{
  "id": 453062,
  "title": "Want to install Lightgbm == 4.0.0",
  "url": "/competitions/child-mind-institute-detect-sleep-states/discussion/453062",
  "author_name": "",
  "post_date": "2023-11-04T17:29:34.817922200Z",
  "votes": 4,
  "comment_count": 8,
  "views": 0,
  "content": "<p>Hi.<br>\nI want to install Lightgbm == 4.0.0 to kaggle notebook but you can't pip install because of internet off competition.My kaggle notebook version is Lightgbm == 3.2.2 .<br>\nIs there any good idea to tackle this problem?</p>",
  "messages": [
    {
      "id": "2512619",
      "postDate": "11/04/2023 17:29:34",
      "content": "<p>Hi.<br>\nI want to install Lightgbm == 4.0.0 to kaggle notebook but you can't pip install because of internet off competition.My kaggle notebook version is Lightgbm == 3.2.2 .<br>\nIs there any good idea to tackle this problem?</p>",
      "rawMarkdown": "Hi.\nI want to install Lightgbm == 4.0.0 to kaggle notebook but you can't pip install because of internet off competition.My kaggle notebook version is Lightgbm == 3.2.2 .\nIs there any good idea to tackle this problem?",
      "votes": null
    },
    {
      "id": "2512625",
      "postDate": "11/04/2023 17:36:42",
      "content": "<p>Use the dataset <a href=\"https://www.kaggle.com/datasets/mingccc/lightgbm410\" target=\"_blank\">https://www.kaggle.com/datasets/mingccc/lightgbm410</a> for the latest version of light GBM <a href=\"https://www.kaggle.com/kazumamori\" target=\"_blank\">@kazumamori</a> </p>",
      "rawMarkdown": "Use the dataset https://www.kaggle.com/datasets/mingccc/lightgbm410 for the latest version of light GBM @kazumamori",
      "votes": null
    },
    {
      "id": "2513450",
      "postDate": "11/05/2023 13:07:46",
      "content": "<p>Hi Ravi!<br>\nThank you for sharing!</p>",
      "rawMarkdown": "Hi Ravi!\nThank you for sharing!",
      "votes": null
    },
    {
      "id": "2518814",
      "postDate": "11/09/2023 15:05:57",
      "content": "<p><a href=\"https://www.kaggle.com/kazumamori\" target=\"_blank\">@kazumamori</a> you can also use another notebook output as a data source. In this another notebook (it can be with the internet on) you could just write <code>!pip download lightgbm==4.0.0</code> and receive all the necessary dependencies for LightGBM installation.</p>",
      "rawMarkdown": "kazumamori you can also use another notebook output as a data source. In this another notebook (it can be with the internet on) you could just write `!pip download lightgbm==4.0.0` and receive all the necessary dependencies for LightGBM installation.",
      "votes": null
    },
    {
      "id": "2518959",
      "postDate": "11/09/2023 16:19:09",
      "content": "<p>Hi, Alexander<br>\nThank you so much for sharing your tip.I didn't know that.<br>\nI did <code>!pip download lightgbm==4.0.0</code> and get three files.<br>\nDo you know the differnce of these?</p>\n<p>scipy-1.11.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</p>\n<p>numpy-1.26.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</p>\n<p>lightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl<br>\n<a href=\"https://www.kaggle.com/alexryzhkov\" target=\"_blank\">@alexryzhkov</a> </p>",
      "rawMarkdown": "Hi, Alexander\nThank you so much for sharing your tip.I didn't know that.\nI did `!pip download lightgbm==4.0.0` and get three files.\nDo you know the differnce of these?\n\nscipy-1.11.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl\n\nnumpy-1.26.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl\n\nlightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl\n@alexryzhkov",
      "votes": null
    },
    {
      "id": "2518980",
      "postDate": "11/09/2023 16:30:41",
      "content": "<p><a href=\"https://www.kaggle.com/kazumamori\" target=\"_blank\">@kazumamori</a> Yes, I know - here is lightgbm with its dependencies. As I think that numpy and scipy are already installed in Kaggle env you can simply run <code>pip install lightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl</code> in your submission kernel, changing the path to whl to the right one</p>",
      "rawMarkdown": "kazumamori Yes, I know - here is lightgbm with its dependencies. As I think that numpy and scipy are already installed in Kaggle env you can simply run `pip install lightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl` in your submission kernel, changing the path to whl to the right one",
      "votes": null
    },
    {
      "id": "2520667",
      "postDate": "11/11/2023 03:53:16",
      "content": "<p>Got it! Thank you!</p>",
      "rawMarkdown": "Got it! Thank you!",
      "votes": null
    },
    {
      "id": "2531359",
      "postDate": "11/20/2023 06:34:06",
      "content": "<p>Not sure whether you still need help, but you can use GitHub Repositories:<br>\nIf the LightGBM version you need is available in a GitHub repository, you can clone the repository into a Kaggle dataset and then add this dataset to your notebook.</p>",
      "rawMarkdown": "Not sure whether you still need help, but you can use GitHub Repositories:\nIf the LightGBM version you need is available in a GitHub repository, you can clone the repository into a Kaggle dataset and then add this dataset to your notebook.",
      "votes": null
    },
    {
      "id": "2531652",
      "postDate": "11/20/2023 11:42:59",
      "content": "<p>Hi! Z Chen3<br>\nThanks for sharing your knowledge.  I didn't know that. I will try next time.</p>",
      "rawMarkdown": "Hi! Z Chen3\nThanks for sharing your knowledge.  I didn't know that. I will try next time.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2512625,
