{
  "id": 312505,
  "title": "How did you download cropped datasets and use it on colab.",
  "url": "/competitions/happy-whale-and-dolphin/discussion/312505",
  "author_name": "",
  "post_date": "2022-03-12T13:12:00.098030100Z",
  "votes": 3,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I'm new for Kaggle.<br>\nI would like to ask you how to download cropped datasets from <a href=\"https://www.kaggle.com/phalanx\" target=\"_blank\">@phalanx</a> and use it on colab.<br>\nWhen you create model on colab using customized datasets, you press the button 'download' and move the datasets to google drive?? <br>\nWhen I do it, it takes a lot of time to move huge image datasets…<br>\nplease give me some advice.</p>",
  "messages": [
    {
      "id": "1720100",
      "postDate": "03/12/2022 13:12:00",
      "content": "<p>I'm new for Kaggle.<br>\nI would like to ask you how to download cropped datasets from <a href=\"https://www.kaggle.com/phalanx\" target=\"_blank\">@phalanx</a> and use it on colab.<br>\nWhen you create model on colab using customized datasets, you press the button 'download' and move the datasets to google drive?? <br>\nWhen I do it, it takes a lot of time to move huge image datasets…<br>\nplease give me some advice.</p>",
      "rawMarkdown": "I'm new for Kaggle.\nI would like to ask you how to download cropped datasets from @phalanx and use it on colab.\nWhen you create model on colab using customized datasets, you press the button 'download' and move the datasets to google drive?? \nWhen I do it, it takes a lot of time to move huge image datasets...\nplease give me some advice.",
      "votes": null
    },
    {
      "id": "1720125",
      "postDate": "03/12/2022 13:46:19",
      "content": "<p>I'm not a Colab power user but from what I understand using GCS might help here. IIRC it was discussed <a href=\"https://www.kaggle.com/c/happy-whale-and-dolphin/discussion/307995\" target=\"_blank\">here</a></p>\n<p>I hope this helps! :)</p>",
      "rawMarkdown": "I'm not a Colab power user but from what I understand using GCS might help here. IIRC it was discussed [here](https://www.kaggle.com/c/happy-whale-and-dolphin/discussion/307995)\n\nI hope this helps! :)",
      "votes": null
    },
    {
      "id": "1720130",
      "postDate": "03/12/2022 13:50:44",
      "content": "<p>Thank you for comment!<br>\nI usually use pytorch so i did not know the method.<br>\nI gonna try it!</p>",
      "rawMarkdown": "Thank you for comment!\nI usually use pytorch so i did not know the method.\nI gonna try it!",
      "votes": null
    },
    {
      "id": "1720211",
      "postDate": "03/12/2022 15:06:38",
      "content": "<p>Keep us posted on how it goes! </p>\n<p>Btw, it seems it's a bit confusing for all-I would also suggest sharing a tutorial once you set it up</p>",
      "rawMarkdown": "Keep us posted on how it goes! \n\nBtw, it seems it's a bit confusing for all-I would also suggest sharing a tutorial once you set it up",
      "votes": null
    },
    {
      "id": "1722376",
      "postDate": "03/14/2022 13:30:53",
      "content": "<p>if tfrecords, you can use gcs path directly.</p>\n<p>below is how to copy to gdrive.</p>\n<p>also you can use kaggle api to download to gdrive, however if dataset is too big, there will be problem for disconnection.</p>\n<p><a href=\"https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300601#1648291\" target=\"_blank\">https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300601#1648291</a></p>\n<p>upvote if It helps.</p>",
      "rawMarkdown": "if tfrecords, you can use gcs path directly.\n\nbelow is how to copy to gdrive.\n\nalso you can use kaggle api to download to gdrive, however if dataset is too big, there will be problem for disconnection.\n\n\nhttps://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300601#1648291\n\nupvote if It helps.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1720125,
      "author_name": "init27",
      "author_url": "",
      "post_date": "03/12/2022 13:46:19",
      "content": "<p>I'm not a Colab power user but from what I understand using GCS might help here. IIRC it was discussed <a href=\"https://www.kaggle.com/c/happy-whale-and-dolphin/discussion/307995\" target=\"_blank\">here</a></p>\n<p>I hope this helps! :)</p>",
      "votes": null,
      "replies": [
        {
          "id": 1720130,
          "author_name": "haruki741",
          "author_url": "",
          "post_date": "03/12/2022 13:50:44",
          "content": "<p>Thank you for comment!<br>\nI usually use pytorch so i did not know the method.<br>\nI gonna try it!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1720211,
          "author_name": "init27",
          "author_url": "",
          "post_date": "03/12/2022 15:06:38",
          "content": "<p>Keep us posted on how it goes! </p>\n<p>Btw, it seems it's a bit confusing for all-I would also suggest sharing a tutorial once you set it up</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1722376,
      "author_name": "dragonzhang",
      "author_url": "",
      "post_date": "03/14/2022 13:30:53",
      "content": "<p>if tfrecords, you can use gcs path directly.</p>\n<p>below is how to copy to gdrive.</p>\n<p>also you can use kaggle api to download to gdrive, however if dataset is too big, there will be problem for disconnection.</p>\n<p><a href=\"https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300601#1648291\" target=\"_blank\">https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300601#1648291</a></p>\n<p>upvote if It helps.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1720100": "I'm new for Kaggle.\nI would like to ask you how to download cropped datasets from @phalanx and use it on colab.\nWhen you create model on colab using customized datasets, you press the button 'download' and move the datasets to google drive?? \nWhen I do it, it takes a lot of time to move huge image datasets...\nplease give me some advice.",
    "1720125": "I'm not a Colab power user but from what I understand using GCS might help here. IIRC it was discussed [here](https://www.kaggle.com/c/happy-whale-and-dolphin/discussion/307995)\n\nI hope this helps! :)",
    "1720130": "Thank you for comment!\nI usually use pytorch so i did not know the method.\nI gonna try it!",
    "1720211": "Keep us posted on how it goes! \n\nBtw, it seems it's a bit confusing for all-I would also suggest sharing a tutorial once you set it up",
    "1722376": "if tfrecords, you can use gcs path directly.\n\nbelow is how to copy to gdrive.\n\nalso you can use kaggle api to download to gdrive, however if dataset is too big, there will be problem for disconnection.\n\n\nhttps://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300601#1648291\n\nupvote if It helps."
  },
  "source": "meta"
}