{
  "id": 226590,
  "title": "How to shake down from 28 to 80 ))",
  "url": "/competitions/ranzcr-clip-catheter-line-classification/writeups/best-overfitting-how-to-shake-down-from-28-to-80",
  "author_name": "",
  "post_date": "2021-03-17T12:31:22.833Z",
  "votes": 34,
  "comment_count": 20,
  "views": 0,
  "content": "<p>Yesterday I was really lost between three submissions (HIGH LB, HIGH CV, ROBUST MODEL). So I decided to choose the HIGH LB as my first submission But what about my second choice??. I was really confused by the decision of allowing all external data even NIH, which leads me to the possibility that the private testing set has a dependent type of data with a different distribution of classes. Remembering this discussion <a href=\"https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/216219\" target=\"_blank\">https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/216219</a>. In this case, the BEST CV will not help so I went with the ROBUST MODEL choice leading me to a bronze medal instead of a silver one. :((<br>\n Well, It was a really interesting competition and I really learned a lot from all of you guys. I am really glad to hear that my work also helped some of you. <br>\nFor those who are interested in the generation process of the starting points <a href=\"https://www.kaggle.com/morizin\" target=\"_blank\">@morizin</a> , <a href=\"https://www.kaggle.com/projdev\" target=\"_blank\">@projdev</a> you can find the notebook here <a href=\"https://www.kaggle.com/ammarali32/starting-points-generation\" target=\"_blank\">https://www.kaggle.com/ammarali32/starting-points-generation</a> Credits to <a href=\"https://www.kaggle.com/yasufuminakama\" target=\"_blank\">@yasufuminakama</a> implementation. <br>\nUPD: Great thanks to my teammate and my girlfriend <a href=\"https://www.kaggle.com/walaaothman\" target=\"_blank\">@walaaothman</a> </p>",
  "messages": [
    {
      "id": "1241432",
      "postDate": "03/17/2021 03:28:01",
      "content": "<p>Yesterday I was really lost between three submissions (HIGH LB, HIGH CV, ROBUST MODEL). So I decided to choose the HIGH LB as my first submission But what about my second choice??. I was really confused by the decision of allowing all external data even NIH, which leads me to the possibility that the private testing set has a dependent type of data with a different distribution of classes. Remembering this discussion <a href=\"https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/216219\" target=\"_blank\">https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/216219</a>. In this case, the BEST CV will not help so I went with the ROBUST MODEL choice leading me to a bronze medal instead of a silver one. :((<br>\n Well, It was a really interesting competition and I really learned a lot from all of you guys. I am really glad to hear that my work also helped some of you. <br>\nFor those who are interested in the generation process of the starting points <a href=\"https://www.kaggle.com/morizin\" target=\"_blank\">@morizin</a> , <a href=\"https://www.kaggle.com/projdev\" target=\"_blank\">@projdev</a> you can find the notebook here <a href=\"https://www.kaggle.com/ammarali32/starting-points-generation\" target=\"_blank\">https://www.kaggle.com/ammarali32/starting-points-generation</a> Credits to <a href=\"https://www.kaggle.com/yasufuminakama\" target=\"_blank\">@yasufuminakama</a> implementation. <br>\nUPD: Great thanks to my teammate and my girlfriend <a href=\"https://www.kaggle.com/walaaothman\" target=\"_blank\">@walaaothman</a> </p>",
      "rawMarkdown": "Yesterday I was really lost between three submissions (HIGH LB, HIGH CV, ROBUST MODEL). So I decided to choose the HIGH LB as my first submission But what about my second choice??. I was really confused by the decision of allowing all external data even NIH, which leads me to the possibility that the private testing set has a dependent type of data with a different distribution of classes. Remembering this discussion https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/216219. In this case, the BEST CV will not help so I went with the ROBUST MODEL choice leading me to a bronze medal instead of a silver one. :((\n Well, It was a really interesting competition and I really learned a lot from all of you guys. I am really glad to hear that my work also helped some of you. \nFor those who are interested in the generation process of the starting points @morizin , @projdev you can find the notebook here https://www.kaggle.com/ammarali32/starting-points-generation Credits to @yasufuminakama implementation. \nUPD: Great thanks to my teammate and my girlfriend @walaaothman",
      "votes": null
    },
    {
      "id": "1241446",
      "postDate": "03/17/2021 03:35:58",
