{
  "id": 284057,
  "title": "I visualized every training image so you don't have to",
  "url": "/competitions/sartorius-cell-instance-segmentation/discussion/284057",
  "author_name": "",
  "post_date": "2021-10-29T10:25:28.695219500Z",
  "votes": 83,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I noticed I'm losing too much time when I'm visualizing images and their annotations while working on my local machine. I decided make a notebook which contains every training image and annotations visualized and grouped by their cell types.</p>\n<p>Link to the notebook -&gt; <a href=\"https://www.kaggle.com/gunesevitan/sartorius-cell-instance-segmentation-eda\" target=\"_blank\">https://www.kaggle.com/gunesevitan/sartorius-cell-instance-segmentation-eda</a></p>",
  "messages": [
    {
      "id": "1564537",
      "postDate": "10/29/2021 10:25:28",
      "content": "<p>I noticed I'm losing too much time when I'm visualizing images and their annotations while working on my local machine. I decided make a notebook which contains every training image and annotations visualized and grouped by their cell types.</p>\n<p>Link to the notebook -&gt; <a href=\"https://www.kaggle.com/gunesevitan/sartorius-cell-instance-segmentation-eda\" target=\"_blank\">https://www.kaggle.com/gunesevitan/sartorius-cell-instance-segmentation-eda</a></p>",
      "rawMarkdown": "I noticed I'm losing too much time when I'm visualizing images and their annotations while working on my local machine. I decided make a notebook which contains every training image and annotations visualized and grouped by their cell types.\n\nLink to the notebook -> https://www.kaggle.com/gunesevitan/sartorius-cell-instance-segmentation-eda",
      "votes": null
    },
    {
      "id": "1582825",
      "postDate": "11/15/2021 10:11:50",
      "content": "<p>Thx for your work mate, i have one question:<br>\nYou have plotted the distribution of cell type in the train data set and the semi supervised and both thoghter.<br>\nBecause it is a counting distribution shouldn't be the last histogram the sum of the other two?</p>",
      "rawMarkdown": "Thx for your work mate, i have one question:\nYou have plotted the distribution of cell type in the train data set and the semi supervised and both thoghter.\nBecause it is a counting distribution shouldn't be the last histogram the sum of the other two?",
      "votes": null
    },
    {
      "id": "1583478",
      "postDate": "11/15/2021 21:18:13",
      "content": "<p>Ok I checked it myself and it's due to the labels in the first bar plot. The labels are wrong named there but I think it's no big deal for the bigger picture. </p>",
      "rawMarkdown": "Ok I checked it myself and it's due to the labels in the first bar plot. The labels are wrong named there but I think it's no big deal for the bigger picture.",
      "votes": null
    },
    {
      "id": "1588187",
      "postDate": "11/19/2021 09:08:15",
      "content": "<p>Thanks for the feedback. I'm fixing the issue.</p>",
      "rawMarkdown": "Thanks for the feedback. I'm fixing the issue.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1582825,
      "author_name": "andreashorlbeck",
      "author_url": "",
      "post_date": "11/15/2021 10:11:50",
      "content": "<p>Thx for your work mate, i have one question:<br>\nYou have plotted the distribution of cell type in the train data set and the semi supervised and both thoghter.<br>\nBecause it is a counting distribution shouldn't be the last histogram the sum of the other two?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1583478,
          "author_name": "andreashorlbeck",
          "author_url": "",
          "post_date": "11/15/2021 21:18:13",
          "content": "<p>Ok I checked it myself and it's due to the labels in the first bar plot. The labels are wrong named there but I think it's no big deal for the bigger picture. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1588187,
          "author_name": "gunesevitan",
          "author_url": "",
          "post_date": "11/19/2021 09:08:15",
          "content": "<p>Thanks for the feedback. I'm fixing the issue.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1564537": "I noticed I'm losing too much time when I'm visualizing images and their annotations while working on my local machine. I decided make a notebook which contains every training image and annotations visualized and grouped by their cell types.\n\nLink to the notebook -> https://www.kaggle.com/gunesevitan/sartorius-cell-instance-segmentation-eda",
    "1582825": "Thx for your work mate, i have one question:\nYou have plotted the distribution of cell type in the train data set and the semi supervised and both thoghter.\nBecause it is a counting distribution shouldn't be the last histogram the sum of the other two?",
    "1583478": "Ok I checked it myself and it's due to the labels in the first bar plot. The labels are wrong named there but I think it's no big deal for the bigger picture.",
    "1588187": "Thanks for the feedback. I'm fixing the issue."
  },
  "source": "meta"
}