{
  "id": 428316,
  "title": "What is your best private score? ",
  "url": "/competitions/hubmap-hacking-the-human-vasculature/discussion/428316",
  "author_name": "Duong Nguyen",
  "post_date": "2023-08-01T02:14:55.078000",
  "votes": 1,
  "comment_count": 13,
  "views": 0,
  "content": "<p>Let's show your best leaderboard-private score.</p>",
  "messages": [
    {
      "id": 2368095,
      "postDate": "2023-08-01T02:24:40.423Z",
      "content": "<p>My best private score: 0.558 with yolov8</p>",
      "rawMarkdown": "My best private score: 0.558 with yolov8",
      "votes": 1,
      "replies": [
        {
          "id": 2368123,
          "postDate": "2023-08-01T02:49:20.267Z",
          "content": "<p>I tried the same model (yolov8), but never got 0.5+ in private ,can u share something with me </p>",
          "rawMarkdown": "I tried the same model (yolov8), but never got 0.5+ in private ,can u share something with me "
        }
      ]
    },
    {
      "id": 2368090,
      "postDate": "2023-08-01T02:21:14.710Z",
      "content": "<p>0.549, yolov8, almost default parameters. I just used it to see if dilation works. The higher the public score, the lower the private score. </p>",
      "rawMarkdown": "0.549, yolov8, almost default parameters. I just used it to see if dilation works. The higher the public score, the lower the private score. ",
      "votes": 1,
      "replies": [
        {
          "id": 2368097,
          "postDate": "2023-08-01T02:25:14.007Z",
          "content": "<p>Every one using dilate will be overfitting bro</p>",
          "rawMarkdown": "Every one using dilate will be overfitting bro\n"
        }
      ]
    },
    {
      "id": 2368084,
      "postDate": "2023-08-01T02:16:52.997Z",
      "content": "<p>PrivateScore: 0.597 -_-</p>",
      "rawMarkdown": "PrivateScore: 0.597 -_-",
      "votes": 1,
      "replies": [
        {
          "id": 2368106,
          "postDate": "2023-08-01T02:35:06.797Z",
          "content": "<p>What a sad story. Good luck in your next competition.</p>",
          "rawMarkdown": "What a sad story. Good luck in your next competition.",
          "votes": 1
        },
        {
          "id": 2368132,
          "postDate": "2023-08-01T02:57:12.637Z",
          "content": "<p>good luck bro</p>",
          "rawMarkdown": "good luck bro\n"
        }
      ]
    },
    {
      "id": 2368081,
      "postDate": "2023-08-01T02:14:55.077Z",
      "content": "<p>Let's show your best leaderboard-private score.</p>",
      "rawMarkdown": "Let's show your best leaderboard-private score.",
      "votes": 1
    },
    {
      "id": 2368148,
      "postDate": "2023-08-01T03:09:41.510Z",
      "content": "<p>My best score is 552 (difference in epochs).</p>",
      "rawMarkdown": "My best score is 552 (difference in epochs)."
