{
  "id": 158564,
  "title": "Is anyone using Yolo here ? If yes, please reply to this thread. I need some help !!!",
  "url": "/competitions/global-wheat-detection/discussion/158564",
  "author_name": "",
  "post_date": "2020-06-14T17:34:42.850376700Z",
  "votes": null,
  "comment_count": 10,
  "views": 0,
  "content": "",
  "messages": [
    {
      "id": "886084",
      "postDate": "06/14/2020 17:34:42",
      "content": "",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "886207",
      "postDate": "06/14/2020 19:38:09",
      "content": "<p>Yes!!</p>",
      "rawMarkdown": "Yes!!",
      "votes": null
    },
    {
      "id": "886289",
      "postDate": "06/14/2020 21:42:16",
      "content": "<p>Can you please give me the code to convert train.csv file to yolo annotations format?\ni tried everywhere but no luck</p>",
      "rawMarkdown": "Can you please give me the code to convert train.csv file to yolo annotations format?\ni tried everywhere but no luck",
      "votes": null
    },
    {
      "id": "888573",
      "postDate": "06/16/2020 12:50:03",
      "content": "<p>i am done with training and my model is working quite nicely , i just want to know to export predicted bounding box co-ordinates in csv file to submit it on kaggle.</p>",
      "rawMarkdown": "i am done with training and my model is working quite nicely , i just want to know to export predicted bounding box co-ordinates in csv file to submit it on kaggle.",
      "votes": null
    },
    {
      "id": "888575",
      "postDate": "06/16/2020 12:50:42",
      "content": "<p>i can't find any way to output bounding box coordinates in text.</p>",
      "rawMarkdown": "i can't find any way to output bounding box coordinates in text.",
      "votes": null
    },
    {
      "id": "890982",
      "postDate": "06/17/2020 19:58:26",
      "content": "<p>there is no need for complex coding . just remember the data given to us in tran.csv contains\n[xmin, ymin, w , h]\nwe need to convert those into \n[centerX, centerY , width , height ] following the official documentation in darknet </p>\n\n<p>So in order to get these coordinates according to yolo format \ncenter X = ( xmin + (w/2) / absolute width of the image)\ncenter Y = ( ymin + (h/2) / absolute height of the image)\nwidth =  (w/ absolute width of the image)\nheight =  (h/ absolute height of the image)</p>\n\n<p>follow this logic and youll get the desired yolo coordintes</p>",
      "rawMarkdown": "there is no need for complex coding . just remember the data given to us in tran.csv contains\n[xmin, ymin, w , h]\nwe need to convert those into \n[centerX, centerY , width , height ] following the official documentation in darknet \n\nSo in order to get these coordinates according to yolo format \ncenter X = ( xmin + (w/2) / absolute width of the image)\ncenter Y = ( ymin + (h/2) / absolute height of the image)\nwidth =  (w/ absolute width of the image)\nheight =  (h/ absolute height of the image)\n\nfollow this logic and youll get the desired yolo coordintes",
      "votes": null
    },
    {
      "id": "892761",
      "postDate": "06/19/2020 06:15:20",
      "content": "<p>Are you using opencv to make predicitions or darknet?</p>",
      "rawMarkdown": "Are you using opencv to make predicitions or darknet?",
      "votes": null
    },
    {
      "id": "893051",
      "postDate": "06/19/2020 10:48:20",
      "content": "<p><a href=\"/akshaydureja\">@akshaydureja</a> Take a look at this post from <a href=\"/pabloberhauser\">@pabloberhauser</a>: \n<a href=\"https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4\">https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4</a></p>",
      "rawMarkdown": "akshaydureja Take a look at this post from @pabloberhauser: \nhttps://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4",
      "votes": null
    },
    {
      "id": "893357",
      "postDate": "06/19/2020 14:51:58",
      "content": "<p>Hi <a href=\"/akshaydureja\">@akshaydureja</a> ,</p>\n\n<p><a href=\"https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4\">https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4</a> is using YOLO\n<a href=\"https://www.kaggle.com/orkatz2/yolov5-fake-or-real-single-model-l-b-0-73\">https://www.kaggle.com/orkatz2/yolov5-fake-or-real-single-model-l-b-0-73</a> is using YOlov5 and showing some conversion, maybe this is what you are looking for?</p>",
