{
  "id": 305180,
  "title": "Handling with grayscale pics",
  "url": "/competitions/happy-whale-and-dolphin/discussion/305180",
  "author_name": "",
  "post_date": "2022-02-04T05:44:48.432227100Z",
  "votes": 7,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Good morning! I have noticed that default dataset has &gt;100 grayscale pictures. How I can handle with it? Just drop it or convert all pictures (50000+) to grayscale format and deal with it? Thanks!</p>",
  "messages": [
    {
      "id": "1675231",
      "postDate": "02/04/2022 05:44:48",
      "content": "<p>Good morning! I have noticed that default dataset has &gt;100 grayscale pictures. How I can handle with it? Just drop it or convert all pictures (50000+) to grayscale format and deal with it? Thanks!</p>",
      "rawMarkdown": "Good morning! I have noticed that default dataset has >100 grayscale pictures. How I can handle with it? Just drop it or convert all pictures (50000+) to grayscale format and deal with it? Thanks!",
      "votes": null
    },
    {
      "id": "1675268",
      "postDate": "02/04/2022 06:21:19",
      "content": "<p>There is only one way to find out :) </p>",
      "rawMarkdown": "There is only one way to find out :)",
      "votes": null
    },
    {
      "id": "1675481",
      "postDate": "02/04/2022 09:48:38",
      "content": "<p>Or inverted approach - convert it to RGB image from grayscale (obviously, you will preserve gray tones, but technically you will work with a 3-channel image). </p>\n<p>Anyway, as <a href=\"https://www.kaggle.com/drhabib\" target=\"_blank\">@drhabib</a> mentioned - there is only one way to figure out which approach it best</p>",
      "rawMarkdown": "Or inverted approach - convert it to RGB image from grayscale (obviously, you will preserve gray tones, but technically you will work with a 3-channel image). \n\nAnyway, as @drhabib mentioned - there is only one way to figure out which approach it best",
      "votes": null
    },
    {
      "id": "1676897",
      "postDate": "02/05/2022 11:35:52",
      "content": "<p>There are 2033 grayscale images in training dataset and 333 in test dataset BTW</p>",
      "rawMarkdown": "There are 2033 grayscale images in training dataset and 333 in test dataset BTW",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1675268,
      "author_name": "drhabib",
      "author_url": "",
      "post_date": "02/04/2022 06:21:19",
      "content": "<p>There is only one way to find out :) </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1675481,
      "author_name": "meowmeowmeowmeowmeow",
      "author_url": "",
      "post_date": "02/04/2022 09:48:38",
      "content": "<p>Or inverted approach - convert it to RGB image from grayscale (obviously, you will preserve gray tones, but technically you will work with a 3-channel image). </p>\n<p>Anyway, as <a href=\"https://www.kaggle.com/drhabib\" target=\"_blank\">@drhabib</a> mentioned - there is only one way to figure out which approach it best</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1676897,
      "author_name": "omomo17",
      "author_url": "",
      "post_date": "02/05/2022 11:35:52",
      "content": "<p>There are 2033 grayscale images in training dataset and 333 in test dataset BTW</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1675231": "Good morning! I have noticed that default dataset has >100 grayscale pictures. How I can handle with it? Just drop it or convert all pictures (50000+) to grayscale format and deal with it? Thanks!",
    "1675268": "There is only one way to find out :)",
    "1675481": "Or inverted approach - convert it to RGB image from grayscale (obviously, you will preserve gray tones, but technically you will work with a 3-channel image). \n\nAnyway, as @drhabib mentioned - there is only one way to figure out which approach it best",
    "1676897": "There are 2033 grayscale images in training dataset and 333 in test dataset BTW"
  },
  "source": "meta"
}