{
  "id": 125512,
  "title": "Can I train model on my local server and submit it?",
  "url": "/competitions/deepfake-detection-challenge/discussion/125512",
  "author_name": "",
  "post_date": "2020-01-11T10:18:13.039319100Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I suppose the whole training dataset is too huge to process on Kaggle kernel. \nSo what my plan is that training on my local server and submits the output without using Kaggle kernel. I think this competition is not kernel only, maybe I can use it but I wanted to make sure about this.</p>",
  "messages": [
    {
      "id": "716147",
      "postDate": "01/11/2020 10:18:13",
      "content": "<p>I suppose the whole training dataset is too huge to process on Kaggle kernel. \nSo what my plan is that training on my local server and submits the output without using Kaggle kernel. I think this competition is not kernel only, maybe I can use it but I wanted to make sure about this.</p>",
      "rawMarkdown": "I suppose the whole training dataset is too huge to process on Kaggle kernel. \nSo what my plan is that training on my local server and submits the output without using Kaggle kernel. I think this competition is not kernel only, maybe I can use it but I wanted to make sure about this.",
      "votes": null
    },
    {
      "id": "716156",
      "postDate": "01/11/2020 10:30:10",
      "content": "<p>This is a code competition. You can train on your local server on the whole training set and then upload your model to perform inference on the test set, but you have to do it in a Kaggle kernel. You can't just submit the output.</p>",
      "rawMarkdown": "This is a code competition. You can train on your local server on the whole training set and then upload your model to perform inference on the test set, but you have to do it in a Kaggle kernel. You can't just submit the output.",
      "votes": null
    },
    {
      "id": "717341",
      "postDate": "01/13/2020 03:32:22",
      "content": "<p>Thanks for your reply! </p>",
      "rawMarkdown": "Thanks for your reply!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 716156,
      "author_name": "nicobonne",
      "author_url": "",
      "post_date": "01/11/2020 10:30:10",
      "content": "<p>This is a code competition. You can train on your local server on the whole training set and then upload your model to perform inference on the test set, but you have to do it in a Kaggle kernel. You can't just submit the output.</p>",
      "votes": null,
      "replies": [
        {
          "id": 717341,
          "author_name": "choonje",
          "author_url": "",
          "post_date": "01/13/2020 03:32:22",
          "content": "<p>Thanks for your reply! </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "716147": "I suppose the whole training dataset is too huge to process on Kaggle kernel. \nSo what my plan is that training on my local server and submits the output without using Kaggle kernel. I think this competition is not kernel only, maybe I can use it but I wanted to make sure about this.",
    "716156": "This is a code competition. You can train on your local server on the whole training set and then upload your model to perform inference on the test set, but you have to do it in a Kaggle kernel. You can't just submit the output.",
    "717341": "Thanks for your reply!"
  },
  "source": "meta"
}