      "rawMarkdown": "Hi @akshaydureja ,\n\nhttps://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4 is using YOLO\nhttps://www.kaggle.com/orkatz2/yolov5-fake-or-real-single-model-l-b-0-73 is using YOlov5 and showing some conversion, maybe this is what you are looking for?",
      "votes": null
    },
    {
      "id": "901411",
      "postDate": "06/25/2020 13:07:40",
      "content": "<p>I'm trying to run v5. I'm getting this error:\nFile \"/kaggle/working/utils/datasets.py\", line 298, in init</p>\n\n<p>It's a write error. I need to write /train/images.shapes to the folder but input is not writeable. Is anyone running into this or how are you getting around it?</p>",
      "rawMarkdown": "I'm trying to run v5. I'm getting this error:\nFile \"/kaggle/working/utils/datasets.py\", line 298, in init\n\nIt's a write error. I need to write /train/images.shapes to the folder but input is not writeable. Is anyone running into this or how are you getting around it?",
      "votes": null
    },
    {
      "id": "921418",
      "postDate": "07/09/2020 08:58:17",
      "content": "<p><a href=\"/vishnukv64\">@vishnukv64</a> are you using AlexeyAB/darknet's YOLOv4? If yes, can you please share how to make prediction on test images and prepare the submission file. I tried to run <code>!chmod +x darknet</code> but it says chmod: changing permissions of 'darknet': Read-only file system while running the test command <code>!./darknet detector test..</code> command.</p>",
      "rawMarkdown": "vishnukv64 are you using AlexeyAB/darknet's YOLOv4? If yes, can you please share how to make prediction on test images and prepare the submission file. I tried to run `!chmod +x darknet` but it says chmod: changing permissions of 'darknet': Read-only file system while running the test command `!./darknet detector test..` command.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 886207,
      "author_name": "vishnukv64",
      "author_url": "",
      "post_date": "06/14/2020 19:38:09",
      "content": "<p>Yes!!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 886289,
      "author_name": "akshaydureja",
      "author_url": "",
      "post_date": "06/14/2020 21:42:16",
      "content": "<p>Can you please give me the code to convert train.csv file to yolo annotations format?\ni tried everywhere but no luck</p>",
      "votes": null,
      "replies": [
        {
          "id": 890982,
          "author_name": "vishnukv64",
          "author_url": "",
          "post_date": "06/17/2020 19:58:26",
          "content": "<p>there is no need for complex coding . just remember the data given to us in tran.csv contains\n[xmin, ymin, w , h]\nwe need to convert those into \n[centerX, centerY , width , height ] following the official documentation in darknet </p>\n\n<p>So in order to get these coordinates according to yolo format \ncenter X = ( xmin + (w/2) / absolute width of the image)\ncenter Y = ( ymin + (h/2) / absolute height of the image)\nwidth =  (w/ absolute width of the image)\nheight =  (h/ absolute height of the image)</p>\n\n<p>follow this logic and youll get the desired yolo coordintes</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 893051,
          "author_name": "msjose",
          "author_url": "",
          "post_date": "06/19/2020 10:48:20",
          "content": "<p><a href=\"/akshaydureja\">@akshaydureja</a> Take a look at this post from <a href=\"/pabloberhauser\">@pabloberhauser</a>: \n<a href=\"https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4\">https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4</a></p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 921418,
          "author_name": "iamprateek",
          "author_url": "",
          "post_date": "07/09/2020 08:58:17",
          "content": "<p><a href=\"/vishnukv64\">@vishnukv64</a> are you using AlexeyAB/darknet's YOLOv4? If yes, can you please share how to make prediction on test images and prepare the submission file. I tried to run <code>!chmod +x darknet</code> but it says chmod: changing permissions of 'darknet': Read-only file system while running the test command <code>!./darknet detector test..</code> command.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 888573,
