{
  "id": 307628,
  "title": "Lost my silver medal and my first lesson on trusting public LB 😥",
  "url": "/competitions/tensorflow-great-barrier-reef/writeups/bao-loc-pham-lost-my-silver-medal-and-my-first-les",
  "author_name": "",
  "post_date": "2022-02-15T03:11:54.273336800Z",
  "votes": 17,
  "comment_count": 11,
  "views": 0,
  "content": "<p><img src=\"https://drive.google.com/file/d/1LSvpkQHzLCNZRbFoMmS_MvkOeXpYlcY-/view?usp=sharing\" alt=\"img\"></p>\n<p>I'm a newbie to kaggle but recently try a lot of submit in order to get my first medal but turn out my method is wrong. </p>\n<ul>\n<li>I chose my best lb instead of my cv, because i only test my model on training time.</li>\n<li>I didn't test my ensembles to get cv score.<br>\nResult: I'm got kicked out of medal zone even though I got some kernels goes up to silver zone.<br>\nAfter high score kernels got publish, I'm like going on roller coaster of emotion because of rapidly changes in lb, now i got my lesson on create a proper validating your cv.</li>\n</ul>",
  "messages": [
    {
      "id": "1690620",
      "postDate": "02/15/2022 03:11:54",
      "content": "<p><img src=\"https://drive.google.com/file/d/1LSvpkQHzLCNZRbFoMmS_MvkOeXpYlcY-/view?usp=sharing\" alt=\"img\"></p>\n<p>I'm a newbie to kaggle but recently try a lot of submit in order to get my first medal but turn out my method is wrong. </p>\n<ul>\n<li>I chose my best lb instead of my cv, because i only test my model on training time.</li>\n<li>I didn't test my ensembles to get cv score.<br>\nResult: I'm got kicked out of medal zone even though I got some kernels goes up to silver zone.<br>\nAfter high score kernels got publish, I'm like going on roller coaster of emotion because of rapidly changes in lb, now i got my lesson on create a proper validating your cv.</li>\n</ul>",
      "rawMarkdown": "![img](https://drive.google.com/file/d/1LSvpkQHzLCNZRbFoMmS_MvkOeXpYlcY-/view?usp=sharing)\n\nI'm a newbie to kaggle but recently try a lot of submit in order to get my first medal but turn out my method is wrong. \n- I chose my best lb instead of my cv, because i only test my model on training time.\n- I didn't test my ensembles to get cv score.\nResult: I'm got kicked out of medal zone even though I got some kernels goes up to silver zone.\nAfter high score kernels got publish, I'm like going on roller coaster of emotion because of rapidly changes in lb, now i got my lesson on create a proper validating your cv.",
      "votes": null
    },
    {
      "id": "1690628",
      "postDate": "02/15/2022 03:19:45",
      "content": "<p>Me too. The first time to deal with data inconsistency in kaggle for me.  This is a valuable lesson.</p>",
      "rawMarkdown": "Me too. The first time to deal with data inconsistency in kaggle for me.  This is a valuable lesson.",
      "votes": null
    },
    {
      "id": "1690637",
      "postDate": "02/15/2022 03:29:59",
      "content": "<p>Same here. I lucked out and one of my other models landed a bronze medal, but this will be a nice lesson for future competitions down the road.</p>",
      "rawMarkdown": "Same here. I lucked out and one of my other models landed a bronze medal, but this will be a nice lesson for future competitions down the road.",
      "votes": null
    },
    {
      "id": "1690649",
      "postDate": "02/15/2022 03:40:34",
      "content": "<p>Many of my submits with high resolution were also overfitted to public LB. But I am so lucky because one of four was submitted with original shape. We should use the same shape as training shape…</p>",
      "rawMarkdown": "Many of my submits with high resolution were also overfitted to public LB. But I am so lucky because one of four was submitted with original shape. We should use the same shape as training shape...",
      "votes": null
    },
    {
      "id": "1690671",
      "postDate": "02/15/2022 03:52:43",
      "content": "<p>Like you, I mistakenly didn't choose to be in the silver medal. I believe everyone will become a master in the future!</p>",
      "rawMarkdown": "Like you, I mistakenly didn't choose to be in the silver medal. I believe everyone will become a master in the future!",
      "votes": null
    },
    {
      "id": "1690693",
      "postDate": "02/15/2022 04:08:04",
      "content": "<p>Keep working for the next competition <a href=\"https://www.kaggle.com/locbaop\" target=\"_blank\">@locbaop</a> </p>",
      "rawMarkdown": "Keep working for the next competition @locbaop",
      "votes": null
