{
  "id": 183197,
  "title": "First competition handle shake-up well. CV made the magic",
  "url": "/competitions/birdsong-recognition/discussion/183197",
  "author_name": "",
  "post_date": "2020-09-16T00:07:49.392787300Z",
  "votes": 4,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Thanks to my team <a href=\"https://www.kaggle.com/gopidurgaprasad\" target=\"_blank\">@gopidurgaprasad</a>. Thanks to everyone sharing your knowledge help us try all different combinations. Our CV Staggery journey - <a href=\"https://www.kaggle.com/c/birdsong-recognition/discussion/183209\" target=\"_blank\">https://www.kaggle.com/c/birdsong-recognition/discussion/183209</a><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F761268%2F3ba3c17c8daef10d0b38eb977534d4dd%2FScreenshot%202020-09-16%20at%205.41.59%20AM.png?generation=1600215220181847&amp;alt=media\" alt=\"\n\"></p>\n<p><strong>Inspiration</strong><br>\n<img src=\"https://1.bp.blogspot.com/-FZa1l-V1xUk/Xz2Mus9zzUI/AAAAAAAAGas/pcMTYi0RCf41dzn9evv4D3mxgy-R9ih_ACLcBGAsYHQ/s830/image2.gif\" alt=\"\"><br>\nPaper: <a href=\"https://arxiv.org/abs/1911.09781\" target=\"_blank\">Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels</a><br>\nCode: <a href=\"https://github.com/google-research/google-research/tree/master/mentormix\" target=\"_blank\">https://github.com/google-research/google-research/tree/master/mentormix</a> ( not used, go through above discussion to know our cv staggery and code )</p>",
  "messages": [
    {
      "id": "1012130",
      "postDate": "09/16/2020 00:07:49",
      "content": "<p>Thanks to my team <a href=\"https://www.kaggle.com/gopidurgaprasad\" target=\"_blank\">@gopidurgaprasad</a>. Thanks to everyone sharing your knowledge help us try all different combinations. Our CV Staggery journey - <a href=\"https://www.kaggle.com/c/birdsong-recognition/discussion/183209\" target=\"_blank\">https://www.kaggle.com/c/birdsong-recognition/discussion/183209</a><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F761268%2F3ba3c17c8daef10d0b38eb977534d4dd%2FScreenshot%202020-09-16%20at%205.41.59%20AM.png?generation=1600215220181847&amp;alt=media\" alt=\"\n\"></p>\n<p><strong>Inspiration</strong><br>\n<img src=\"https://1.bp.blogspot.com/-FZa1l-V1xUk/Xz2Mus9zzUI/AAAAAAAAGas/pcMTYi0RCf41dzn9evv4D3mxgy-R9ih_ACLcBGAsYHQ/s830/image2.gif\" alt=\"\"><br>\nPaper: <a href=\"https://arxiv.org/abs/1911.09781\" target=\"_blank\">Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels</a><br>\nCode: <a href=\"https://github.com/google-research/google-research/tree/master/mentormix\" target=\"_blank\">https://github.com/google-research/google-research/tree/master/mentormix</a> ( not used, go through above discussion to know our cv staggery and code )</p>",
      "rawMarkdown": "Thanks to my team @gopidurgaprasad. Thanks to everyone sharing your knowledge help us try all different combinations. Our CV Staggery journey - https://www.kaggle.com/c/birdsong-recognition/discussion/183209\n![\n](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F761268%2F3ba3c17c8daef10d0b38eb977534d4dd%2FScreenshot%202020-09-16%20at%205.41.59%20AM.png?generation=1600215220181847&alt=media)\n\n**Inspiration**\n![](https://1.bp.blogspot.com/-FZa1l-V1xUk/Xz2Mus9zzUI/AAAAAAAAGas/pcMTYi0RCf41dzn9evv4D3mxgy-R9ih_ACLcBGAsYHQ/s830/image2.gif)\nPaper: [Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels](https://arxiv.org/abs/1911.09781)\nCode: https://github.com/google-research/google-research/tree/master/mentormix ( not used, go through above discussion to know our cv staggery and code )",
      "votes": null
    },
    {
      "id": "1012132",
      "postDate": "09/16/2020 00:10:11",
