{
  "id": 600369,
  "title": "Grand X-Ray Slam Division B Official Discussion Thread",
  "url": "/competitions/grand-xray-slam-division-b/discussion/600369",
  "author_name": "Guntas Dhanjal",
  "post_date": "2025-08-22T14:30:42.676000",
  "votes": 0,
  "comment_count": 7,
  "views": 0,
  "content": "<p>Hi all and welcome to Division B of the Grand X-Ray Slam — the second challenge in this 2-part Kaggle hackathon by Blue and Gold Healthcare Inc.</p>\n<p>This round focuses again on 14 chest X-ray conditions, but with a new dataset split to really test how well your models generalize. It’s your chance to refine ideas from Division A or try out fresh strategies — and of course, aim for the Grand Slam Leaderboard.</p>\n<h3>Prizes</h3>\n<ul>\n<li>1st Place: $750  </li>\n<li>2nd Place: $500  </li>\n<li>3rd Place: $250  </li>\n<li><strong>Grand Slam Bonus:</strong> $2,500 shared among the top 3 performers across both Divisions  </li>\n</ul>\n<h3>How to Begin</h3>\n<ul>\n<li>Dive into the <a href=\"https://www.kaggle.com/competitions/grand-xray-slam-division-b\" target=\"_blank\">Division B</a> dataset, notice the differences vs. <a href=\"https://www.kaggle.com/competitions/grand-xray-slam-division-a\" target=\"_blank\">Division A</a>.</li>\n<li>Start with a baseline, then move toward transfer learning or ensembles.</li>\n<li>Use this discussion space to ask questions, exchange ideas, or share notebooks. </li>\n</ul>\n<p>We’re looking forward to seeing how you take on this new stage.<br>\nLet’s keep pushing the limits of healthcare AI together. </p>\n<p>— The Grand X-Ray Slam Team  </p>",
  "messages": [
    {
      "id": 3301025,
      "postDate": "2025-10-12T09:31:05.697Z",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/guntasdhanjal\" target=\"_blank\">@guntasdhanjal</a> can I use dataset from Div A to train model for Div B or I only allow to use public dataset?</p>",
      "rawMarkdown": "Hi @guntasdhanjal can I use dataset from Div A to train model for Div B or I only allow to use public dataset?",
      "replies": [
        {
          "id": 3301550,
          "postDate": "2025-10-13T16:00:22.120Z",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/sonnguyenk17\" target=\"_blank\">@sonnguyenk17</a> ,</p>\n<p>The datasets in Division A and Division B are fully separate — each contains different images.<br>\nFor Division B, you should train your model only on Division B’s data.<br>\nUsing data from Division A to train a model for Division B isn’t allowed, since it would mix data across divisions.</p>\n<p>You’re welcome to reuse your code or model architecture, but the training data must come from the current division.</p>",
          "rawMarkdown": "Hi @sonnguyenk17 ,\n\nThe datasets in Division A and Division B are fully separate — each contains different images.\nFor Division B, you should train your model only on Division B’s data.\nUsing data from Division A to train a model for Division B isn’t allowed, since it would mix data across divisions.\n\nYou’re welcome to reuse your code or model architecture, but the training data must come from the current division."
        }
      ]
    },
    {
      "id": 3295493,
      "postDate": "2025-09-29T00:02:16.730Z",
      "content": "<p>Hi - just to clarify: Is it ok for the same person to compete in both Division A as well as Division B, or are we supposed to pick one?</p>",
      "rawMarkdown": "Hi - just to clarify: Is it ok for the same person to compete in both Division A as well as Division B, or are we supposed to pick one?",
      "replies": [
        {
          "id": 3295612,
          "postDate": "2025-09-29T08:29:31.930Z",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/optimistix\" target=\"_blank\">@optimistix</a> ,<br>\nYes, you can compete in both Division A and Division B.  <br>\nPrizes are awarded separately for each division, and the shared \"Grand Slam\" prizes require participation in both divisions under the same account.</p>",
          "rawMarkdown": "Hi @optimistix ,\nYes, you can compete in both Division A and Division B.  \nPrizes are awarded separately for each division, and the shared \"Grand Slam\" prizes require participation in both divisions under the same account.\n",
          "votes": 1,
          "replies": [
            {
              "id": 3295961,
              "postDate": "2025-09-29T21:32:36.647Z",
              "content": "<p>Thanks for confirming/clarifying.</p>",
              "rawMarkdown": "Thanks for confirming/clarifying."