    },
    {
      "id": 2368124,
      "postDate": "2023-08-01T02:49:54.970Z",
      "content": "<p>The private test data seems very unstable itself, my very simple model single fold can get 0.536 private. When ensemble of multiple folds got lower</p>",
      "rawMarkdown": "The private test data seems very unstable itself, my very simple model single fold can get 0.536 private. When ensemble of multiple folds got lower",
      "replies": [
        {
          "id": 2368135,
          "postDate": "2023-08-01T02:58:46.020Z",
          "content": "<p>Maybe we overfitting on dataset 2. </p>",
          "rawMarkdown": "Maybe we overfitting on dataset 2. "
        }
      ]
    },
    {
      "id": 2368092,
      "postDate": "2023-08-01T02:22:49.043Z",
      "content": "<p>best private: 0.561😭</p>",
      "rawMarkdown": "best private: 0.561😭",
      "replies": [
        {
          "id": 2368121,
          "postDate": "2023-08-01T02:49:02.543Z",
          "rawMarkdown": "",
          "isDeleted": true
        }
      ]
    },
    {
      "id": 2368083,
      "postDate": "2023-08-01T02:16:14.580Z",
      "content": "<p>PrivateScore: 0.537 (Yolov8 size 1024 with pseudolabel) (:(( we didn't selected it)</p>",
      "rawMarkdown": "PrivateScore: 0.537 (Yolov8 size 1024 with pseudolabel) (:(( we didn't selected it)"
    }
  ],
  "comments": [
    {
      "id": 2368095,
      "author_name": "Mr.CheckerChichChich",
      "author_url": "",
      "post_date": "2023-08-01T02:24:40.423000",
      "content": "<p>My best private score: 0.558 with yolov8</p>",
      "votes": 1,
      "replies": [
        {
          "id": 2368123,
          "author_name": "Arvin",
          "author_url": "",
          "post_date": "2023-08-01T02:49:20.267000",
          "content": "<p>I tried the same model (yolov8), but never got 0.5+ in private ,can u share something with me </p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2368090,
      "author_name": "WQ",
      "author_url": "",
      "post_date": "2023-08-01T02:21:14.710000",
      "content": "<p>0.549, yolov8, almost default parameters. I just used it to see if dilation works. The higher the public score, the lower the private score. </p>",
      "votes": 1,
      "replies": [
        {
          "id": 2368097,
          "author_name": "Mr.CheckerChichChich",
          "author_url": "",
          "post_date": "2023-08-01T02:25:14.007000",
          "content": "<p>Every one using dilate will be overfitting bro</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2368084,
      "author_name": "Quan Vu",
      "author_url": "",
      "post_date": "2023-08-01T02:16:52.997000",
      "content": "<p>PrivateScore: 0.597 -_-</p>",
      "votes": 1,
      "replies": [
        {
          "id": 2368106,
          "author_name": "WQ",
          "author_url": "",
          "post_date": "2023-08-01T02:35:06.797000",
          "content": "<p>What a sad story. Good luck in your next competition.</p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 2368132,
          "author_name": "Duong Nguyen",
          "author_url": "",
          "post_date": "2023-08-01T02:57:12.637000",
          "content": "<p>good luck bro</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2368148,
      "author_name": "devchopin",
      "author_url": "",
      "post_date": "2023-08-01T03:09:41.510000",
      "content": "<p>My best score is 552 (difference in epochs).</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2368124,
      "author_name": "Phat Tran",
      "author_url": "",
      "post_date": "2023-08-01T02:49:54.970000",
      "content": "<p>The private test data seems very unstable itself, my very simple model single fold can get 0.536 private. When ensemble of multiple folds got lower</p>",
      "votes": 0,
      "replies": [
        {
          "id": 2368135,
          "author_name": "Duong Nguyen",
          "author_url": "",
          "post_date": "2023-08-01T02:58:46.020000",
          "content": "<p>Maybe we overfitting on dataset 2. </p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2368092,
      "author_name": "TanjiroLL",
      "author_url": "",
      "post_date": "2023-08-01T02:22:49.043000",
      "content": "<p>best private: 0.561😭</p>",
      "votes": 0,
      "replies": [
        {
          "id": 2368121,
          "author_name": "",
          "author_url": "",
          "post_date": "2023-08-01T02:49:02.543000",
          "content": "",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2368083,
      "author_name": "Duong Nguyen",
      "author_url": "",
      "post_date": "2023-08-01T02:16:14.580000",
      "content": "<p>PrivateScore: 0.537 (Yolov8 size 1024 with pseudolabel) (:(( we didn't selected it)</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2368095": "My best private score: 0.558 with yolov8",
    "2368090": "0.549, yolov8, almost default parameters. I just used it to see if dilation works. The higher the public score, the lower the private score. ",
    "2368084": "PrivateScore: 0.597 -_-",
    "2368081": "Let's show your best leaderboard-private score.",
    "2368148": "My best score is 552 (difference in epochs).",
    "2368124": "The private test data seems very unstable itself, my very simple model single fold can get 0.536 private. When ensemble of multiple folds got lower",
    "2368092": "best private: 0.561😭",
    "2368083": "PrivateScore: 0.537 (Yolov8 size 1024 with pseudolabel) (:(( we didn't selected it)"
  }
}