{
  "id": 156533,
  "title": "Augmentations and Data Storage",
  "url": "/competitions/siim-isic-melanoma-classification/discussion/156533",
  "author_name": "",
  "post_date": "2020-06-06T15:33:46.818220600Z",
  "votes": 2,
  "comment_count": 5,
  "views": 0,
  "content": "<p>From the kaggle NoteBook I have prepared over 10000 image augments for class 1. Instead of doing it again can I save the images somewhere permanently in the kaggle or can I publish it as a Dataset so that it can be used again later? Is it possible to publish dataset from kaggle kernel? Because whenever I store the images under kaggle/working it disappears whenever I restart the session.</p>",
  "messages": [
    {
      "id": "876269",
      "postDate": "06/06/2020 15:33:46",
      "content": "<p>From the kaggle NoteBook I have prepared over 10000 image augments for class 1. Instead of doing it again can I save the images somewhere permanently in the kaggle or can I publish it as a Dataset so that it can be used again later? Is it possible to publish dataset from kaggle kernel? Because whenever I store the images under kaggle/working it disappears whenever I restart the session.</p>",
      "rawMarkdown": "From the kaggle NoteBook I have prepared over 10000 image augments for class 1. Instead of doing it again can I save the images somewhere permanently in the kaggle or can I publish it as a Dataset so that it can be used again later? Is it possible to publish dataset from kaggle kernel? Because whenever I store the images under kaggle/working it disappears whenever I restart the session.",
      "votes": null
    },
    {
      "id": "876356",
      "postDate": "06/06/2020 16:32:23",
      "content": "<p>You should commit your notebook. But before commiting it dont save as it is. Compress the augumented images to a zip file and then store in the working directory then commit.So that after the notebook commits you will not run out of memory and you notebook wont get crashed.\nAnd finally make your dataset public😊  </p>",
      "rawMarkdown": "You should commit your notebook. But before commiting it dont save as it is. Compress the augumented images to a zip file and then store in the working directory then commit.So that after the notebook commits you will not run out of memory and you notebook wont get crashed.\nAnd finally make your dataset public😊",
      "votes": null
    },
    {
      "id": "876720",
      "postDate": "06/07/2020 01:00:05",
      "content": "<p>Thanks a lot Sharuk</p>",
      "rawMarkdown": "Thanks a lot Sharuk",
      "votes": null
    },
    {
      "id": "877065",
      "postDate": "06/07/2020 09:38:43",
      "content": "<p>I did it <a href=\"/msharuk589\">@msharuk589</a> Thanks. I created Image augments from the external dataset and saved it under kaggle/working and then hit commit. Now the kernel shows the output visualizations but there is no link to download/add new dataset. What is the problem? Link to the kernel <a href=\"https://www.kaggle.com/aakashveera/dataset-creater-siim/\">https://www.kaggle.com/aakashveera/dataset-creater-siim/</a></p>",
      "rawMarkdown": "I did it @msharuk589 Thanks. I created Image augments from the external dataset and saved it under kaggle/working and then hit commit. Now the kernel shows the output visualizations but there is no link to download/add new dataset. What is the problem? Link to the kernel [https://www.kaggle.com/aakashveera/dataset-creater-siim/](https://www.kaggle.com/aakashveera/dataset-creater-siim/)",
      "votes": null
    },
    {
      "id": "877077",
      "postDate": "06/07/2020 09:54:22",
      "content": "<p>Yeah when the output dataset exceeds 500MB i also couldnt create a dataset\nBut you could use this in a kernel by the \"Add Data\" option in the kernel\nSo that you can use it\nEDIT : BTW your link isnt working. Check it da😑 </p>",
      "rawMarkdown": "Yeah when the output dataset exceeds 500MB i also couldnt create a dataset\nBut you could use this in a kernel by the \"Add Data\" option in the kernel\nSo that you can use it\nEDIT : BTW your link isnt working. Check it da😑",
      "votes": null
    },
    {
      "id": "897058",
      "postDate": "06/22/2020 15:39:41",
      "content": "<p>You can write the augmented images into new folder and create zip file of that folder so you are able to see the download or copy option exactly right side to that folder.</p>",
      "rawMarkdown": "You can write the augmented images into new folder and create zip file of that folder so you are able to see the download or copy option exactly right side to that folder.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 876356,
      "author_name": "msharuk589",
      "author_url": "",
      "post_date": "06/06/2020 16:32:23",
      "content": "<p>You should commit your notebook. But before commiting it dont save as it is. Compress the augumented images to a zip file and then store in the working directory then commit.So that after the notebook commits you will not run out of memory and you notebook wont get crashed.\nAnd finally make your dataset public😊  </p>",
      "votes": null,
      "replies": [
        {
          "id": 876720,
          "author_name": "aakashveera",
          "author_url": "",
          "post_date": "06/07/2020 01:00:05",
          "content": "<p>Thanks a lot Sharuk</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 877065,
          "author_name": "aakashveera",
          "author_url": "",
          "post_date": "06/07/2020 09:38:43",
          "content": "<p>I did it <a href=\"/msharuk589\">@msharuk589</a> Thanks. I created Image augments from the external dataset and saved it under kaggle/working and then hit commit. Now the kernel shows the output visualizations but there is no link to download/add new dataset. What is the problem? Link to the kernel <a href=\"https://www.kaggle.com/aakashveera/dataset-creater-siim/\">https://www.kaggle.com/aakashveera/dataset-creater-siim/</a></p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 877077,
          "author_name": "msharuk589",
          "author_url": "",
          "post_date": "06/07/2020 09:54:22",
          "content": "<p>Yeah when the output dataset exceeds 500MB i also couldnt create a dataset\nBut you could use this in a kernel by the \"Add Data\" option in the kernel\nSo that you can use it\nEDIT : BTW your link isnt working. Check it da😑 </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 897058,
          "author_name": "margipatel",
          "author_url": "",
          "post_date": "06/22/2020 15:39:41",
          "content": "<p>You can write the augmented images into new folder and create zip file of that folder so you are able to see the download or copy option exactly right side to that folder.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "876269": "From the kaggle NoteBook I have prepared over 10000 image augments for class 1. Instead of doing it again can I save the images somewhere permanently in the kaggle or can I publish it as a Dataset so that it can be used again later? Is it possible to publish dataset from kaggle kernel? Because whenever I store the images under kaggle/working it disappears whenever I restart the session.",
    "876356": "You should commit your notebook. But before commiting it dont save as it is. Compress the augumented images to a zip file and then store in the working directory then commit.So that after the notebook commits you will not run out of memory and you notebook wont get crashed.\nAnd finally make your dataset public😊",
    "876720": "Thanks a lot Sharuk",
    "877065": "I did it @msharuk589 Thanks. I created Image augments from the external dataset and saved it under kaggle/working and then hit commit. Now the kernel shows the output visualizations but there is no link to download/add new dataset. What is the problem? Link to the kernel [https://www.kaggle.com/aakashveera/dataset-creater-siim/](https://www.kaggle.com/aakashveera/dataset-creater-siim/)",
    "877077": "Yeah when the output dataset exceeds 500MB i also couldnt create a dataset\nBut you could use this in a kernel by the \"Add Data\" option in the kernel\nSo that you can use it\nEDIT : BTW your link isnt working. Check it da😑",
    "897058": "You can write the augmented images into new folder and create zip file of that folder so you are able to see the download or copy option exactly right side to that folder."
  },
  "source": "meta"
}