            }
          ]
        }
      ]
    },
    {
      "id": 3276946,
      "postDate": "2025-08-27T03:51:36.573Z",
      "content": "<p>While training, my code showed that 00052495_001_001.jpg is 0 bytes. It is in line <strong>49072</strong> in the train2.csv. Please confirm if this is my error or the file is truely missing.</p>\n<p><a href=\"https://www.kaggle.com/guntasdhanjal\" target=\"_blank\">@guntasdhanjal</a> </p>",
      "rawMarkdown": "While training, my code showed that 00052495_001_001.jpg is 0 bytes. It is in line **49072** in the train2.csv. Please confirm if this is my error or the file is truely missing.\n\n@guntasdhanjal ",
      "replies": [
        {
          "id": 3278716,
          "postDate": "2025-08-30T12:31:45.393Z",
          "content": "<p>i'm goana check them for you and get abck to u asap </p>",
          "rawMarkdown": "i'm goana check them for you and get abck to u asap \n"
        }
      ]
    },
    {
      "id": 3273309,
      "postDate": "2025-08-22T14:30:42.677Z",
      "content": "<p>Hi all and welcome to Division B of the Grand X-Ray Slam — the second challenge in this 2-part Kaggle hackathon by Blue and Gold Healthcare Inc.</p>\n<p>This round focuses again on 14 chest X-ray conditions, but with a new dataset split to really test how well your models generalize. It’s your chance to refine ideas from Division A or try out fresh strategies — and of course, aim for the Grand Slam Leaderboard.</p>\n<h3>Prizes</h3>\n<ul>\n<li>1st Place: $750  </li>\n<li>2nd Place: $500  </li>\n<li>3rd Place: $250  </li>\n<li><strong>Grand Slam Bonus:</strong> $2,500 shared among the top 3 performers across both Divisions  </li>\n</ul>\n<h3>How to Begin</h3>\n<ul>\n<li>Dive into the <a href=\"https://www.kaggle.com/competitions/grand-xray-slam-division-b\" target=\"_blank\">Division B</a> dataset, notice the differences vs. <a href=\"https://www.kaggle.com/competitions/grand-xray-slam-division-a\" target=\"_blank\">Division A</a>.</li>\n<li>Start with a baseline, then move toward transfer learning or ensembles.</li>\n<li>Use this discussion space to ask questions, exchange ideas, or share notebooks. </li>\n</ul>\n<p>We’re looking forward to seeing how you take on this new stage.<br>\nLet’s keep pushing the limits of healthcare AI together. </p>\n<p>— The Grand X-Ray Slam Team  </p>",
      "rawMarkdown": "Hi all and welcome to Division B of the Grand X-Ray Slam — the second challenge in this 2-part Kaggle hackathon by Blue and Gold Healthcare Inc.\n\nThis round focuses again on 14 chest X-ray conditions, but with a new dataset split to really test how well your models generalize. It’s your chance to refine ideas from Division A or try out fresh strategies — and of course, aim for the Grand Slam Leaderboard.\n\n###  Prizes\n- 1st Place: $750  \n- 2nd Place: $500  \n- 3rd Place: $250  \n- **Grand Slam Bonus:** $2,500 shared among the top 3 performers across both Divisions  \n\n###  How to Begin\n- Dive into the [Division B](https://www.kaggle.com/competitions/grand-xray-slam-division-b) dataset, notice the differences vs. [Division A](https://www.kaggle.com/competitions/grand-xray-slam-division-a).\n- Start with a baseline, then move toward transfer learning or ensembles.\n- Use this discussion space to ask questions, exchange ideas, or share notebooks. \n\nWe’re looking forward to seeing how you take on this new stage.\nLet’s keep pushing the limits of healthcare AI together. \n\n— The Grand X-Ray Slam Team  "
    }
  ],
  "comments": [
    {
      "id": 3301025,
      "author_name": "SonNguyenK17",
      "author_url": "",
      "post_date": "2025-10-12T09:31:05.697000",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/guntasdhanjal\" target=\"_blank\">@guntasdhanjal</a> can I use dataset from Div A to train model for Div B or I only allow to use public dataset?</p>",
      "votes": 0,
      "replies": [
        {
          "id": 3301550,
          "author_name": "Guntas Dhanjal",
          "author_url": "",