      "content": "<p>Thanks for participating in our challenge! Looking forward to learning more about what did and didn't work for you!</p>",
      "rawMarkdown": "Thanks for participating in our challenge! Looking forward to learning more about what did and didn't work for you!",
      "votes": null
    },
    {
      "id": "1012152",
      "postDate": "09/16/2020 00:24:22",
      "content": "<p>You really have balls to choose that 0.533LB ! Well deserved the good result! <br>\nCant wait to see yr CV strategy</p>",
      "rawMarkdown": "You really have balls to choose that 0.533LB ! Well deserved the good result! \nCant wait to see yr CV strategy",
      "votes": null
    },
    {
      "id": "1012181",
      "postDate": "09/16/2020 00:41:10",
      "content": "<p>my team mate shared (<a href=\"https://www.kaggle.com/c/birdsong-recognition/discussion/183015\" target=\"_blank\">https://www.kaggle.com/c/birdsong-recognition/discussion/183015</a>) filter stage 3 dataset for training the model. we will share our final CV strategy soon. Thank you and congrats to all</p>",
      "rawMarkdown": "my team mate shared (https://www.kaggle.com/c/birdsong-recognition/discussion/183015) filter stage 3 dataset for training the model. we will share our final CV strategy soon. Thank you and congrats to all",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1012132,
      "author_name": "holgerklinck",
      "author_url": "",
      "post_date": "09/16/2020 00:10:11",
      "content": "<p>Thanks for participating in our challenge! Looking forward to learning more about what did and didn't work for you!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1012152,
      "author_name": "fiyeroleung",
      "author_url": "",
      "post_date": "09/16/2020 00:24:22",
      "content": "<p>You really have balls to choose that 0.533LB ! Well deserved the good result! <br>\nCant wait to see yr CV strategy</p>",
      "votes": null,
      "replies": [
        {
          "id": 1012181,
          "author_name": "seshurajup",
          "author_url": "",
          "post_date": "09/16/2020 00:41:10",
          "content": "<p>my team mate shared (<a href=\"https://www.kaggle.com/c/birdsong-recognition/discussion/183015\" target=\"_blank\">https://www.kaggle.com/c/birdsong-recognition/discussion/183015</a>) filter stage 3 dataset for training the model. we will share our final CV strategy soon. Thank you and congrats to all</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1012130": "Thanks to my team @gopidurgaprasad. Thanks to everyone sharing your knowledge help us try all different combinations. Our CV Staggery journey - https://www.kaggle.com/c/birdsong-recognition/discussion/183209\n![\n](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F761268%2F3ba3c17c8daef10d0b38eb977534d4dd%2FScreenshot%202020-09-16%20at%205.41.59%20AM.png?generation=1600215220181847&alt=media)\n\n**Inspiration**\n![](https://1.bp.blogspot.com/-FZa1l-V1xUk/Xz2Mus9zzUI/AAAAAAAAGas/pcMTYi0RCf41dzn9evv4D3mxgy-R9ih_ACLcBGAsYHQ/s830/image2.gif)\nPaper: [Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels](https://arxiv.org/abs/1911.09781)\nCode: https://github.com/google-research/google-research/tree/master/mentormix ( not used, go through above discussion to know our cv staggery and code )",
    "1012132": "Thanks for participating in our challenge! Looking forward to learning more about what did and didn't work for you!",
    "1012152": "You really have balls to choose that 0.533LB ! Well deserved the good result! \nCant wait to see yr CV strategy",
    "1012181": "my team mate shared (https://www.kaggle.com/c/birdsong-recognition/discussion/183015) filter stage 3 dataset for training the model. we will share our final CV strategy soon. Thank you and congrats to all"
  },
  "source": "meta"
}