if you  prefer.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "968140": "I'm a bit confused by the `Data` description page. I see the training data, which is well organized and understandable. \n\nBut then it says that the test data is hidden. The only `test` files I see are two labeled \"example\". So where is the data that I perform inference on for my submission? \n\nThis is my first \"real\" competition after doing a couple of the smaller beginner's comps. Perhaps this is my noobness speaking, but the explanation of the data is very convoluted and hard to understand.",
    "968616": "There have been a few posts in discussion about how to handle test audio and submission and the confusion on it, if you want to search for them.  Most people are just using  a dummy set of data to make sure their notebook generates a file.  \n\nThis inference notebook can give you the idea of how to deal with generating a simple submission file so you can do submit once your inference notebook finishes (in output the Submit button will be next to submission.csv).  Submit will then run on the hidden test set -\n\nhttps://www.kaggle.com/ttahara/inference-birdsong-baseline-resnest50-fast\nsee code under section  \"read data\" \nif test_audio does not exist use ../birdcall-check/test_audio - this is a dataset using audio clips from the start of train.  Use add data in your notebook to include it - \nhttps://www.kaggle.com/shonenkov/birdcall-check\n\nshonenkov has a notebook that created it if you want to look at that, copy and create a dataset of your own if you  prefer."
  },
  "source": "meta"
}