    },
    {
      "id": "1690708",
      "postDate": "02/15/2022 04:16:47",
      "content": "<p>The first time I had faced a shakeup, I was so sad and didn't come back to comp for many months. I hope you take the learning and win 🏅 s in the future :)</p>",
      "rawMarkdown": "The first time I had faced a shakeup, I was so sad and didn't come back to comp for many months. I hope you take the learning and win 🏅 s in the future :)",
      "votes": null
    },
    {
      "id": "1690833",
      "postDate": "02/15/2022 05:54:36",
      "content": "<p>It was only very late in the competition that I learned there's only 5 video. 3 for training, 1 public lb, 1 private lb. That was a wake up call for me, telling me I shouldn't pick based on public lb alone. I'd be in the shake down hell were it not for that one comment I read less than 36 hours away from deadline.</p>",
      "rawMarkdown": "It was only very late in the competition that I learned there's only 5 video. 3 for training, 1 public lb, 1 private lb. That was a wake up call for me, telling me I shouldn't pick based on public lb alone. I'd be in the shake down hell were it not for that one comment I read less than 36 hours away from deadline.",
      "votes": null
    },
    {
      "id": "1690860",
      "postDate": "02/15/2022 06:13:35",
      "content": "<p>I think many people experienced something similar for this competition. Have submissions that would have done well on the private lb but didn't choose to submit it.</p>",
      "rawMarkdown": "I think many people experienced something similar for this competition. Have submissions that would have done well on the private lb but didn't choose to submit it.",
      "votes": null
    },
    {
      "id": "1690865",
      "postDate": "02/15/2022 06:16:51",
      "content": "<p>it's bitter, but this lesson we as a ML engineer can not forget when building products. Btw, your performance in public LB inspires me.</p>",
      "rawMarkdown": "it's bitter, but this lesson we as a ML engineer can not forget when building products. Btw, your performance in public LB inspires me.",
      "votes": null
    },
    {
      "id": "1692727",
      "postDate": "02/16/2022 07:52:58",
      "content": "<p>Learning from mistake is the best! Enjoy  </p>",
      "rawMarkdown": "Learning from mistake is the best! Enjoy",
      "votes": null
    },
    {
      "id": "1695518",
      "postDate": "02/18/2022 07:38:01",
      "content": "<p>Most importantly, don't give up! Don't be afraid doing mistakes, because only mistakes, made by you is the general lesson! Good luck, <a href=\"https://www.kaggle.com/locbaop\" target=\"_blank\">@locbaop</a>! </p>",
      "rawMarkdown": "Most importantly, don't give up! Don't be afraid doing mistakes, because only mistakes, made by you is the general lesson! Good luck, @locbaop!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1690628,
      "author_name": "xinjiali",
      "author_url": "",
      "post_date": "02/15/2022 03:19:45",
      "content": "<p>Me too. The first time to deal with data inconsistency in kaggle for me.  This is a valuable lesson.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690637,
      "author_name": "lueznaybors",
      "author_url": "",
      "post_date": "02/15/2022 03:29:59",
      "content": "<p>Same here. I lucked out and one of my other models landed a bronze medal, but this will be a nice lesson for future competitions down the road.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690649,
      "author_name": "ttkagglett",
      "author_url": "",
      "post_date": "02/15/2022 03:40:34",
      "content": "<p>Many of my submits with high resolution were also overfitted to public LB. But I am so lucky because one of four was submitted with original shape. We should use the same shape as training shape…</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690671,
      "author_name": "fxalll",
      "author_url": "",
      "post_date": "02/15/2022 03:52:43",
      "content": "<p>Like you, I mistakenly didn't choose to be in the silver medal. I believe everyone will become a master in the future!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690693,
      "author_name": "duykhanh99",
      "author_url": "",
      "post_date": "02/15/2022 04:08:04",
      "content": "<p>Keep working for the next competition <a href=\"https://www.kaggle.com/locbaop\" target=\"_blank\">@locbaop</a> </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690708,
      "author_name": "init27",
