{
  "id": 446009,
  "title": "Keras is not accepting the folder structure of `train`",
  "url": "/competitions/UBC-OCEAN/discussion/446009",
  "author_name": "hbhutta",
  "post_date": "2023-10-09T22:52:59.716000",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Keras expects the folder to contain subdirectories each being named after one of the labels, so there should be five subdirectories since we have five labels.</p>\n<p>The problem is that I'm trying to change the folder structure of `train' so that Keras accepts it, but I'm getting an error that the file structure is read only. I also can't download the data and work with it locally since it's too much for my computer.</p>\n<p>What to do?</p>",
  "messages": [
    {
      "id": 2475425,
      "postDate": "2023-10-09T22:52:59.717Z",
      "content": "<p>Keras expects the folder to contain subdirectories each being named after one of the labels, so there should be five subdirectories since we have five labels.</p>\n<p>The problem is that I'm trying to change the folder structure of `train' so that Keras accepts it, but I'm getting an error that the file structure is read only. I also can't download the data and work with it locally since it's too much for my computer.</p>\n<p>What to do?</p>",
      "rawMarkdown": "Keras expects the folder to contain subdirectories each being named after one of the labels, so there should be five subdirectories since we have five labels.\n\nThe problem is that I'm trying to change the folder structure of `train' so that Keras accepts it, but I'm getting an error that the file structure is read only. I also can't download the data and work with it locally since it's too much for my computer.\n\nWhat to do?",
      "votes": 1
    },
    {
      "id": 2486268,
      "postDate": "2023-10-17T19:37:00.810Z",
      "content": "<p>The train folders i.e  \"..kaggle/input/working\"is read only. You can't modify it. You will have to make your own pipeline for train/validation data. I think the data is too big to download - haven't looked at it yet. </p>\n<p>Another option is to work with any dataset other competitor has uploaded.</p>",
      "rawMarkdown": "The train folders i.e  \"..kaggle/input/working\"is read only. You can't modify it. You will have to make your own pipeline for train/validation data. I think the data is too big to download - haven't looked at it yet. \n\nAnother option is to work with any dataset other competitor has uploaded."
    },
    {
      "id": 2476686,
      "postDate": "2023-10-10T20:55:59.460Z",
      "content": "<p>Anyone have an idea? I can't proceed otherwise..</p>",
      "rawMarkdown": "Anyone have an idea? I can't proceed otherwise..",
      "replies": [
        {
          "id": 2486271,
          "postDate": "2023-10-17T19:41:29.667Z",
          "content": "<p>check the following link. Someone has made a dataset.</p>\n<p><a href=\"https://www.kaggle.com/competitions/UBC-OCEAN/discussion/446104\" target=\"_blank\">https://www.kaggle.com/competitions/UBC-OCEAN/discussion/446104</a></p>",
          "rawMarkdown": "check the following link. Someone has made a dataset.\n\nhttps://www.kaggle.com/competitions/UBC-OCEAN/discussion/446104"
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 2486268,
      "author_name": "Sandy",
      "author_url": "",
      "post_date": "2023-10-17T19:37:00.810000",
      "content": "<p>The train folders i.e  \"..kaggle/input/working\"is read only. You can't modify it. You will have to make your own pipeline for train/validation data. I think the data is too big to download - haven't looked at it yet. </p>\n<p>Another option is to work with any dataset other competitor has uploaded.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2476686,
      "author_name": "hbhutta",
      "author_url": "",
      "post_date": "2023-10-10T20:55:59.460000",
      "content": "<p>Anyone have an idea? I can't proceed otherwise..</p>",
      "votes": 0,
      "replies": [
        {
          "id": 2486271,
          "author_name": "Sandy",
          "author_url": "",
          "post_date": "2023-10-17T19:41:29.667000",
          "content": "<p>check the following link. Someone has made a dataset.</p>\n<p><a href=\"https://www.kaggle.com/competitions/UBC-OCEAN/discussion/446104\" target=\"_blank\">https://www.kaggle.com/competitions/UBC-OCEAN/discussion/446104</a></p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2475425": "Keras expects the folder to contain subdirectories each being named after one of the labels, so there should be five subdirectories since we have five labels.\n\nThe problem is that I'm trying to change the folder structure of `train' so that Keras accepts it, but I'm getting an error that the file structure is read only. I also can't download the data and work with it locally since it's too much for my computer.\n\nWhat to do?",
    "2486268": "The train folders i.e  \"..kaggle/input/working\"is read only. You can't modify it. You will have to make your own pipeline for train/validation data. I think the data is too big to download - haven't looked at it yet. \n\nAnother option is to work with any dataset other competitor has uploaded.",
    "2476686": "Anyone have an idea? I can't proceed otherwise.."
  }
}