      "author_name": "ravi20076",
      "author_url": "",
      "post_date": "11/04/2023 17:36:42",
      "content": "<p>Use the dataset <a href=\"https://www.kaggle.com/datasets/mingccc/lightgbm410\" target=\"_blank\">https://www.kaggle.com/datasets/mingccc/lightgbm410</a> for the latest version of light GBM <a href=\"https://www.kaggle.com/kazumamori\" target=\"_blank\">@kazumamori</a> </p>",
      "votes": null,
      "replies": [
        {
          "id": 2513450,
          "author_name": "kazumamori",
          "author_url": "",
          "post_date": "11/05/2023 13:07:46",
          "content": "<p>Hi Ravi!<br>\nThank you for sharing!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 2518814,
      "author_name": "alexryzhkov",
      "author_url": "",
      "post_date": "11/09/2023 15:05:57",
      "content": "<p><a href=\"https://www.kaggle.com/kazumamori\" target=\"_blank\">@kazumamori</a> you can also use another notebook output as a data source. In this another notebook (it can be with the internet on) you could just write <code>!pip download lightgbm==4.0.0</code> and receive all the necessary dependencies for LightGBM installation.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2518959,
          "author_name": "kazumamori",
          "author_url": "",
          "post_date": "11/09/2023 16:19:09",
          "content": "<p>Hi, Alexander<br>\nThank you so much for sharing your tip.I didn't know that.<br>\nI did <code>!pip download lightgbm==4.0.0</code> and get three files.<br>\nDo you know the differnce of these?</p>\n<p>scipy-1.11.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</p>\n<p>numpy-1.26.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl</p>\n<p>lightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl<br>\n<a href=\"https://www.kaggle.com/alexryzhkov\" target=\"_blank\">@alexryzhkov</a> </p>",
          "votes": null,
          "replies": [
            {
              "id": 2518980,
              "author_name": "alexryzhkov",
              "author_url": "",
              "post_date": "11/09/2023 16:30:41",
              "content": "<p><a href=\"https://www.kaggle.com/kazumamori\" target=\"_blank\">@kazumamori</a> Yes, I know - here is lightgbm with its dependencies. As I think that numpy and scipy are already installed in Kaggle env you can simply run <code>pip install lightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl</code> in your submission kernel, changing the path to whl to the right one</p>",
              "votes": null,
              "replies": [
                {
                  "id": 2520667,
                  "author_name": "kazumamori",
                  "author_url": "",
                  "post_date": "11/11/2023 03:53:16",
                  "content": "<p>Got it! Thank you!</p>",
                  "votes": null,
                  "replies": []
                }
              ]
            }
          ]
        }
      ]
    },
    {
      "id": 2531359,
      "author_name": "ziyuechen3",
      "author_url": "",
      "post_date": "11/20/2023 06:34:06",
      "content": "<p>Not sure whether you still need help, but you can use GitHub Repositories:<br>\nIf the LightGBM version you need is available in a GitHub repository, you can clone the repository into a Kaggle dataset and then add this dataset to your notebook.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2531652,
          "author_name": "kazumamori",
          "author_url": "",
          "post_date": "11/20/2023 11:42:59",
          "content": "<p>Hi! Z Chen3<br>\nThanks for sharing your knowledge.  I didn't know that. I will try next time.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2512619": "Hi.\nI want to install Lightgbm == 4.0.0 to kaggle notebook but you can't pip install because of internet off competition.My kaggle notebook version is Lightgbm == 3.2.2 .\nIs there any good idea to tackle this problem?",
    "2512625": "Use the dataset https://www.kaggle.com/datasets/mingccc/lightgbm410 for the latest version of light GBM @kazumamori",
    "2513450": "Hi Ravi!\nThank you for sharing!",
    "2518814": "kazumamori you can also use another notebook output as a data source. In this another notebook (it can be with the internet on) you could just write `!pip download lightgbm==4.0.0` and receive all the necessary dependencies for LightGBM installation.",
    "2518959": "Hi, Alexander\nThank you so much for sharing your tip.I didn't know that.\nI did `!pip download lightgbm==4.0.0` and get three files.\nDo you know the differnce of these?\n\nscipy-1.11.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl\n\nnumpy-1.26.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl\n\nlightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl\n@alexryzhkov",
    "2518980": "kazumamori Yes, I know - here is lightgbm with its dependencies. As I think that numpy and scipy are already installed in Kaggle env you can simply run `pip install lightgbm-4.0.0-py3-none-manylinux_2_28_x86_64.whl` in your submission kernel, changing the path to whl to the right one",
    "2520667": "Got it! Thank you!",
    "2531359": "Not sure whether you still need help, but you can use GitHub Repositories:\nIf the LightGBM version you need is available in a GitHub repository, you can clone the repository into a Kaggle dataset and then add this dataset to your notebook.",
    "2531652": "Hi! Z Chen3\nThanks for sharing your knowledge.  I didn't know that. I will try next time."
  },
  "source": "meta"
}