      "content": "<p>I am sorry to hear that you experienced such a shakedown - 80th is still an impressive leaderboard position. Many Kagglers benefited from your contributions during this competition, and for that, you should be proud. You'll get that silver / gold next time.</p>",
      "rawMarkdown": "I am sorry to hear that you experienced such a shakedown - 80th is still an impressive leaderboard position. Many Kagglers benefited from your contributions during this competition, and for that, you should be proud. You'll get that silver / gold next time.",
      "votes": null
    },
    {
      "id": "1241452",
      "postDate": "03/17/2021 03:43:39",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> thank you. I think the notebook is still in private.</p>",
      "rawMarkdown": "ammarali32 thank you. I think the notebook is still in private.",
      "votes": null
    },
    {
      "id": "1241457",
      "postDate": "03/17/2021 03:49:18",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> thank you. Sorry to hear that a shakedown.</p>",
      "rawMarkdown": "ammarali32 thank you. Sorry to hear that a shakedown.",
      "votes": null
    },
    {
      "id": "1241465",
      "postDate": "03/17/2021 03:58:42",
      "content": "<p>Thanks and congrats for your high standing hope you will get gold at the end</p>",
      "rawMarkdown": "Thanks and congrats for your high standing hope you will get gold at the end",
      "votes": null
    },
    {
      "id": "1241466",
      "postDate": "03/17/2021 03:59:02",
      "content": "<p>Welcome. Sorry for that fixed and congrats ))</p>",
      "rawMarkdown": "Welcome. Sorry for that fixed and congrats ))",
      "votes": null
    },
    {
      "id": "1241468",
      "postDate": "03/17/2021 03:59:56",
      "content": "<p>Thanks a lot for the motivating comment)) and congrats</p>",
      "rawMarkdown": "Thanks a lot for the motivating comment)) and congrats",
      "votes": null
    },
    {
      "id": "1241478",
      "postDate": "03/17/2021 04:13:21",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> as you review some of the discussions for how they placed, what they used in ensembles, for models, etc., think you will see how much your work here helped many people.  So kind of a silver or gold by proxy.  Hope that makes you feel kind of proud.  And next time it will be yours. </p>",
      "rawMarkdown": "ammarali32 as you review some of the discussions for how they placed, what they used in ensembles, for models, etc., think you will see how much your work here helped many people.  So kind of a silver or gold by proxy.  Hope that makes you feel kind of proud.  And next time it will be yours.",
      "votes": null
    },
    {
      "id": "1241502",
      "postDate": "03/17/2021 04:28:48",
      "content": "<p>Thanks)) that is really motivating. </p>",
      "rawMarkdown": "Thanks)) that is really motivating.",
      "votes": null
    },
    {
      "id": "1241508",
      "postDate": "03/17/2021 04:34:06",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> I am so sorry for your shake down. I would like to say much of thank you for your contribution to RANZCR.</p>",
      "rawMarkdown": "ammarali32 I am so sorry for your shake down. I would like to say much of thank you for your contribution to RANZCR.",
      "votes": null
    },
    {
      "id": "1241520",
      "postDate": "03/17/2021 04:46:37",
      "content": "<p>Thanks a lot, I am really glad to hear that ))</p>",
      "rawMarkdown": "Thanks a lot, I am really glad to hear that ))",
      "votes": null
    },
    {
      "id": "1241680",
      "postDate": "03/17/2021 06:42:22",
      "content": "<p>Sorry about the shakedown. I am one of those who benefit from your work, since the starting point you shared gave me better score than imagenet pretrained. I would not have gotten my score without your work, so thank you and hope you have better luck next time. </p>",
      "rawMarkdown": "Sorry about the shakedown. I am one of those who benefit from your work, since the starting point you shared gave me better score than imagenet pretrained. I would not have gotten my score without your work, so thank you and hope you have better luck next time.",
      "votes": null
    },
    {
      "id": "1241943",
      "postDate": "03/17/2021 09:47:23",
      "content": "<p>Thanks a lot ))</p>",
      "rawMarkdown": "Thanks a lot ))",
      "votes": null
    },
    {
      "id": "1242021",
      "postDate": "03/17/2021 10:53:09",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> Sorry about the shakedown . Your contributions were critical to this competition . Wishing you the best to win Gold in the next competition </p>",
      "rawMarkdown": "ammarali32 Sorry about the shakedown . Your contributions were critical to this competition . Wishing you the best to win Gold in the next competition",
      "votes": null
    },
