{
  "id": 261490,
  "title": "Solution to our first NLP competition🥉",
  "content": "<p>Thank you Kaggle and CommonLit for exciting competition!</p>\n<p>our team (cpptake nishipy)  was in the silver medal range in the public LB, but in the private LB, the shake down happened and we fall into the bronze medal range.<br>\nIt's frustrating, but we learned a lot, so we'll leave you with the solution</p>\n<p>I would like to get a better ranking in the next competition, so if you have any ideas for improvement, please comment.</p>\n<h3>Pipe Line &amp; ideas</h3>\n<h5>see the following notebook:</h5>\n<p><a href=\"https://www.kaggle.com/takeshikobayashi/bronze-medal-solution-roberta-stacking-ensemble\" target=\"_blank\">our final sumit notebook</a></p>\n<h3>reference¶</h3>\n<ul>\n<li><a href=\"https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-infer-3\" target=\"_blank\">https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-infer-3</a></li>\n<li><a href=\"https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-itpt\" target=\"_blank\">https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-itpt</a></li>\n<li><a href=\"https://www.kaggle.com/andretugan/pre-trained-roberta-solution-in-pytorch\" target=\"_blank\">https://www.kaggle.com/andretugan/pre-trained-roberta-solution-in-pytorch</a></li>\n<li><a href=\"https://www.kaggle.com/maunish/clrp-roberta-svm\" target=\"_blank\">https://www.kaggle.com/maunish/clrp-roberta-svm</a></li>\n<li><a href=\"https://www.kaggle.com/jcesquiveld/best-transformer-representations\" target=\"_blank\">https://www.kaggle.com/jcesquiveld/best-transformer-representations</a></li>\n<li><a href=\"https://www.kaggle.com/gilfernandes/commonlit-pytorch-t5-large-svm\" target=\"_blank\">https://www.kaggle.com/gilfernandes/commonlit-pytorch-t5-large-svm</a><br>\nThank you <a href=\"https://www.kaggle.com/rhtsingh\" target=\"_blank\">@rhtsingh</a> <a href=\"https://www.kaggle.com/andretugan\" target=\"_blank\">@andretugan</a>  <a href=\"https://www.kaggle.com/maunish\" target=\"_blank\">@maunish</a> <a href=\"https://www.kaggle.com/jcesquiveld\" target=\"_blank\">@jcesquiveld</a> <a href=\"https://www.kaggle.com/gilfernandes\" target=\"_blank\">@gilfernandes</a> and the comunity for sharing great notebooks!</li>\n</ul>",
  "comments": [
    {
      "id": 1449388,
      "author_name": "Sinan Calisir",
      "author_url": "",
      "post_date": "2021-08-04T21:26:19.207000",
      "content": "<p>Congrats on winning a medal in your first NLP competition!</p>",
      "votes": 3,
      "replies": [
        {
          "id": 1449608,
          "author_name": "cpptake",
          "author_url": "",
          "post_date": "2021-08-04T23:21:10.487000",
          "content": "<p>Thank you!</p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 1447617,
      "author_name": "Chris Deotte",
      "author_url": "",
      "post_date": "2021-08-04T15:27:57.207000",
      "content": "<p>Great job earning a medal in your first NLP competition!</p>",
      "votes": 3,
      "replies": [
        {
          "id": 1447994,
          "author_name": "cpptake",
          "author_url": "",
          "post_date": "2021-08-04T16:28:53.830000",
          "content": "<p>Thank you for reply!<br>\nWe will continue trying hard to get better place in competitions and to contribute for the Kaggle community</p>",
          "votes": 1,
          "replies": []
        }
      ]
    }
  ],
  "comment_count": 4,
  "messages": [
    {
      "id": 1447483,
      "postDate": "2021-08-04T15:02:40.870Z",
      "content": "<p>Thank you Kaggle and CommonLit for exciting competition!</p>\n<p>our team (cpptake nishipy)  was in the silver medal range in the public LB, but in the private LB, the shake down happened and we fall into the bronze medal range.<br>\nIt's frustrating, but we learned a lot, so we'll leave you with the solution</p>\n<p>I would like to get a better ranking in the next competition, so if you have any ideas for improvement, please comment.