{
  "id": 545871,
  "title": "is kaggle kernels enough?",
  "url": "/competitions/jane-street-real-time-market-data-forecasting/discussion/545871",
  "author_name": "",
  "post_date": "2024-11-12T15:08:50.616371200Z",
  "votes": 1,
  "comment_count": 5,
  "views": 0,
  "content": "<p>I don't have any computation power, and I want to get into this competition to improve my Time series skills, could I get good score and perform well without any offline training what do you guys think?</p>",
  "messages": [
    {
      "id": "3043657",
      "postDate": "11/12/2024 15:08:50",
      "content": "<p>I don't have any computation power, and I want to get into this competition to improve my Time series skills, could I get good score and perform well without any offline training what do you guys think?</p>",
      "rawMarkdown": "I don't have any computation power, and I want to get into this competition to improve my Time series skills, could I get good score and perform well without any offline training what do you guys think?",
      "votes": null
    },
    {
      "id": "3043697",
      "postDate": "11/12/2024 15:42:38",
      "content": "<p>Unfortunately not possible to train on full data - not enough RAM on kaggle. If you want   just to learn - you can train on a part of the data.</p>",
      "rawMarkdown": "Unfortunately not possible to train on full data - not enough RAM on kaggle. If you want   just to learn - you can train on a part of the data.",
      "votes": null
    },
    {
      "id": "3043917",
      "postDate": "11/12/2024 19:56:51",
      "content": "<p><a href=\"https://www.kaggle.com/tahaalshatiri\" target=\"_blank\">@tahaalshatiri</a> I suggest you could learn from other time series competitions if learning is your priority. The below competition is a good choice and can be trained on Kaggle- <br>\n<a href=\"https://www.kaggle.com/competitions/brist1d\" target=\"_blank\">https://www.kaggle.com/competitions/brist1d</a></p>",
      "rawMarkdown": "tahaalshatiri I suggest you could learn from other time series competitions if learning is your priority. The below competition is a good choice and can be trained on Kaggle- \nhttps://www.kaggle.com/competitions/brist1d",
      "votes": null
    },
    {
      "id": "3045709",
      "postDate": "11/14/2024 17:57:07",
      "content": "<p>It can be done, you just need to be smart with how you load your data and only keep what's necessary.</p>",
      "rawMarkdown": "It can be done, you just need to be smart with how you load your data and only keep what's necessary.",
      "votes": null
    },
    {
      "id": "3045859",
      "postDate": "11/14/2024 21:26:11",
      "content": "<p>Hi could you please give an example of how to do this please - I'm new to kaggle. Thank you !</p>",
      "rawMarkdown": "Hi could you please give an example of how to do this please - I'm new to kaggle. Thank you !",
      "votes": null
    },
    {
      "id": "3046198",
      "postDate": "11/15/2024 08:06:16",
      "content": "<p>I mentioned - \"not possible to train on FULL data\", but to train on some part of the data - yes we can</p>",
      "rawMarkdown": "I mentioned - \"not possible to train on FULL data\", but to train on some part of the data - yes we can",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3043697,
      "author_name": "eu1234",
      "author_url": "",
      "post_date": "11/12/2024 15:42:38",
      "content": "<p>Unfortunately not possible to train on full data - not enough RAM on kaggle. If you want   just to learn - you can train on a part of the data.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3045709,
          "author_name": "yanisfalaki",
          "author_url": "",
          "post_date": "11/14/2024 17:57:07",
          "content": "<p>It can be done, you just need to be smart with how you load your data and only keep what's necessary.</p>",
          "votes": null,
          "replies": [
            {
              "id": 3045859,
              "author_name": "denthe6",
              "author_url": "",
              "post_date": "11/14/2024 21:26:11",
              "content": "<p>Hi could you please give an example of how to do this please - I'm new to kaggle. Thank you !</p>",
              "votes": null,
              "replies": []
            },
            {
              "id": 3046198,
              "author_name": "eu1234",
              "author_url": "",
              "post_date": "11/15/2024 08:06:16",
              "content": "<p>I mentioned - \"not possible to train on FULL data\", but to train on some part of the data - yes we can</p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    },
    {
      "id": 3043917,
      "author_name": "ravi20076",
      "author_url": "",
      "post_date": "11/12/2024 19:56:51",
      "content": "<p><a href=\"https://www.kaggle.com/tahaalshatiri\" target=\"_blank\">@tahaalshatiri</a> I suggest you could learn from other time series competitions if learning is your priority. The below competition is a good choice and can be trained on Kaggle- <br>\n<a href=\"https://www.kaggle.com/competitions/brist1d\" target=\"_blank\">https://www.kaggle.com/competitions/brist1d</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3043657": "I don't have any computation power, and I want to get into this competition to improve my Time series skills, could I get good score and perform well without any offline training what do you guys think?",
    "3043697": "Unfortunately not possible to train on full data - not enough RAM on kaggle. If you want   just to learn - you can train on a part of the data.",
    "3043917": "tahaalshatiri I suggest you could learn from other time series competitions if learning is your priority. The below competition is a good choice and can be trained on Kaggle- \nhttps://www.kaggle.com/competitions/brist1d",
    "3045709": "It can be done, you just need to be smart with how you load your data and only keep what's necessary.",
    "3045859": "Hi could you please give an example of how to do this please - I'm new to kaggle. Thank you !",
    "3046198": "I mentioned - \"not possible to train on FULL data\", but to train on some part of the data - yes we can"
  },
  "source": "meta"
}