    {
      "id": "1242137",
      "postDate": "03/17/2021 12:30:26",
      "content": "<p>Thanks. I really appreciate that and hope that competition will not be so far. </p>",
      "rawMarkdown": "Thanks. I really appreciate that and hope that competition will not be so far.",
      "votes": null
    },
    {
      "id": "1242260",
      "postDate": "03/17/2021 13:58:33",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> I am sure you will earn the well-deserved silver/gold in the next competitions! Joining the big crowd of people who benefited from your ideas and contributions during the course of this competition. Привет Петербургу! </p>",
      "rawMarkdown": "ammarali32 I am sure you will earn the well-deserved silver/gold in the next competitions! Joining the big crowd of people who benefited from your ideas and contributions during the course of this competition. Привет Петербургу!",
      "votes": null
    },
    {
      "id": "1242306",
      "postDate": "03/17/2021 14:36:52",
      "content": "<p>I'm sorry for your shake down …</p>\n<p>I'd have given up this competition if there weren't your discussion/notebook/dataset.  So I'd like to say great thanks.</p>",
      "rawMarkdown": "I'm sorry for your shake down ...\n\nI'd have given up this competition if there weren't your discussion/notebook/dataset.  So I'd like to say great thanks.",
      "votes": null
    },
    {
      "id": "1242333",
      "postDate": "03/17/2021 14:52:20",
      "content": "<p>Thanks)) I really appreciate it. It is ok my happiness about winning first place in the MagNet competition \"Driven Data\" still some outcome from these three months. My fault was going on a lot of competitions during my Master's degree and a full-time job.  You really did a great job and I really wanted to send you a team request but we had submissions over the limit. hope we will be in a competition together one day. Good Luck  </p>",
      "rawMarkdown": "Thanks)) I really appreciate it. It is ok my happiness about winning first place in the MagNet competition \"Driven Data\" still some outcome from these three months. My fault was going on a lot of competitions during my Master's degree and a full-time job.  You really did a great job and I really wanted to send you a team request but we had submissions over the limit. hope we will be in a competition together one day. Good Luck",
      "votes": null
    },
    {
      "id": "1242337",
      "postDate": "03/17/2021 14:54:12",
      "content": "<p>большое спасибо за вашу поддержку. I hope so , congratulations )) </p>",
      "rawMarkdown": "большое спасибо за вашу поддержку. I hope so , congratulations ))",
      "votes": null
    },
    {
      "id": "1242859",
      "postDate": "03/17/2021 21:31:07",
      "content": "<p>So sorry for your shake down. </p>\n<p>Your contributions helped us a lot. Thanks for the amazing work. I'm sure that you will win silvers/golds in the upcoming challenges. Keep up the great work. 🙏</p>",
      "rawMarkdown": "So sorry for your shake down. \n\nYour contributions helped us a lot. Thanks for the amazing work. I'm sure that you will win silvers/golds in the upcoming challenges. Keep up the great work. 🙏",
      "votes": null
    },
    {
      "id": "1242950",
      "postDate": "03/17/2021 23:16:32",
      "content": "<p>This competition was fun despite the shakedown. I have learned a lot from your ideas.<br>\nThank you very much for coaching me. I am sure that you will get gold next time.</p>",
      "rawMarkdown": "This competition was fun despite the shakedown. I have learned a lot from your ideas.\nThank you very much for coaching me. I am sure that you will get gold next time.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1241446,
      "author_name": "tuckerarrants",
      "author_url": "",
      "post_date": "03/17/2021 03:35:58",
      "content": "<p>I am sorry to hear that you experienced such a shakedown - 80th is still an impressive leaderboard position. Many Kagglers benefited from your contributions during this competition, and for that, you should be proud. You'll get that silver / gold next time.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1241468,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 03:59:56",
          "content": "<p>Thanks a lot for the motivating comment)) and congrats</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1241452,
      "author_name": "projdev",
      "author_url": "",
      "post_date": "03/17/2021 03:43:39",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> thank you. I think the notebook is still in private.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1241466,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 03:59:02",
          "content": "<p>Welcome. Sorry for that fixed and congrats ))</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1241457,