</p>\n<h3>Pipe Line &amp; ideas</h3>\n<h5>see the following notebook:</h5>\n<p><a href=\"https://www.kaggle.com/takeshikobayashi/bronze-medal-solution-roberta-stacking-ensemble\" target=\"_blank\">our final sumit notebook</a></p>\n<h3>reference¶</h3>\n<ul>\n<li><a href=\"https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-infer-3\" target=\"_blank\">https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-infer-3</a></li>\n<li><a href=\"https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-itpt\" target=\"_blank\">https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-itpt</a></li>\n<li><a href=\"https://www.kaggle.com/andretugan/pre-trained-roberta-solution-in-pytorch\" target=\"_blank\">https://www.kaggle.com/andretugan/pre-trained-roberta-solution-in-pytorch</a></li>\n<li><a href=\"https://www.kaggle.com/maunish/clrp-roberta-svm\" target=\"_blank\">https://www.kaggle.com/maunish/clrp-roberta-svm</a></li>\n<li><a href=\"https://www.kaggle.com/jcesquiveld/best-transformer-representations\" target=\"_blank\">https://www.kaggle.com/jcesquiveld/best-transformer-representations</a></li>\n<li><a href=\"https://www.kaggle.com/gilfernandes/commonlit-pytorch-t5-large-svm\" target=\"_blank\">https://www.kaggle.com/gilfernandes/commonlit-pytorch-t5-large-svm</a><br>\nThank you <a href=\"https://www.kaggle.com/rhtsingh\" target=\"_blank\">@rhtsingh</a> <a href=\"https://www.kaggle.com/andretugan\" target=\"_blank\">@andretugan</a>  <a href=\"https://www.kaggle.com/maunish\" target=\"_blank\">@maunish</a> <a href=\"https://www.kaggle.com/jcesquiveld\" target=\"_blank\">@jcesquiveld</a> <a href=\"https://www.kaggle.com/gilfernandes\" target=\"_blank\">@gilfernandes</a> and the comunity for sharing great notebooks!</li>\n</ul>",
      "rawMarkdown": "Thank you Kaggle and CommonLit for exciting competition!\n\nour team (cpptake nishipy)  was in the silver medal range in the public LB, but in the private LB, the shake down happened and we fall into the bronze medal range.\nIt's frustrating, but we learned a lot, so we'll leave you with the solution\n\nI would like to get a better ranking in the next competition, so if you have any ideas for improvement, please comment.\n\n### Pipe Line & ideas\n##### see the following notebook:\n[our final sumit notebook](https://www.kaggle.com/takeshikobayashi/bronze-medal-solution-roberta-stacking-ensemble)\n\n\n\n###reference¶\n- https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-infer-3\n- https://www.kaggle.com/rhtsingh/commonlit-readability-prize-roberta-torch-itpt\n- https://www.kaggle.com/andretugan/pre-trained-roberta-solution-in-pytorch\n- https://www.kaggle.com/maunish/clrp-roberta-svm\n- https://www.kaggle.com/jcesquiveld/best-transformer-representations\n- https://www.kaggle.com/gilfernandes/commonlit-pytorch-t5-large-svm\nThank you @rhtsingh @andretugan  @maunish @jcesquiveld @gilfernandes and the comunity for sharing great notebooks!\n\n",
      "votes": 8
    },
    {
      "id": 1449388,
      "postDate": "2021-08-04T21:26:19.207Z",
      "content": "<p>Congrats on winning a medal in your first NLP competition!</p>",
      "rawMarkdown": "Congrats on winning a medal in your first NLP competition!",
      "votes": 3,
      "replies": [
        {
          "id": 1449608,
          "postDate": "2021-08-04T23:21:10.487Z",
          "content": "<p>Thank you!</p>",
          "rawMarkdown": "Thank you!",
          "votes": 1
        }
      ]
    },
    {
      "id": 1447617,
      "postDate": "2021-08-04T15:27:57.207Z",
      "content": "<p>Great job earning a medal in your first NLP competition!</p>",
      "rawMarkdown": "Great job earning a medal in your first NLP competition!",
      "votes": 3,
      "replies": [
        {
          "id": 1447994,
          "postDate": "2021-08-04T16:28:53.830Z",
          "content": "<p>Thank you for reply!<br>\nWe will continue trying hard to get better place in competitions and to contribute for the Kaggle community</p>",
          "rawMarkdown": "Thank you for reply!\nWe will continue trying hard to get better place in competitions and to contribute for the Kaggle community",
          "votes": 1
        }
      ]
    }
  ],
  "complete": true,
  "source": "live",
  "fetched_at": "2026-10-19T23:49:28Z",
  "competition_id": 25914,
  "competition_slug": "commonlitreadabilityprize",
  "team_id": 6852025,
  "solution_url": "/competitions/commonlitreadabilityprize/writeups/cpptake-nishipy-solution-to-our-first-nlp-competit"
}