{
  "id": 238325,
  "title": "PyTorch NFNet_l0 Baseline [LB: 0.95]",
  "url": "/competitions/seti-breakthrough-listen/discussion/238325",
  "author_name": "Y.Nakama",
  "post_date": "2021-05-11T23:06:28.416000",
  "votes": 73,
  "comment_count": 6,
  "views": 0,
  "content": "<p>I prepared PyTorch NFNet_l0 Baseline.</p>\n<ul>\n<li>Model: NFNet_l0</li>\n<li>Split: StratifiedKFold 4 folds</li>\n<li>Size: 224x224</li>\n<li>CV: 0.968, LB: 0.95</li>\n<li><a href=\"https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-training\" target=\"_blank\">training notebook</a></li>\n<li><a href=\"https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-inference\" target=\"_blank\">inference notebook</a></li>\n</ul>\n<p>Hope this helps, happy kaggling!</p>",
  "messages": [
    {
      "id": 1303110,
      "postDate": "2021-05-11T23:06:28.417Z",
      "content": "<p>I prepared PyTorch NFNet_l0 Baseline.</p>\n<ul>\n<li>Model: NFNet_l0</li>\n<li>Split: StratifiedKFold 4 folds</li>\n<li>Size: 224x224</li>\n<li>CV: 0.968, LB: 0.95</li>\n<li><a href=\"https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-training\" target=\"_blank\">training notebook</a></li>\n<li><a href=\"https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-inference\" target=\"_blank\">inference notebook</a></li>\n</ul>\n<p>Hope this helps, happy kaggling!</p>",
      "rawMarkdown": "I prepared PyTorch NFNet_l0 Baseline.\n\n- Model: NFNet_l0\n- Split: StratifiedKFold 4 folds\n- Size: 224x224\n- CV: 0.968, LB: 0.95\n- [training notebook](https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-training)\n- [inference notebook](https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-inference)\n\nHope this helps, happy kaggling!",
      "votes": 73
    },
    {
      "id": 1303990,
      "postDate": "2021-05-12T11:19:24.533Z",
      "content": "<p>You are always creating useful public notebooks for different competitions, thanks for all you've done! I think the idea sharing is what makes kaggle so different!</p>",
      "rawMarkdown": "You are always creating useful public notebooks for different competitions, thanks for all you've done! I think the idea sharing is what makes kaggle so different!",
      "votes": 2,
      "replies": [
        {
          "id": 1304802,
          "postDate": "2021-05-12T22:19:00.603Z",
          "content": "<p>Thanks! I learned a lot from public notebooks, that's why I publish starter codes for other competitions too :)</p>",
          "rawMarkdown": "Thanks! I learned a lot from public notebooks, that's why I publish starter codes for other competitions too :)",
          "votes": 3
        }
      ]
    },
    {
      "id": 1303880,
      "postDate": "2021-05-12T10:02:29.713Z",
      "content": "<p>Fantastic kernel!</p>",
      "rawMarkdown": "Fantastic kernel!",
      "votes": 2
    },
    {
      "id": 1335417,
      "postDate": "2021-06-04T07:46:55.273Z",
      "content": "<p>Thank you! Your code was very easy to read!</p>",
      "rawMarkdown": "Thank you! Your code was very easy to read!"
    },
    {
      "id": 1316729,
      "postDate": "2021-05-20T20:09:23.687Z",
      "content": "<p>Thank you!!!</p>",
      "rawMarkdown": "Thank you!!!\n"
    },
    {
      "id": 1316382,
      "postDate": "2021-05-20T13:54:38.870Z",
      "content": "<p>Thank you for your wonderful baseline!</p>",
      "rawMarkdown": "Thank you for your wonderful baseline!"
    }
  ],
  "comments": [
    {
      "id": 1303990,
      "author_name": "YTEP (Jiazhi Yang)",
      "author_url": "",
      "post_date": "2021-05-12T11:19:24.533000",
      "content": "<p>You are always creating useful public notebooks for different competitions, thanks for all you've done! I think the idea sharing is what makes kaggle so different!</p>",
      "votes": 2,
      "replies": [
        {
          "id": 1304802,
          "author_name": "Y.Nakama",
          "author_url": "",
          "post_date": "2021-05-12T22:19:00.603000",
          "content": "<p>Thanks! I learned a lot from public notebooks, that's why I publish starter codes for other competitions too :)</p>",
          "votes": 3,
          "replies": []
        }
      ]
    },
    {
      "id": 1303880,
      "author_name": "James Howard",
      "author_url": "",
      "post_date": "2021-05-12T10:02:29.713000",
      "content": "<p>Fantastic kernel!</p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 1335417,
      "author_name": "Y.Ts",
      "author_url": "",
      "post_date": "2021-06-04T07:46:55.273000",
      "content": "<p>Thank you! Your code was very easy to read!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1316729,
      "author_name": "Gustavo Corradi",
      "author_url": "",
      "post_date": "2021-05-20T20:09:23.687000",
      "content": "<p>Thank you!!!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1316382,
      "author_name": "HyoJin Kim",
      "author_url": "",
      "post_date": "2021-05-20T13:54:38.870000",
      "content": "<p>Thank you for your wonderful baseline!</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1303110": "I prepared PyTorch NFNet_l0 Baseline.\n\n- Model: NFNet_l0\n- Split: StratifiedKFold 4 folds\n- Size: 224x224\n- CV: 0.968, LB: 0.95\n- [training notebook](https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-training)\n- [inference notebook](https://www.kaggle.com/yasufuminakama/seti-nfnet-l0-starter-inference)\n\nHope this helps, happy kaggling!",
    "1303990": "You are always creating useful public notebooks for different competitions, thanks for all you've done! I think the idea sharing is what makes kaggle so different!",
    "1303880": "Fantastic kernel!",
    "1335417": "Thank you! Your code was very easy to read!",
    "1316729": "Thank you!!!\n",
    "1316382": "Thank you for your wonderful baseline!"
  }
}