{
  "id": 205885,
  "title": "How far can End2End CNN model can achieve🏆",
  "url": "/competitions/ranzcr-clip-catheter-line-classification/discussion/205885",
  "author_name": "",
  "post_date": "2020-12-22T10:38:26.976359500Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p><strong>A QUICK SURVEY⚡</strong><br>\n<strong>Generalizing Intuition about End2End-CNN classification model Performance.</strong></p>\n<p>The first-shot idea strikes to fresher-participant about the competition is to build a CNN-biased classification model.<br>\nBut we know to achieve more accurate classification, considering annotation will be <strong>worthwhile</strong>.</p>\n<p>For the sake of this topic, Just <strong>comment-down about your End2End-CNN classification model performance to help the community to generalize their ideas</strong>.</p>\n<p>It will be a great help∆<br>\nThank You🙌</p>",
  "messages": [
    {
      "id": "1122295",
      "postDate": "12/22/2020 10:38:26",
      "content": "<p><strong>A QUICK SURVEY⚡</strong><br>\n<strong>Generalizing Intuition about End2End-CNN classification model Performance.</strong></p>\n<p>The first-shot idea strikes to fresher-participant about the competition is to build a CNN-biased classification model.<br>\nBut we know to achieve more accurate classification, considering annotation will be <strong>worthwhile</strong>.</p>\n<p>For the sake of this topic, Just <strong>comment-down about your End2End-CNN classification model performance to help the community to generalize their ideas</strong>.</p>\n<p>It will be a great help∆<br>\nThank You🙌</p>",
      "rawMarkdown": "**A QUICK SURVEY⚡**\n**Generalizing Intuition about End2End-CNN classification model Performance.**\n\nThe first-shot idea strikes to fresher-participant about the competition is to build a CNN-biased classification model.\nBut we know to achieve more accurate classification, considering annotation will be **worthwhile**.\n\nFor the sake of this topic, Just **comment-down about your End2End-CNN classification model performance to help the community to generalize their ideas**.\n\nIt will be a great help∆\nThank You🙌",
      "votes": null
    },
    {
      "id": "1122302",
      "postDate": "12/22/2020 10:42:28",
      "content": "<p>I achieved <strong>SCORE-0.95</strong> using End2End Efficientnet-B7 CNN model.<br>\n<a href=\"https://www.kaggle.com/akhileshdkapse/efficientnetb7-tpu-validation-auc-0-93\" target=\"_blank\">Notebook</a><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4684168%2Ff01f2864c48d399a2cb043c24fa66e69%2F20201222_143321.jpg?generation=1608633734417572&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "I achieved **SCORE-0.95** using End2End Efficientnet-B7 CNN model.\n[Notebook](https://www.kaggle.com/akhileshdkapse/efficientnetb7-tpu-validation-auc-0-93)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4684168%2Ff01f2864c48d399a2cb043c24fa66e69%2F20201222_143321.jpg?generation=1608633734417572&alt=media)",
      "votes": null
    },
    {
      "id": "1123124",
      "postDate": "12/22/2020 23:55:51",
      "content": "<p>Currently, I have</p>\n<ul>\n<li>B2-256-noTTA-fold0: 0.934LB</li>\n<li>B4-380-noTTA-fold0: 0.953LB</li>\n</ul>",
      "rawMarkdown": "Currently, I have\n- B2-256-noTTA-fold0: 0.934LB\n- B4-380-noTTA-fold0: 0.953LB",
      "votes": null
    },
    {
      "id": "1123204",
      "postDate": "12/23/2020 03:13:59",
      "content": "<p>Great! I'm now trying to implement training with k-fold.<br>\nReally thanks for the notebook regarding this 🙌.</p>",
      "rawMarkdown": "Great! I'm now trying to implement training with k-fold.\nReally thanks for the notebook regarding this 🙌.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1122302,
      "author_name": "akhileshdkapse",
      "author_url": "",
      "post_date": "12/22/2020 10:42:28",
      "content": "<p>I achieved <strong>SCORE-0.95</strong> using End2End Efficientnet-B7 CNN model.<br>\n<a href=\"https://www.kaggle.com/akhileshdkapse/efficientnetb7-tpu-validation-auc-0-93\" target=\"_blank\">Notebook</a><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4684168%2Ff01f2864c48d399a2cb043c24fa66e69%2F20201222_143321.jpg?generation=1608633734417572&amp;alt=media\" alt=\"\"></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1123124,
      "author_name": "underwearfitting",
      "author_url": "",
      "post_date": "12/22/2020 23:55:51",
      "content": "<p>Currently, I have</p>\n<ul>\n<li>B2-256-noTTA-fold0: 0.934LB</li>\n<li>B4-380-noTTA-fold0: 0.953LB</li>\n</ul>",
      "votes": null,
      "replies": [
        {
          "id": 1123204,
          "author_name": "akhileshdkapse",
          "author_url": "",
          "post_date": "12/23/2020 03:13:59",
          "content": "<p>Great! I'm now trying to implement training with k-fold.<br>\nReally thanks for the notebook regarding this 🙌.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1122295": "**A QUICK SURVEY⚡**\n**Generalizing Intuition about End2End-CNN classification model Performance.**\n\nThe first-shot idea strikes to fresher-participant about the competition is to build a CNN-biased classification model.\nBut we know to achieve more accurate classification, considering annotation will be **worthwhile**.\n\nFor the sake of this topic, Just **comment-down about your End2End-CNN classification model performance to help the community to generalize their ideas**.\n\nIt will be a great help∆\nThank You🙌",
    "1122302": "I achieved **SCORE-0.95** using End2End Efficientnet-B7 CNN model.\n[Notebook](https://www.kaggle.com/akhileshdkapse/efficientnetb7-tpu-validation-auc-0-93)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4684168%2Ff01f2864c48d399a2cb043c24fa66e69%2F20201222_143321.jpg?generation=1608633734417572&alt=media)",
    "1123124": "Currently, I have\n- B2-256-noTTA-fold0: 0.934LB\n- B4-380-noTTA-fold0: 0.953LB",
    "1123204": "Great! I'm now trying to implement training with k-fold.\nReally thanks for the notebook regarding this 🙌."
  },
  "source": "meta"
}