{
  "id": 64370,
  "title": "Appending context to images",
  "url": "/competitions/airbus-ship-detection/discussion/64370",
  "author_name": "MingRuiWang",
  "post_date": "2018-08-28T17:21:22.681000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi all, a small minority of images have land and port in them (like <code>00c996565.jpg</code> from the test set).</p>\n\n<p>It would be very helpful if we know the setting of each image, ie. sea, land, port, island. The objective is to target a images with land and ports for more meaningful training.</p>\n\n<p>I'd like to check if anyone has already done such preprocessing before.</p>",
  "messages": [
    {
      "id": 377172,
      "postDate": "2018-08-28T17:21:22.680Z",
      "content": "<p>Hi all, a small minority of images have land and port in them (like <code>00c996565.jpg</code> from the test set).</p>\n\n<p>It would be very helpful if we know the setting of each image, ie. sea, land, port, island. The objective is to target a images with land and ports for more meaningful training.</p>\n\n<p>I'd like to check if anyone has already done such preprocessing before.</p>",
      "rawMarkdown": "Hi all, a small minority of images have land and port in them (like `00c996565.jpg` from the test set).\n\nIt would be very helpful if we know the setting of each image, ie. sea, land, port, island. The objective is to target a images with land and ports for more meaningful training.\n\nI'd like to check if anyone has already done such preprocessing before.",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "377172": "Hi all, a small minority of images have land and port in them (like `00c996565.jpg` from the test set).\n\nIt would be very helpful if we know the setting of each image, ie. sea, land, port, island. The objective is to target a images with land and ports for more meaningful training.\n\nI'd like to check if anyone has already done such preprocessing before."
  }
}