      "author_url": "",
      "post_date": "02/15/2022 04:16:47",
      "content": "<p>The first time I had faced a shakeup, I was so sad and didn't come back to comp for many months. I hope you take the learning and win 🏅 s in the future :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690833,
      "author_name": "liminchen1",
      "author_url": "",
      "post_date": "02/15/2022 05:54:36",
      "content": "<p>It was only very late in the competition that I learned there's only 5 video. 3 for training, 1 public lb, 1 private lb. That was a wake up call for me, telling me I shouldn't pick based on public lb alone. I'd be in the shake down hell were it not for that one comment I read less than 36 hours away from deadline.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690860,
      "author_name": "kuanweichen",
      "author_url": "",
      "post_date": "02/15/2022 06:13:35",
      "content": "<p>I think many people experienced something similar for this competition. Have submissions that would have done well on the private lb but didn't choose to submit it.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1690865,
      "author_name": "aengusng",
      "author_url": "",
      "post_date": "02/15/2022 06:16:51",
      "content": "<p>it's bitter, but this lesson we as a ML engineer can not forget when building products. Btw, your performance in public LB inspires me.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1692727,
      "author_name": "phanttan",
      "author_url": "",
      "post_date": "02/16/2022 07:52:58",
      "content": "<p>Learning from mistake is the best! Enjoy  </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1695518,
      "author_name": "vad13irt",
      "author_url": "",
      "post_date": "02/18/2022 07:38:01",
      "content": "<p>Most importantly, don't give up! Don't be afraid doing mistakes, because only mistakes, made by you is the general lesson! Good luck, <a href=\"https://www.kaggle.com/locbaop\" target=\"_blank\">@locbaop</a>! </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1690620": "![img](https://drive.google.com/file/d/1LSvpkQHzLCNZRbFoMmS_MvkOeXpYlcY-/view?usp=sharing)\n\nI'm a newbie to kaggle but recently try a lot of submit in order to get my first medal but turn out my method is wrong. \n- I chose my best lb instead of my cv, because i only test my model on training time.\n- I didn't test my ensembles to get cv score.\nResult: I'm got kicked out of medal zone even though I got some kernels goes up to silver zone.\nAfter high score kernels got publish, I'm like going on roller coaster of emotion because of rapidly changes in lb, now i got my lesson on create a proper validating your cv.",
    "1690628": "Me too. The first time to deal with data inconsistency in kaggle for me.  This is a valuable lesson.",
    "1690637": "Same here. I lucked out and one of my other models landed a bronze medal, but this will be a nice lesson for future competitions down the road.",
    "1690649": "Many of my submits with high resolution were also overfitted to public LB. But I am so lucky because one of four was submitted with original shape. We should use the same shape as training shape...",
    "1690671": "Like you, I mistakenly didn't choose to be in the silver medal. I believe everyone will become a master in the future!",
    "1690693": "Keep working for the next competition @locbaop",
    "1690708": "The first time I had faced a shakeup, I was so sad and didn't come back to comp for many months. I hope you take the learning and win 🏅 s in the future :)",
    "1690833": "It was only very late in the competition that I learned there's only 5 video. 3 for training, 1 public lb, 1 private lb. That was a wake up call for me, telling me I shouldn't pick based on public lb alone. I'd be in the shake down hell were it not for that one comment I read less than 36 hours away from deadline.",
    "1690860": "I think many people experienced something similar for this competition. Have submissions that would have done well on the private lb but didn't choose to submit it.",
    "1690865": "it's bitter, but this lesson we as a ML engineer can not forget when building products. Btw, your performance in public LB inspires me.",
    "1692727": "Learning from mistake is the best! Enjoy",
    "1695518": "Most importantly, don't give up! Don't be afraid doing mistakes, because only mistakes, made by you is the general lesson! Good luck, @locbaop!"
  },
  "source": "meta"
}