{
  "fetched_at": "2026-10-19T07:43:44Z",
  "source_url": "https://www.kaggle.com/competitions/bird-watch-spring-2022/data",
  "pages": [
    {
      "name": "data-description",
      "content": "<p>You can use the train.csv and all images in the dataset to score as high as you can on the leaderboard. Below are descriptions of the data provided. </p>\n\n<h2>File descriptions</h2>\n<ul>\n<li><strong>train.csv</strong> - the training set, including labels</li>\n<li><strong>test.csv</strong> - the test set without labels</li>\n<li><strong>sample_submission.csv</strong> - a sample submission file in the correct format</li>\n<li><strong>metadata.csv</strong> - the actual classes of each label</li>\n</ul>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<h2>\n    Welcome to MIDS W251 team! You’re in the right place.\n</h2>\n<p>\n    This is the legendary Bird Watch competition – for you to dive into ML\n    competitions and familiarize yourself improving deep learning models.\n</p>\n<p>\n    The competition is simple: create a model that predicts which class of\n    species a bird is.\n</p>\n<p>\n    Once you’re ready to start competing, click on the Join the Competition\nbutton in the data page to create an account and gain access to the competition data. Then\n    check out the <a href=\"https://www.kaggle.com/darraghdog/beat-the-benchmark\">Beat the Benchmark</a> script in the code page for an examples\n    how to make your first submission!\n</p>\n<h2>\n    Overview of How Kaggle’s Competitions Work\n</h2>\n<p>\n    1. Join the Competition\n</p>\n<p>\n    Read about the challenge description, accept the Competition Rules and gain\n    access to the competition dataset.\n</p>\n<p>\n    2. Get to Work\n</p>\n<p>\n    Download the data, build models on it locally or on Kaggle Kernels (our\n    no-setup, customizable Jupyter Notebooks environment with free GPUs) and\n    generate a prediction file. See the Code page of the competition.\n</p>\n<p>\n    3. Make a Submission\n</p>\n<p>\n    Upload your prediction as a submission on Kaggle and receive an accuracy\n    score. Alternatively you can create a kaggle script in the code section of the compeition and submit the output of that script to the competition. To do this, you must write your predictions out to a file, as seen in the <a href=\"https://www.kaggle.com/darraghdog/beat-the-benchmark\">Beat the Benchmark</a> script, and save the script. This will run in the background, you will need to wait some time. Once finished, go to the output section of the script, select the output file, and press Submit. \n</p>\n<p>\n    4. Check the Leaderboard\n</p>\n<p>\n    See how your model ranks against other Kagglers on our leaderboard.\n</p>\n<p>\n    5. Improve Your Score\n</p>\n<h2>\n    What Data Will I Use in This Competition?\n</h2>\n<p>\n    In this competition, you’ll gain access to two similar datasets which\n    include image file. One dataset is titled `train.csv` and the other is\n    titled `test.csv`.\n</p>\n<p>\n    Train.csv will contain the filenames and will reveal the label of the bird\n    species, also known as the “ground truth”. The file metadata.csv can be\n    used to see the real species name of each of the 270 labels.\n</p>\n<p>\n    The `test.csv` dataset contains similar information but does not disclose\n    the “ground truth” for each passenger. It’s your job to predict these\n    outcomes.\n</p>\n<p>\n    Using the patterns you find in the train.csv image data, predict whether\n    the class of the test image files.\n</p>\n<p>\n    Check out the Data tab to explore the datasets even further. Once you feel\n    you’ve created a competitive model, submit it to Kaggle to see where your\n    model stands on our leaderboard against other Kagglers.\n</p>\n<h2>\n    How to Submit your Prediction to Kaggle\n</h2>\n<p>\n    Once you’re ready to make a submission and get on the leaderboard:\n</p>\n<p>\n    1. Click on the “Submit Predictions” button\n</p>\n<p>\n    2. Upload a CSV file in the submission file format. You’re able to submit 5\n    submissions a day.\n</p>\n<p>\n    Submission File Format:\n</p>\n<p>\n    For every image in the test dataset, submission files should contain two\n    columns: `filename` and `label`. Filename should be a string and label\n    should be an integer between 0 and 269.\n</p>\n<p>The file should contain a header and have the following format:</p>\n<pre>filename,label<br>test/6fbed82c07cd.jpg,0<br>test/d8399f767c45.jpg,0<br>...<br></pre>\n<h2>\n    A Last Word on Kaggle Notebooks\n</h2>\n<p>\n    As we mentioned before, Kaggle Notebooks is our no-setup, customizable,\n    Jupyter Notebooks environment with free GPUs and a huge repository of\n    community published data &amp;amp; code.\n</p>\n<p>\n    In every competition, you’ll find many Kernels publically shared with\n    incredible insights. It’s an invaluable resource worth becoming familiar\nwith. Check out this competition’s Kernel’s    <a href=\"https://www.kaggle.com/darraghdog/beat-the-benchmark\">here</a>.\n</p>\n<p>\n    To create a notebook, go to the Code page and create a new notebook. Make\n    sure to set GPU as your accelerator once created. Each person has\n    approximately 30 hours of GPU compute time each week, I believe it resets\n    on Fridays.\n</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<h2>Goal</h2>\nIt is your job to predict the `label`, or species, of each image in the test dataset.\nFor each image in the test set, you must predict a 0 to 269 value for the label, where 270 is the number of bird species. The bird species can be seen in the `metadata.csv` file on the data tab.<p></p>\n<p>The evaluation metric will be multi class accuracy, where discrete class labels must be predicted.</p>\n\n\n<h2>Submission Format</h2>\n<p><strong>For every image in the test dataset</strong>, submission files should contain two columns: `filename` and `label`. Filename should be a string and label should be an integer between 0 and 269.</p>\n<p>The file should contain a header and have the following format:</p>\n<pre>filename,label<br>test/6fbed82c07cd.jpg,0<br>test/d8399f767c45.jpg,0<br>...<br></pre>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "  - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}