{
  "id": 493845,
  "title": "Importing models while submiting",
  "url": "/competitions/birdclef-2024/discussion/493845",
  "author_name": "",
  "post_date": "2024-04-15T05:41:42.724180300Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I have a small doubt regarding the submission, I do not know if someone has already answered this question, but I would appreciate any help. </p>\n<p>While submitting does the training time should also be under 2 hours or just the testing phase? </p>\n<p>If everything should be done in 2 hours, can I train my model on my local system and import it to Kaggle notebook while submitting. </p>",
  "messages": [
    {
      "id": "2752708",
      "postDate": "04/15/2024 05:41:42",
      "content": "<p>I have a small doubt regarding the submission, I do not know if someone has already answered this question, but I would appreciate any help. </p>\n<p>While submitting does the training time should also be under 2 hours or just the testing phase? </p>\n<p>If everything should be done in 2 hours, can I train my model on my local system and import it to Kaggle notebook while submitting. </p>",
      "rawMarkdown": "I have a small doubt regarding the submission, I do not know if someone has already answered this question, but I would appreciate any help. \n\nWhile submitting does the training time should also be under 2 hours or just the testing phase? \n\nIf everything should be done in 2 hours, can I train my model on my local system and import it to Kaggle notebook while submitting.",
      "votes": null
    },
    {
      "id": "2752888",
      "postDate": "04/15/2024 07:46:08",
      "content": "<p>You can train your model however you like. This can be done on Kaggle using a notebook, or externally using your own compute resources - no matter how long it takes for you to train. For submission, upload and import your trained model into your submission notebook and then run inference in under 2 hours.</p>",
      "rawMarkdown": "You can train your model however you like. This can be done on Kaggle using a notebook, or externally using your own compute resources - no matter how long it takes for you to train. For submission, upload and import your trained model into your submission notebook and then run inference in under 2 hours.",
      "votes": null
    },
    {
      "id": "2758535",
      "postDate": "04/18/2024 07:47:13",
      "content": "<p>Thank you for responding </p>",
      "rawMarkdown": "Thank you for responding",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2752888,
      "author_name": "stefankahl",
      "author_url": "",
      "post_date": "04/15/2024 07:46:08",
      "content": "<p>You can train your model however you like. This can be done on Kaggle using a notebook, or externally using your own compute resources - no matter how long it takes for you to train. For submission, upload and import your trained model into your submission notebook and then run inference in under 2 hours.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2758535,
          "author_name": "manideepbharadwaj10",
          "author_url": "",
          "post_date": "04/18/2024 07:47:13",
          "content": "<p>Thank you for responding </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2752708": "I have a small doubt regarding the submission, I do not know if someone has already answered this question, but I would appreciate any help. \n\nWhile submitting does the training time should also be under 2 hours or just the testing phase? \n\nIf everything should be done in 2 hours, can I train my model on my local system and import it to Kaggle notebook while submitting.",
    "2752888": "You can train your model however you like. This can be done on Kaggle using a notebook, or externally using your own compute resources - no matter how long it takes for you to train. For submission, upload and import your trained model into your submission notebook and then run inference in under 2 hours.",
    "2758535": "Thank you for responding"
  },
  "source": "meta"
}