{
  "id": 590644,
  "title": "Is this data trainable in the kaggle notebook? Or should I use other resources?",
  "url": "/competitions/brain-to-text-25/discussion/590644",
  "author_name": "",
  "post_date": "2025-07-22T02:50:41.962643100Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I am currently using runpod server to load the data and go through the baseline RNN codes. </p>\n<p>Because the RAM limit of kaggle is 30GB to my knowledge, will it be possible to train the model in the kaggle notebook? or should I continue using the runpod server.</p>\n<p>Thank you.</p>",
  "messages": [
    {
      "id": "3252093",
      "postDate": "07/22/2025 02:50:41",
      "content": "<p>I am currently using runpod server to load the data and go through the baseline RNN codes. </p>\n<p>Because the RAM limit of kaggle is 30GB to my knowledge, will it be possible to train the model in the kaggle notebook? or should I continue using the runpod server.</p>\n<p>Thank you.</p>",
      "rawMarkdown": "I am currently using runpod server to load the data and go through the baseline RNN codes. \n\nBecause the RAM limit of kaggle is 30GB to my knowledge, will it be possible to train the model in the kaggle notebook? or should I continue using the runpod server.\n\nThank you.",
      "votes": null
    },
    {
      "id": "3252596",
      "postDate": "07/23/2025 04:36:25",
      "content": "<p>Hi, I personally do not know much about Kaggle's server compute. We run all of this code (training, LM, eval) on a dedicated local Ubuntu machine. I realize that the current LM has hefty resource requirements, but maybe through this competition someone will streamline a better solution 😁</p>",
      "rawMarkdown": "Hi, I personally do not know much about Kaggle's server compute. We run all of this code (training, LM, eval) on a dedicated local Ubuntu machine. I realize that the current LM has hefty resource requirements, but maybe through this competition someone will streamline a better solution 😁",
      "votes": null
    },
    {
      "id": "3253129",
      "postDate": "07/24/2025 04:20:34",
      "content": "<p>Thank you Nick</p>",
      "rawMarkdown": "Thank you Nick",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3252596,
      "author_name": "notnickc",
      "author_url": "",
      "post_date": "07/23/2025 04:36:25",
      "content": "<p>Hi, I personally do not know much about Kaggle's server compute. We run all of this code (training, LM, eval) on a dedicated local Ubuntu machine. I realize that the current LM has hefty resource requirements, but maybe through this competition someone will streamline a better solution 😁</p>",
      "votes": null,
      "replies": [
        {
          "id": 3253129,
          "author_name": "kendontcare11",
          "author_url": "",
          "post_date": "07/24/2025 04:20:34",
          "content": "<p>Thank you Nick</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3252093": "I am currently using runpod server to load the data and go through the baseline RNN codes. \n\nBecause the RAM limit of kaggle is 30GB to my knowledge, will it be possible to train the model in the kaggle notebook? or should I continue using the runpod server.\n\nThank you.",
    "3252596": "Hi, I personally do not know much about Kaggle's server compute. We run all of this code (training, LM, eval) on a dedicated local Ubuntu machine. I realize that the current LM has hefty resource requirements, but maybe through this competition someone will streamline a better solution 😁",
    "3253129": "Thank you Nick"
  },
  "source": "meta"
}