Hi all, I want to use torch geometric but it is not included in the env, how can I install it or submit a version which allow me to use this package? Because it says the internet is off when you submit the file.
", "messages": [ { "id": "2692053", "postDate": "03/11/2024 16:35:54", "content": "Hi all, I want to use torch geometric but it is not included in the env, how can I install it or submit a version which allow me to use this package? Because it says the internet is off when you submit the file.
", "rawMarkdown": "Hi all, I want to use torch geometric but it is not included in the env, how can I install it or submit a version which allow me to use this package? Because it says the internet is off when you submit the file.", "votes": null }, { "id": "2692115", "postDate": "03/11/2024 17:19:22", "content": "one example how to do it is to download and save the pip-wheel of torch geometric and create a dataset with it. Then in the submission kernel put this dataset as input and pip install from that wheel
", "rawMarkdown": "one example how to do it is to download and save the pip-wheel of torch geometric and create a dataset with it. Then in the submission kernel put this dataset as input and pip install from that wheel", "votes": null }, { "id": "2694303", "postDate": "03/13/2024 01:37:23", "content": "How to do that is here by CHRIS DEOTTE .
\nHow to Install Without Internet
The most neat way is to install packages into a supplemental notebook, mark this notebook as Utility Script, then attach it as a data source to any competition kernel. The detailed tutorial on this and other methods is available in pip-install-no-internet notebook. Hope it would be of a help, @enningyang.
", "rawMarkdown": "The most neat way is to install packages into a supplemental notebook, mark this notebook as Utility Script, then attach it as a data source to any competition kernel. The detailed tutorial on this and other methods is available in [pip-install-no-internet](https://www.kaggle.com/code/kononenko/pip-install-no-internet) notebook. Hope it would be of a help, @enningyang.", "votes": null } ], "comments": [ { "id": 2692115, "author_name": "christofhenkel", "author_url": "", "post_date": "03/11/2024 17:19:22", "content": "one example how to do it is to download and save the pip-wheel of torch geometric and create a dataset with it. Then in the submission kernel put this dataset as input and pip install from that wheel
", "votes": null, "replies": [] }, { "id": 2694303, "author_name": "clearwaterkzk", "author_url": "", "post_date": "03/13/2024 01:37:23", "content": "How to do that is here by CHRIS DEOTTE .
\nHow to Install Without Internet
The most neat way is to install packages into a supplemental notebook, mark this notebook as Utility Script, then attach it as a data source to any competition kernel. The detailed tutorial on this and other methods is available in pip-install-no-internet notebook. Hope it would be of a help, @enningyang.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2692053": "Hi all, I want to use torch geometric but it is not included in the env, how can I install it or submit a version which allow me to use this package? Because it says the internet is off when you submit the file.", "2692115": "one example how to do it is to download and save the pip-wheel of torch geometric and create a dataset with it. Then in the submission kernel put this dataset as input and pip install from that wheel", "2694303": "How to do that is here by CHRIS DEOTTE .\n[How to Install Without Internet](https://www.kaggle.com/c/severstal-steel-defect-detection/discussion/113195)", "2699363": "The most neat way is to install packages into a supplemental notebook, mark this notebook as Utility Script, then attach it as a data source to any competition kernel. The detailed tutorial on this and other methods is available in [pip-install-no-internet](https://www.kaggle.com/code/kononenko/pip-install-no-internet) notebook. Hope it would be of a help, @enningyang." }, "source": "meta" }