{
  "id": 228840,
  "title": "TfRecord Dataset 1024 x 1024",
  "url": "/competitions/hubmap-kidney-segmentation/discussion/228840",
  "author_name": "Md. Masud Rana",
  "post_date": "2021-03-26T21:13:57.391000",
  "votes": 6,
  "comment_count": 4,
  "views": 0,
  "content": "<p>Hi, all I made these two dataset size 1024*1024 TfRecord format base on <a href=\"https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs\" target=\"_blank\">this</a> kernel.</p>\n<ol>\n<li><a href=\"https://www.kaggle.com/durbin164/hubmap-data-1024-1024-tfrecord\" target=\"_blank\">Orignal dataset's TfRecord</a></li>\n<li><a href=\"https://www.kaggle.com/durbin164/hubmap-data-1024-1024-shift-1024\" target=\"_blank\">Shift-1024 TfRecord</a></li>\n</ol>\n<p>Thanks a lot, <a href=\"https://www.kaggle.com/wrrosa\" target=\"_blank\">@wrrosa</a> for your great kernel. </p>",
  "messages": [
    {
      "id": 1253549,
      "postDate": "2021-03-26T21:13:57.390Z",
      "content": "<p>Hi, all I made these two dataset size 1024*1024 TfRecord format base on <a href=\"https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs\" target=\"_blank\">this</a> kernel.</p>\n<ol>\n<li><a href=\"https://www.kaggle.com/durbin164/hubmap-data-1024-1024-tfrecord\" target=\"_blank\">Orignal dataset's TfRecord</a></li>\n<li><a href=\"https://www.kaggle.com/durbin164/hubmap-data-1024-1024-shift-1024\" target=\"_blank\">Shift-1024 TfRecord</a></li>\n</ol>\n<p>Thanks a lot, <a href=\"https://www.kaggle.com/wrrosa\" target=\"_blank\">@wrrosa</a> for your great kernel. </p>",
      "rawMarkdown": "Hi, all I made these two dataset size 1024*1024 TfRecord format base on [this](https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs) kernel.\n\n1. [Orignal dataset's TfRecord](https://www.kaggle.com/durbin164/hubmap-data-1024-1024-tfrecord)\n2. [Shift-1024 TfRecord](https://www.kaggle.com/durbin164/hubmap-data-1024-1024-shift-1024)\n\nThanks a lot, @wrrosa for your great kernel. \n",
      "votes": 6
    },
    {
      "id": 1253698,
      "postDate": "2021-03-27T00:41:52.783Z",
      "content": "<p>Have you tried using it in kaggle kernel?</p>",
      "rawMarkdown": "Have you tried using it in kaggle kernel?\n",
      "votes": 1,
      "replies": [
        {
          "id": 1253994,
          "postDate": "2021-03-27T08:39:55.820Z",
          "content": "<p>Hi, I tried it in Kaggle kernel with the TPU environment. </p>",
          "rawMarkdown": "Hi, I tried it in Kaggle kernel with the TPU environment. "
        }
      ]
    },
    {
      "id": 1253807,
      "postDate": "2021-03-27T03:53:56.420Z",
      "rawMarkdown": "",
      "votes": 1,
      "isDeleted": true,
      "replies": [
        {
          "id": 1253997,
          "postDate": "2021-03-27T08:43:38.187Z",
          "content": "<p>Hi, In <a href=\"https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs#Train-2\" target=\"_blank\">this </a>kernel in the Train2 section, the author shift the dataset size with pad size. and made two separated datasets with original Train data.  </p>",
          "rawMarkdown": "Hi, In [this ](https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs#Train-2)kernel in the Train2 section, the author shift the dataset size with pad size. and made two separated datasets with original Train data.  "
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 1253698,
      "author_name": "William Green",
      "author_url": "",
      "post_date": "2021-03-27T00:41:52.783000",
      "content": "<p>Have you tried using it in kaggle kernel?</p>",
      "votes": 1,
      "replies": [
        {
          "id": 1253994,
          "author_name": "Md. Masud Rana",
          "author_url": "",
          "post_date": "2021-03-27T08:39:55.820000",
          "content": "<p>Hi, I tried it in Kaggle kernel with the TPU environment. </p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1253807,
      "author_name": "",
      "author_url": "",
      "post_date": "2021-03-27T03:53:56.420000",
      "content": "",
      "votes": 1,
      "replies": [
        {
          "id": 1253997,
          "author_name": "Md. Masud Rana",
          "author_url": "",
          "post_date": "2021-03-27T08:43:38.187000",
          "content": "<p>Hi, In <a href=\"https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs#Train-2\" target=\"_blank\">this </a>kernel in the Train2 section, the author shift the dataset size with pad size. and made two separated datasets with original Train data.  </p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1253549": "Hi, all I made these two dataset size 1024*1024 TfRecord format base on [this](https://www.kaggle.com/wrrosa/hubmap-tf-with-tpu-efficientunet-512x512-tfrecs) kernel.\n\n1. [Orignal dataset's TfRecord](https://www.kaggle.com/durbin164/hubmap-data-1024-1024-tfrecord)\n2. [Shift-1024 TfRecord](https://www.kaggle.com/durbin164/hubmap-data-1024-1024-shift-1024)\n\nThanks a lot, @wrrosa for your great kernel. \n",
    "1253698": "Have you tried using it in kaggle kernel?\n",
    "1253807": ""
  }
}