      "author_name": "morizin",
      "author_url": "",
      "post_date": "03/17/2021 03:49:18",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> thank you. Sorry to hear that a shakedown.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1241465,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 03:58:42",
          "content": "<p>Thanks and congrats for your high standing hope you will get gold at the end</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1241478,
      "author_name": "something4kag",
      "author_url": "",
      "post_date": "03/17/2021 04:13:21",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> as you review some of the discussions for how they placed, what they used in ensembles, for models, etc., think you will see how much your work here helped many people.  So kind of a silver or gold by proxy.  Hope that makes you feel kind of proud.  And next time it will be yours. </p>",
      "votes": null,
      "replies": [
        {
          "id": 1241502,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 04:28:48",
          "content": "<p>Thanks)) that is really motivating. </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1241508,
      "author_name": "yoshitaka1105",
      "author_url": "",
      "post_date": "03/17/2021 04:34:06",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> I am so sorry for your shake down. I would like to say much of thank you for your contribution to RANZCR.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1241520,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 04:46:37",
          "content": "<p>Thanks a lot, I am really glad to hear that ))</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1241680,
      "author_name": "nyleve",
      "author_url": "",
      "post_date": "03/17/2021 06:42:22",
      "content": "<p>Sorry about the shakedown. I am one of those who benefit from your work, since the starting point you shared gave me better score than imagenet pretrained. I would not have gotten my score without your work, so thank you and hope you have better luck next time. </p>",
      "votes": null,
      "replies": [
        {
          "id": 1241943,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 09:47:23",
          "content": "<p>Thanks a lot ))</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1242021,
      "author_name": "usharengaraju",
      "author_url": "",
      "post_date": "03/17/2021 10:53:09",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> Sorry about the shakedown . Your contributions were critical to this competition . Wishing you the best to win Gold in the next competition </p>",
      "votes": null,
      "replies": [
        {
          "id": 1242137,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 12:30:26",
          "content": "<p>Thanks. I really appreciate that and hope that competition will not be so far. </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1242260,
      "author_name": "kozodoi",
      "author_url": "",
      "post_date": "03/17/2021 13:58:33",
      "content": "<p><a href=\"https://www.kaggle.com/ammarali32\" target=\"_blank\">@ammarali32</a> I am sure you will earn the well-deserved silver/gold in the next competitions! Joining the big crowd of people who benefited from your ideas and contributions during the course of this competition. Привет Петербургу! </p>",
      "votes": null,
      "replies": [
        {
          "id": 1242337,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 14:54:12",
          "content": "<p>большое спасибо за вашу поддержку. I hope so , congratulations )) </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1242306,
      "author_name": "ttahara",
      "author_url": "",
      "post_date": "03/17/2021 14:36:52",
      "content": "<p>I'm sorry for your shake down …</p>\n<p>I'd have given up this competition if there weren't your discussion/notebook/dataset.  So I'd like to say great thanks.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1242333,
          "author_name": "ammarali32",
          "author_url": "",
          "post_date": "03/17/2021 14:52:20",
          "content": "<p>Thanks)) I really appreciate it. It is ok my happiness about winning first place in the MagNet competition \"Driven Data\" still some outcome from these three months. My fault was going on a lot of competitions during my Master's degree and a full-time job.  You really did a great job and I really wanted to send you a team request but we had submissions over the limit. hope we will be in a competition together one day. Good Luck  </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1242859,
      "author_name": "snnclsr",
      "author_url": "",
      "post_date": "03/17/2021 21:31:07",
