{
  "id": 408054,
  "title": "Do we allow to use GPU at all?",
  "url": "/competitions/predict-student-performance-from-game-play/discussion/408054",
  "author_name": "",
  "post_date": "2023-05-09T09:02:07.065834500Z",
  "votes": null,
  "comment_count": 4,
  "views": 0,
  "content": "<p><code>Note: this competition is aimed at producing models that are small and lightweight. We have introduced compute constraints to match - your VMs will have only 2 CPUs, 8GB of RAM, and no GPU available. You will still have a maximum of 9 hours to complete the task, but between the constraints and the efficiency prize there will be some interesting sub-problems to solve. Good luck!</code></p>\n<p>I read this from the part where it says 'efficiency prize'. Using gpu is prohibited or is it only for the people who aim for this 'efficiency prize'?</p>",
  "messages": [
    {
      "id": "2251326",
      "postDate": "05/09/2023 09:02:07",
      "content": "<p><code>Note: this competition is aimed at producing models that are small and lightweight. We have introduced compute constraints to match - your VMs will have only 2 CPUs, 8GB of RAM, and no GPU available. You will still have a maximum of 9 hours to complete the task, but between the constraints and the efficiency prize there will be some interesting sub-problems to solve. Good luck!</code></p>\n<p>I read this from the part where it says 'efficiency prize'. Using gpu is prohibited or is it only for the people who aim for this 'efficiency prize'?</p>",
      "rawMarkdown": "`Note: this competition is aimed at producing models that are small and lightweight. We have introduced compute constraints to match - your VMs will have only 2 CPUs, 8GB of RAM, and no GPU available. You will still have a maximum of 9 hours to complete the task, but between the constraints and the efficiency prize there will be some interesting sub-problems to solve. Good luck!`\n\nI read this from the part where it says 'efficiency prize'. Using gpu is prohibited or is it only for the people who aim for this 'efficiency prize'?",
      "votes": null
    },
    {
      "id": "2251331",
      "postDate": "05/09/2023 09:05:04",
      "content": "<p>I think we can't submit if GPUs are enabled, so I think GPUs are not allowed in this competition in general, not only with the Efficiency Prize.</p>",
      "rawMarkdown": "I think we can't submit if GPUs are enabled, so I think GPUs are not allowed in this competition in general, not only with the Efficiency Prize.",
      "votes": null
    },
    {
      "id": "2251421",
      "postDate": "05/09/2023 10:21:28",
      "content": "<p>GPU is not allowed in inference notebook: <em>Inference VMs will have only 2 CPUs, 8GB of RAM, and no GPU available</em>, but you can use GPU to train your models.</p>",
      "rawMarkdown": "GPU is not allowed in inference notebook: *Inference VMs will have only 2 CPUs, 8GB of RAM, and no GPU available*, but you can use GPU to train your models.",
      "votes": null
    },
    {
      "id": "2253129",
      "postDate": "05/10/2023 00:29:43",
      "content": "<p>So in order to do the inference, I have to save the trained weights and load them with non gpu environment?</p>",
      "rawMarkdown": "So in order to do the inference, I have to save the trained weights and load them with non gpu environment?",
      "votes": null
    },
    {
      "id": "2253359",
      "postDate": "05/10/2023 05:49:55",
      "content": "<p>Exactly! you can refer to this example by <a href=\"https://www.kaggle.com/takanashihumbert\" target=\"_blank\">@takanashihumbert</a> </p>\n<ul>\n<li>train: <a href=\"https://www.kaggle.com/code/takanashihumbert/magic-bingo-train-part-lb-0-687\" target=\"_blank\">https://www.kaggle.com/code/takanashihumbert/magic-bingo-train-part-lb-0-687</a> </li>\n<li>inference: <a href=\"https://www.kaggle.com/code/takanashihumbert/magic-bingo-inference-part-lb-0-687/notebook\" target=\"_blank\">https://www.kaggle.com/code/takanashihumbert/magic-bingo-inference-part-lb-0-687/notebook</a></li>\n</ul>",
      "rawMarkdown": "Exactly! you can refer to this example by @takanashihumbert \n- train: https://www.kaggle.com/code/takanashihumbert/magic-bingo-train-part-lb-0-687 \n- inference: https://www.kaggle.com/code/takanashihumbert/magic-bingo-inference-part-lb-0-687/notebook",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2251331,
      "author_name": "shinomoriaoshi",
      "author_url": "",
      "post_date": "05/09/2023 09:05:04",
      "content": "<p>I think we can't submit if GPUs are enabled, so I think GPUs are not allowed in this competition in general, not only with the Efficiency Prize.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2251421,
      "author_name": "steubk",
      "author_url": "",
      "post_date": "05/09/2023 10:21:28",
      "content": "<p>GPU is not allowed in inference notebook: <em>Inference VMs will have only 2 CPUs, 8GB of RAM, and no GPU available</em>, but you can use GPU to train your models.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2253129,
          "author_name": "gilgarad",
          "author_url": "",
          "post_date": "05/10/2023 00:29:43",
          "content": "<p>So in order to do the inference, I have to save the trained weights and load them with non gpu environment?</p>",
          "votes": null,
          "replies": [
            {
              "id": 2253359,
              "author_name": "steubk",
              "author_url": "",
              "post_date": "05/10/2023 05:49:55",
              "content": "<p>Exactly! you can refer to this example by <a href=\"https://www.kaggle.com/takanashihumbert\" target=\"_blank\">@takanashihumbert</a> </p>\n<ul>\n<li>train: <a href=\"https://www.kaggle.com/code/takanashihumbert/magic-bingo-train-part-lb-0-687\" target=\"_blank\">https://www.kaggle.com/code/takanashihumbert/magic-bingo-train-part-lb-0-687</a> </li>\n<li>inference: <a href=\"https://www.kaggle.com/code/takanashihumbert/magic-bingo-inference-part-lb-0-687/notebook\" target=\"_blank\">https://www.kaggle.com/code/takanashihumbert/magic-bingo-inference-part-lb-0-687/notebook</a></li>\n</ul>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2251326": "`Note: this competition is aimed at producing models that are small and lightweight. We have introduced compute constraints to match - your VMs will have only 2 CPUs, 8GB of RAM, and no GPU available. You will still have a maximum of 9 hours to complete the task, but between the constraints and the efficiency prize there will be some interesting sub-problems to solve. Good luck!`\n\nI read this from the part where it says 'efficiency prize'. Using gpu is prohibited or is it only for the people who aim for this 'efficiency prize'?",
    "2251331": "I think we can't submit if GPUs are enabled, so I think GPUs are not allowed in this competition in general, not only with the Efficiency Prize.",
    "2251421": "GPU is not allowed in inference notebook: *Inference VMs will have only 2 CPUs, 8GB of RAM, and no GPU available*, but you can use GPU to train your models.",
    "2253129": "So in order to do the inference, I have to save the trained weights and load them with non gpu environment?",
    "2253359": "Exactly! you can refer to this example by @takanashihumbert \n- train: https://www.kaggle.com/code/takanashihumbert/magic-bingo-train-part-lb-0-687 \n- inference: https://www.kaggle.com/code/takanashihumbert/magic-bingo-inference-part-lb-0-687/notebook"
  },
  "source": "meta"
}