          "post_date": "2025-10-13T16:00:22.120000",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/sonnguyenk17\" target=\"_blank\">@sonnguyenk17</a> ,</p>\n<p>The datasets in Division A and Division B are fully separate — each contains different images.<br>\nFor Division B, you should train your model only on Division B’s data.<br>\nUsing data from Division A to train a model for Division B isn’t allowed, since it would mix data across divisions.</p>\n<p>You’re welcome to reuse your code or model architecture, but the training data must come from the current division.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 3295493,
      "author_name": "Optimistix",
      "author_url": "",
      "post_date": "2025-09-29T00:02:16.730000",
      "content": "<p>Hi - just to clarify: Is it ok for the same person to compete in both Division A as well as Division B, or are we supposed to pick one?</p>",
      "votes": 0,
      "replies": [
        {
          "id": 3295612,
          "author_name": "Guntas Dhanjal",
          "author_url": "",
          "post_date": "2025-09-29T08:29:31.930000",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/optimistix\" target=\"_blank\">@optimistix</a> ,<br>\nYes, you can compete in both Division A and Division B.  <br>\nPrizes are awarded separately for each division, and the shared \"Grand Slam\" prizes require participation in both divisions under the same account.</p>",
          "votes": 1,
          "replies": [
            {
              "id": 3295961,
              "author_name": "Optimistix",
              "author_url": "",
              "post_date": "2025-09-29T21:32:36.647000",
              "content": "<p>Thanks for confirming/clarifying.</p>",
              "votes": 0,
              "replies": []
            }
          ]
        }
      ]
    },
    {
      "id": 3276946,
      "author_name": "Hiếu Lê Ngọc",
      "author_url": "",
      "post_date": "2025-08-27T03:51:36.573000",
      "content": "<p>While training, my code showed that 00052495_001_001.jpg is 0 bytes. It is in line <strong>49072</strong> in the train2.csv. Please confirm if this is my error or the file is truely missing.</p>\n<p><a href=\"https://www.kaggle.com/guntasdhanjal\" target=\"_blank\">@guntasdhanjal</a> </p>",
      "votes": 0,
      "replies": [
        {
          "id": 3278716,
          "author_name": "Guntas Dhanjal",
          "author_url": "",
          "post_date": "2025-08-30T12:31:45.393000",
          "content": "<p>i'm goana check them for you and get abck to u asap </p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3301025": "Hi @guntasdhanjal can I use dataset from Div A to train model for Div B or I only allow to use public dataset?",
    "3295493": "Hi - just to clarify: Is it ok for the same person to compete in both Division A as well as Division B, or are we supposed to pick one?",
    "3276946": "While training, my code showed that 00052495_001_001.jpg is 0 bytes. It is in line **49072** in the train2.csv. Please confirm if this is my error or the file is truely missing.\n\n@guntasdhanjal ",
    "3273309": "Hi all and welcome to Division B of the Grand X-Ray Slam — the second challenge in this 2-part Kaggle hackathon by Blue and Gold Healthcare Inc.\n\nThis round focuses again on 14 chest X-ray conditions, but with a new dataset split to really test how well your models generalize. It’s your chance to refine ideas from Division A or try out fresh strategies — and of course, aim for the Grand Slam Leaderboard.\n\n###  Prizes\n- 1st Place: $750  \n- 2nd Place: $500  \n- 3rd Place: $250  \n- **Grand Slam Bonus:** $2,500 shared among the top 3 performers across both Divisions  \n\n###  How to Begin\n- Dive into the [Division B](https://www.kaggle.com/competitions/grand-xray-slam-division-b) dataset, notice the differences vs. [Division A](https://www.kaggle.com/competitions/grand-xray-slam-division-a).\n- Start with a baseline, then move toward transfer learning or ensembles.\n- Use this discussion space to ask questions, exchange ideas, or share notebooks. \n\nWe’re looking forward to seeing how you take on this new stage.\nLet’s keep pushing the limits of healthcare AI together. \n\n— The Grand X-Ray Slam Team  "
  }
}