{
  "id": 293719,
  "title": "Starting Tips Required",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/293719",
  "author_name": "",
  "post_date": "2021-12-06T16:10:20.983559500Z",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hello! This is my first time competing in a  competition. I would like some guideline regarding code and other stuff</p>\n<ul>\n<li>I would like to know if we have to label dataset manually</li>\n<li>I would also like to know do we have to use TF2 built in object detection API, if so then how to manage directory structure</li>\n</ul>",
  "messages": [
    {
      "id": "1608766",
      "postDate": "12/06/2021 16:10:20",
      "content": "<p>Hello! This is my first time competing in a  competition. I would like some guideline regarding code and other stuff</p>\n<ul>\n<li>I would like to know if we have to label dataset manually</li>\n<li>I would also like to know do we have to use TF2 built in object detection API, if so then how to manage directory structure</li>\n</ul>",
      "rawMarkdown": "Hello! This is my first time competing in a  competition. I would like some guideline regarding code and other stuff\n- I would like to know if we have to label dataset manually\n- I would also like to know do we have to use TF2 built in object detection API, if so then how to manage directory structure",
      "votes": null
    },
    {
      "id": "1611124",
      "postDate": "12/07/2021 18:35:29",
      "content": "<p>Hi,did you get help?</p>",
      "rawMarkdown": "Hi,did you get help?",
      "votes": null
    },
    {
      "id": "1625686",
      "postDate": "12/22/2021 04:27:23",
      "content": "<ol>\n<li>No don't have to label the dataset manually since the boundary boxes ares already labelled in the train.csv under annotations. Of course if you want to web crawl and download photos of COTS to add, then manual labeling would be used there. </li>\n<li>You don't have to use TF2 but it is required for eligibility in the secondary prize pool. That is TF2 for both training and inference. </li>\n</ol>",
      "rawMarkdown": "1. No don't have to label the dataset manually since the boundary boxes ares already labelled in the train.csv under annotations. Of course if you want to web crawl and download photos of COTS to add, then manual labeling would be used there. \n2. You don't have to use TF2 but it is required for eligibility in the secondary prize pool. That is TF2 for both training and inference.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1611124,
      "author_name": "jacklinetum",
      "author_url": "",
      "post_date": "12/07/2021 18:35:29",
      "content": "<p>Hi,did you get help?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1625686,
      "author_name": "wilbertbhtan",
      "author_url": "",
      "post_date": "12/22/2021 04:27:23",
      "content": "<ol>\n<li>No don't have to label the dataset manually since the boundary boxes ares already labelled in the train.csv under annotations. Of course if you want to web crawl and download photos of COTS to add, then manual labeling would be used there. </li>\n<li>You don't have to use TF2 but it is required for eligibility in the secondary prize pool. That is TF2 for both training and inference. </li>\n</ol>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1608766": "Hello! This is my first time competing in a  competition. I would like some guideline regarding code and other stuff\n- I would like to know if we have to label dataset manually\n- I would also like to know do we have to use TF2 built in object detection API, if so then how to manage directory structure",
    "1611124": "Hi,did you get help?",
    "1625686": "1. No don't have to label the dataset manually since the boundary boxes ares already labelled in the train.csv under annotations. Of course if you want to web crawl and download photos of COTS to add, then manual labeling would be used there. \n2. You don't have to use TF2 but it is required for eligibility in the secondary prize pool. That is TF2 for both training and inference."
  },
  "source": "meta"
}