{
  "id": 307800,
  "title": "63rd place solution - Single Cascade R-CNN + Norfair",
  "url": "/competitions/tensorflow-great-barrier-reef/writeups/shinya7y-63rd-place-solution-single-cascade-r-cnn-",
  "author_name": "",
  "post_date": "2022-02-20T07:13:57.263Z",
  "votes": 9,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I would like to thank organizers and all participants.</p>\n<p>Detector: Cascade R-CNN, Res2Net-101, 2560x1440<br>\nTracker: Norfair</p>\n<p>Training config:<br>\n<a href=\"https://github.com/shinya7y/UniverseNet/blob/master/configs/datasets/kaggle_gbr_cots/cascade_rcnn_r2_101_fpn_fp16_4x2_mixup_affine_hsv_1440_lr002_7e_cots_val2.py\" target=\"_blank\">https://github.com/shinya7y/UniverseNet/blob/master/configs/datasets/kaggle_gbr_cots/cascade_rcnn_r2_101_fpn_fp16_4x2_mixup_affine_hsv_1440_lr002_7e_cots_val2.py</a></p>\n<p>Inference notebook:<br>\n<a href=\"https://www.kaggle.com/shinya7y/great-barrier-reef-cascade-r-cnn-private-0-694\" target=\"_blank\">https://www.kaggle.com/shinya7y/great-barrier-reef-cascade-r-cnn-private-0-694</a><br>\nThis notebook version: 0.694 (trained on video 0, 1)<br>\nSubmission version: 0.685 (trained on video 0, 1, 2; I should have tuned hyperparameters 😢)</p>",
  "messages": [
    {
      "id": "1691913",
      "postDate": "02/15/2022 17:21:01",
      "content": "<p>I would like to thank organizers and all participants.</p>\n<p>Detector: Cascade R-CNN, Res2Net-101, 2560x1440<br>\nTracker: Norfair</p>\n<p>Training config:<br>\n<a href=\"https://github.com/shinya7y/UniverseNet/blob/master/configs/datasets/kaggle_gbr_cots/cascade_rcnn_r2_101_fpn_fp16_4x2_mixup_affine_hsv_1440_lr002_7e_cots_val2.py\" target=\"_blank\">https://github.com/shinya7y/UniverseNet/blob/master/configs/datasets/kaggle_gbr_cots/cascade_rcnn_r2_101_fpn_fp16_4x2_mixup_affine_hsv_1440_lr002_7e_cots_val2.py</a></p>\n<p>Inference notebook:<br>\n<a href=\"https://www.kaggle.com/shinya7y/great-barrier-reef-cascade-r-cnn-private-0-694\" target=\"_blank\">https://www.kaggle.com/shinya7y/great-barrier-reef-cascade-r-cnn-private-0-694</a><br>\nThis notebook version: 0.694 (trained on video 0, 1)<br>\nSubmission version: 0.685 (trained on video 0, 1, 2; I should have tuned hyperparameters 😢)</p>",
      "rawMarkdown": "I would like to thank organizers and all participants.\n\nDetector: Cascade R-CNN, Res2Net-101, 2560x1440\nTracker: Norfair\n\nTraining config:\nhttps://github.com/shinya7y/UniverseNet/blob/master/configs/datasets/kaggle_gbr_cots/cascade_rcnn_r2_101_fpn_fp16_4x2_mixup_affine_hsv_1440_lr002_7e_cots_val2.py\n\nInference notebook:\nhttps://www.kaggle.com/shinya7y/great-barrier-reef-cascade-r-cnn-private-0-694\nThis notebook version: 0.694 (trained on video 0, 1)\nSubmission version: 0.685 (trained on video 0, 1, 2; I should have tuned hyperparameters 😢)",
      "votes": null
    },
    {
      "id": "1743649",
      "postDate": "04/03/2022 07:10:27",
      "content": "<p>Great Job!</p>",
      "rawMarkdown": "Great Job!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1743649,
      "author_name": "rishengyang",
      "author_url": "",
      "post_date": "04/03/2022 07:10:27",
      "content": "<p>Great Job!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1691913": "I would like to thank organizers and all participants.\n\nDetector: Cascade R-CNN, Res2Net-101, 2560x1440\nTracker: Norfair\n\nTraining config:\nhttps://github.com/shinya7y/UniverseNet/blob/master/configs/datasets/kaggle_gbr_cots/cascade_rcnn_r2_101_fpn_fp16_4x2_mixup_affine_hsv_1440_lr002_7e_cots_val2.py\n\nInference notebook:\nhttps://www.kaggle.com/shinya7y/great-barrier-reef-cascade-r-cnn-private-0-694\nThis notebook version: 0.694 (trained on video 0, 1)\nSubmission version: 0.685 (trained on video 0, 1, 2; I should have tuned hyperparameters 😢)",
    "1743649": "Great Job!"
  },
  "source": "meta"
}