{
  "id": 398511,
  "title": "Faster Computation using GPU",
  "url": "/competitions/early-detection-of-3d-printing-issues/discussion/398511",
  "author_name": "",
  "post_date": "2023-03-30T11:58:29.439174700Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I am stuck at the issue where the model training takes a lot of time using resources like Kaggle notebook GPUs and Colab GPUs, has anyone been able to train faster using these resources only?</p>",
  "messages": [
    {
      "id": "2202917",
      "postDate": "03/30/2023 11:58:29",
      "content": "<p>I am stuck at the issue where the model training takes a lot of time using resources like Kaggle notebook GPUs and Colab GPUs, has anyone been able to train faster using these resources only?</p>",
      "rawMarkdown": "I am stuck at the issue where the model training takes a lot of time using resources like Kaggle notebook GPUs and Colab GPUs, has anyone been able to train faster using these resources only?",
      "votes": null
    },
    {
      "id": "2202974",
      "postDate": "03/30/2023 12:38:33",
      "content": "<p>I'm afraid the GPUs Kaggle or Colab give you are not powerful enough to run training quickly. I know many students either have access to a good GPU computer in their schools, or have enough cloud computing credit so that they can spin up a GPU VM in the cloud.</p>",
      "rawMarkdown": "I'm afraid the GPUs Kaggle or Colab give you are not powerful enough to run training quickly. I know many students either have access to a good GPU computer in their schools, or have enough cloud computing credit so that they can spin up a GPU VM in the cloud.",
      "votes": null
    },
    {
      "id": "2203053",
      "postDate": "03/30/2023 13:32:13",
      "content": "<p>You can commit the code and leave it running in the background. Try using 16 bit precision </p>",
      "rawMarkdown": "You can commit the code and leave it running in the background. Try using 16 bit precision",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2202974,
      "author_name": "kennethjiangobico",
      "author_url": "",
      "post_date": "03/30/2023 12:38:33",
      "content": "<p>I'm afraid the GPUs Kaggle or Colab give you are not powerful enough to run training quickly. I know many students either have access to a good GPU computer in their schools, or have enough cloud computing credit so that they can spin up a GPU VM in the cloud.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2203053,
      "author_name": "tanmayambadkar",
      "author_url": "",
      "post_date": "03/30/2023 13:32:13",
      "content": "<p>You can commit the code and leave it running in the background. Try using 16 bit precision </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2202917": "I am stuck at the issue where the model training takes a lot of time using resources like Kaggle notebook GPUs and Colab GPUs, has anyone been able to train faster using these resources only?",
    "2202974": "I'm afraid the GPUs Kaggle or Colab give you are not powerful enough to run training quickly. I know many students either have access to a good GPU computer in their schools, or have enough cloud computing credit so that they can spin up a GPU VM in the cloud.",
    "2203053": "You can commit the code and leave it running in the background. Try using 16 bit precision"
  },
  "source": "meta"
}