      "author_name": "akshaydureja",
      "author_url": "",
      "post_date": "06/16/2020 12:50:03",
      "content": "<p>i am done with training and my model is working quite nicely , i just want to know to export predicted bounding box co-ordinates in csv file to submit it on kaggle.</p>",
      "votes": null,
      "replies": [
        {
          "id": 892761,
          "author_name": "vishnukv64",
          "author_url": "",
          "post_date": "06/19/2020 06:15:20",
          "content": "<p>Are you using opencv to make predicitions or darknet?</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 888575,
      "author_name": "akshaydureja",
      "author_url": "",
      "post_date": "06/16/2020 12:50:42",
      "content": "<p>i can't find any way to output bounding box coordinates in text.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 893357,
      "author_name": "ChristianDenich",
      "author_url": "",
      "post_date": "06/19/2020 14:51:58",
      "content": "<p>Hi <a href=\"/akshaydureja\">@akshaydureja</a> ,</p>\n\n<p><a href=\"https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4\">https://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4</a> is using YOLO\n<a href=\"https://www.kaggle.com/orkatz2/yolov5-fake-or-real-single-model-l-b-0-73\">https://www.kaggle.com/orkatz2/yolov5-fake-or-real-single-model-l-b-0-73</a> is using YOlov5 and showing some conversion, maybe this is what you are looking for?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 901411,
      "author_name": "crained",
      "author_url": "",
      "post_date": "06/25/2020 13:07:40",
      "content": "<p>I'm trying to run v5. I'm getting this error:\nFile \"/kaggle/working/utils/datasets.py\", line 298, in init</p>\n\n<p>It's a write error. I need to write /train/images.shapes to the folder but input is not writeable. Is anyone running into this or how are you getting around it?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "886084": "",
    "886207": "Yes!!",
    "886289": "Can you please give me the code to convert train.csv file to yolo annotations format?\ni tried everywhere but no luck",
    "888573": "i am done with training and my model is working quite nicely , i just want to know to export predicted bounding box co-ordinates in csv file to submit it on kaggle.",
    "888575": "i can't find any way to output bounding box coordinates in text.",
    "890982": "there is no need for complex coding . just remember the data given to us in tran.csv contains\n[xmin, ymin, w , h]\nwe need to convert those into \n[centerX, centerY , width , height ] following the official documentation in darknet \n\nSo in order to get these coordinates according to yolo format \ncenter X = ( xmin + (w/2) / absolute width of the image)\ncenter Y = ( ymin + (h/2) / absolute height of the image)\nwidth =  (w/ absolute width of the image)\nheight =  (h/ absolute height of the image)\n\nfollow this logic and youll get the desired yolo coordintes",
    "892761": "Are you using opencv to make predicitions or darknet?",
    "893051": "akshaydureja Take a look at this post from @pabloberhauser: \nhttps://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4",
    "893357": "Hi @akshaydureja ,\n\nhttps://www.kaggle.com/pabloberhauser/creating-label-files-for-use-in-yolov4 is using YOLO\nhttps://www.kaggle.com/orkatz2/yolov5-fake-or-real-single-model-l-b-0-73 is using YOlov5 and showing some conversion, maybe this is what you are looking for?",
    "901411": "I'm trying to run v5. I'm getting this error:\nFile \"/kaggle/working/utils/datasets.py\", line 298, in init\n\nIt's a write error. I need to write /train/images.shapes to the folder but input is not writeable. Is anyone running into this or how are you getting around it?",
    "921418": "vishnukv64 are you using AlexeyAB/darknet's YOLOv4? If yes, can you please share how to make prediction on test images and prepare the submission file. I tried to run `!chmod +x darknet` but it says chmod: changing permissions of 'darknet': Read-only file system while running the test command `!./darknet detector test..` command."
  },
  "source": "meta"
}