      "content": "<p>So sorry for your shake down. </p>\n<p>Your contributions helped us a lot. Thanks for the amazing work. I'm sure that you will win silvers/golds in the upcoming challenges. Keep up the great work. 🙏</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1242950,
      "author_name": "walaaothman",
      "author_url": "",
      "post_date": "03/17/2021 23:16:32",
      "content": "<p>This competition was fun despite the shakedown. I have learned a lot from your ideas.<br>\nThank you very much for coaching me. I am sure that you will get gold next time.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1241432": "Yesterday I was really lost between three submissions (HIGH LB, HIGH CV, ROBUST MODEL). So I decided to choose the HIGH LB as my first submission But what about my second choice??. I was really confused by the decision of allowing all external data even NIH, which leads me to the possibility that the private testing set has a dependent type of data with a different distribution of classes. Remembering this discussion https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/216219. In this case, the BEST CV will not help so I went with the ROBUST MODEL choice leading me to a bronze medal instead of a silver one. :((\n Well, It was a really interesting competition and I really learned a lot from all of you guys. I am really glad to hear that my work also helped some of you. \nFor those who are interested in the generation process of the starting points @morizin , @projdev you can find the notebook here https://www.kaggle.com/ammarali32/starting-points-generation Credits to @yasufuminakama implementation. \nUPD: Great thanks to my teammate and my girlfriend @walaaothman",
    "1241446": "I am sorry to hear that you experienced such a shakedown - 80th is still an impressive leaderboard position. Many Kagglers benefited from your contributions during this competition, and for that, you should be proud. You'll get that silver / gold next time.",
    "1241452": "ammarali32 thank you. I think the notebook is still in private.",
    "1241457": "ammarali32 thank you. Sorry to hear that a shakedown.",
    "1241465": "Thanks and congrats for your high standing hope you will get gold at the end",
    "1241466": "Welcome. Sorry for that fixed and congrats ))",
    "1241468": "Thanks a lot for the motivating comment)) and congrats",
    "1241478": "ammarali32 as you review some of the discussions for how they placed, what they used in ensembles, for models, etc., think you will see how much your work here helped many people.  So kind of a silver or gold by proxy.  Hope that makes you feel kind of proud.  And next time it will be yours.",
    "1241502": "Thanks)) that is really motivating.",
    "1241508": "ammarali32 I am so sorry for your shake down. I would like to say much of thank you for your contribution to RANZCR.",
    "1241520": "Thanks a lot, I am really glad to hear that ))",
    "1241680": "Sorry about the shakedown. I am one of those who benefit from your work, since the starting point you shared gave me better score than imagenet pretrained. I would not have gotten my score without your work, so thank you and hope you have better luck next time.",
    "1241943": "Thanks a lot ))",
    "1242021": "ammarali32 Sorry about the shakedown . Your contributions were critical to this competition . Wishing you the best to win Gold in the next competition",
    "1242137": "Thanks. I really appreciate that and hope that competition will not be so far.",
    "1242260": "ammarali32 I am sure you will earn the well-deserved silver/gold in the next competitions! Joining the big crowd of people who benefited from your ideas and contributions during the course of this competition. Привет Петербургу!",
    "1242306": "I'm sorry for your shake down ...\n\nI'd have given up this competition if there weren't your discussion/notebook/dataset.  So I'd like to say great thanks.",
    "1242333": "Thanks)) I really appreciate it. It is ok my happiness about winning first place in the MagNet competition \"Driven Data\" still some outcome from these three months. My fault was going on a lot of competitions during my Master's degree and a full-time job.  You really did a great job and I really wanted to send you a team request but we had submissions over the limit. hope we will be in a competition together one day. Good Luck",
    "1242337": "большое спасибо за вашу поддержку. I hope so , congratulations ))",
    "1242859": "So sorry for your shake down. \n\nYour contributions helped us a lot. Thanks for the amazing work. I'm sure that you will win silvers/golds in the upcoming challenges. Keep up the great work. 🙏",
    "1242950": "This competition was fun despite the shakedown. I have learned a lot from your ideas.\nThank you very much for coaching me. I am sure that you will get gold next time."
  },
  "source": "meta"
}