{
  "id": 130201,
  "title": "problem with submission",
  "url": "/competitions/deepfake-detection-challenge/discussion/130201",
  "author_name": "",
  "post_date": "2020-02-12T17:27:10.444916700Z",
  "votes": null,
  "comment_count": 5,
  "views": 0,
  "content": "<p>I train and predict labels on my machine and then export my submission.csv into an appropriate folder. I load it on the plateforme and then export my submission output.</p>\n\n<p>When I submit to the competition I have an error : Submission Scoring Error\nI don't understand what it is </p>\n\n<p>Could you help me ?</p>\n\n<p>thanks for your help</p>",
  "messages": [
    {
      "id": "744257",
      "postDate": "02/12/2020 17:27:10",
      "content": "<p>I train and predict labels on my machine and then export my submission.csv into an appropriate folder. I load it on the plateforme and then export my submission output.</p>\n\n<p>When I submit to the competition I have an error : Submission Scoring Error\nI don't understand what it is </p>\n\n<p>Could you help me ?</p>\n\n<p>thanks for your help</p>",
      "rawMarkdown": "I train and predict labels on my machine and then export my submission.csv into an appropriate folder. I load it on the plateforme and then export my submission output.\n\nWhen I submit to the competition I have an error : Submission Scoring Error\nI don't understand what it is \n\nCould you help me ?\n\nthanks for your help",
      "votes": null
    },
    {
      "id": "744370",
      "postDate": "02/12/2020 19:13:14",
      "content": "<p>You can't predict labels on your own machine, you must do it in a notebook on Kaggle. The test set for submission is different than the test set that is available for download.</p>",
      "rawMarkdown": "You can't predict labels on your own machine, you must do it in a notebook on Kaggle. The test set for submission is different than the test set that is available for download.",
      "votes": null
    },
    {
      "id": "754914",
      "postDate": "02/24/2020 08:00:39",
      "content": "<p>I have the following message :</p>\n\n<p>Your notebook used 1708MB of dataset inputs, exceeding the competition 1024MB</p>\n\n<p>What can I do ???</p>",
      "rawMarkdown": "I have the following message :\n\nYour notebook used 1708MB of dataset inputs, exceeding the competition 1024MB\n\nWhat can I do ???",
      "votes": null
    },
    {
      "id": "754916",
      "postDate": "02/24/2020 08:06:10",
      "content": "<p>It's pretty clear what you need to do - reduce the size of your external dataset. What have you got in it? Saved models? If so, upload fewer models, or upload only the model's parameter weights and not the optimizers etc. if they are included.</p>",
      "rawMarkdown": "It's pretty clear what you need to do - reduce the size of your external dataset. What have you got in it? Saved models? If so, upload fewer models, or upload only the model's parameter weights and not the optimizers etc. if they are included.",
      "votes": null
    },
    {
      "id": "755368",
      "postDate": "02/24/2020 17:56:46",
      "content": "<p>Ok, thanks James. I try to fix that by deleting some files ...</p>\n\n<p>but I also have the following message : </p>\n\n<p>Your Notebook cannot use internet access for this competition....</p>",
      "rawMarkdown": "Ok, thanks James. I try to fix that by deleting some files ...\n\nbut I also have the following message : \n\nYour Notebook cannot use internet access for this competition....",
      "votes": null
    },
    {
      "id": "755379",
      "postDate": "02/24/2020 18:14:39",
      "content": "<p>Yes so you need to turn off 'internet' on the right side...</p>",
      "rawMarkdown": "Yes so you need to turn off 'internet' on the right side...",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 744370,
      "author_name": "humananalog",
      "author_url": "",
      "post_date": "02/12/2020 19:13:14",
      "content": "<p>You can't predict labels on your own machine, you must do it in a notebook on Kaggle. The test set for submission is different than the test set that is available for download.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 754914,
      "author_name": "pyxide",
      "author_url": "",
      "post_date": "02/24/2020 08:00:39",
      "content": "<p>I have the following message :</p>\n\n<p>Your notebook used 1708MB of dataset inputs, exceeding the competition 1024MB</p>\n\n<p>What can I do ???</p>",
      "votes": null,
      "replies": [
        {
          "id": 754916,
          "author_name": "jamesphoward",
          "author_url": "",
          "post_date": "02/24/2020 08:06:10",
          "content": "<p>It's pretty clear what you need to do - reduce the size of your external dataset. What have you got in it? Saved models? If so, upload fewer models, or upload only the model's parameter weights and not the optimizers etc. if they are included.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 755368,
      "author_name": "pyxide",
      "author_url": "",
      "post_date": "02/24/2020 17:56:46",
      "content": "<p>Ok, thanks James. I try to fix that by deleting some files ...</p>\n\n<p>but I also have the following message : </p>\n\n<p>Your Notebook cannot use internet access for this competition....</p>",
      "votes": null,
      "replies": [
        {
          "id": 755379,
          "author_name": "jamesphoward",
          "author_url": "",
          "post_date": "02/24/2020 18:14:39",
          "content": "<p>Yes so you need to turn off 'internet' on the right side...</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "744257": "I train and predict labels on my machine and then export my submission.csv into an appropriate folder. I load it on the plateforme and then export my submission output.\n\nWhen I submit to the competition I have an error : Submission Scoring Error\nI don't understand what it is \n\nCould you help me ?\n\nthanks for your help",
    "744370": "You can't predict labels on your own machine, you must do it in a notebook on Kaggle. The test set for submission is different than the test set that is available for download.",
    "754914": "I have the following message :\n\nYour notebook used 1708MB of dataset inputs, exceeding the competition 1024MB\n\nWhat can I do ???",
    "754916": "It's pretty clear what you need to do - reduce the size of your external dataset. What have you got in it? Saved models? If so, upload fewer models, or upload only the model's parameter weights and not the optimizers etc. if they are included.",
    "755368": "Ok, thanks James. I try to fix that by deleting some files ...\n\nbut I also have the following message : \n\nYour Notebook cannot use internet access for this competition....",
    "755379": "Yes so you need to turn off 'internet' on the right side..."
  },
  "source": "meta"
}