{ "id": 177255, "title": "How can I use l5kit without an internet connection?", "url": "/competitions/lyft-motion-prediction-autonomous-vehicles/discussion/177255", "author_name": "", "post_date": "2020-08-25T09:39:35.700316300Z", "votes": 4, "comment_count": 9, "views": 0, "content": "

The Data Description page says it can be used without doing anything, but it doesn't work. Maybe I'm mistaken, does anyone have any advice?

", "messages": [ { "id": "984793", "postDate": "08/25/2020 09:39:35", "content": "

The Data Description page says it can be used without doing anything, but it doesn't work. Maybe I'm mistaken, does anyone have any advice?

", "rawMarkdown": "The Data Description page says it can be used without doing anything, but it doesn't work. Maybe I'm mistaken, does anyone have any advice?", "votes": null }, { "id": "984862", "postDate": "08/25/2020 10:42:23", "content": "

https://www.kaggle.com/philculliton/kaggle-l5kit

", "rawMarkdown": "https://www.kaggle.com/philculliton/kaggle-l5kit", "votes": null }, { "id": "984973", "postDate": "08/25/2020 12:04:42", "content": "

@jpbremer
\nPlease explain in detail how to use that notebook.
\nI can't understand.

", "rawMarkdown": "jpbremer \nPlease explain in detail how to use that notebook.\nI can't understand.", "votes": null }, { "id": "984987", "postDate": "08/25/2020 12:22:28", "content": "

For some Reason I cant add Images to my comment right now…
\nBut this is basicly how you want to do it: @phalanx

\n
    \n
  1. Click Add Data in your kernel
    2. \n
  2. -> Kernel Output Files
    3. \n
  3. search for kaggle_l5kit
    4. \n
  4. Add to your kernel
    5. \n
  5. Then at the top of your kernel add:
    6. \n
\n
import os\n\n## this script transports l5kit and dependencies\nos.system('pip uninstall typing -y')\nos.system('pip install --ignore-installed --target=/kaggle/working l5kit')\n
", "rawMarkdown": "For some Reason I cant add Images to my comment right now...\nBut this is basicly how you want to do it: @phalanx \n1. Click Add Data in your kernel\n2. -> Kernel Output Files\n3. search for kaggle_l5kit\n4. Add to your kernel\n5. Then at the top of your kernel add: \n ```\nimport os\n\n## this script transports l5kit and dependencies\nos.system('pip uninstall typing -y')\nos.system('pip install --ignore-installed --target=/kaggle/working l5kit')\n```", "votes": null }, { "id": "985005", "postDate": "08/25/2020 12:38:11", "content": "

Great, thanks!
\nI try it.

", "rawMarkdown": "Great, thanks!\nI try it.", "votes": null }, { "id": "985212", "postDate": "08/25/2020 15:10:39", "content": "

You don't need 5. After 1-4, you can simply
\nimport l5kit

\n

For #5, you would need an internet connection and it would re-install everything that is already in the kaggle_l5kit utility script.

\n

Unfortunately, there is a bug in the notebook's input/output explorer (it shows too many files and the browser freezes after a few seconds). see this comment So, as long as they don't fix this, we cannot submit anything useful.

", "rawMarkdown": "You don't need 5. After 1-4, you can simply\n`import l5kit`\n\nFor \\#5, you would need an internet connection and it would re-install everything that is already in the `kaggle_l5kit` utility script.\n\nUnfortunately, there is a bug in the notebook's input/output explorer (it shows too many files and the browser freezes after a few seconds). [see this comment](https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125#984977) So, as long as they don't fix this, we cannot submit anything useful.", "votes": null }, { "id": "985315", "postDate": "08/25/2020 16:34:19", "content": "

Hi all. Thanks for your feedback. Yep - there's currently a bug with displaying the utility script - it'll freeze until the utility script finishes loading all of its files in the explorer. I'm working on finding a workaround until it gets fixed.

", "rawMarkdown": "Hi all. Thanks for your feedback. Yep - there's currently a bug with displaying the utility script - it'll freeze until the utility script finishes loading all of its files in the explorer. I'm working on finding a workaround until it gets fixed.", "votes": null }, { "id": "985661", "postDate": "08/25/2020 23:06:09", "content": "

@jpbremer
\nThank you, that's very helpful. Your icon reminds me of the ARC competition;) I wish you even more success!

", "rawMarkdown": "jpbremer \nThank you, that's very helpful. Your icon reminds me of the ARC competition;) I wish you even more success!", "votes": null }, { "id": "985674", "postDate": "08/25/2020 23:31:01", "content": "

@yujiariyasu
\nNow, I remember. I`ve just had a dejavu when I saw your profile pic :D
\nGood Luck

", "rawMarkdown": "yujiariyasu \nNow, I remember. I`ve just had a dejavu when I saw your profile pic :D \nGood Luck", "votes": null }, { "id": "1004135", "postDate": "09/09/2020 13:41:31", "content": "

Thanks! Been hitting my head against the wall for this ;-)

", "rawMarkdown": "Thanks! Been hitting my head against the wall for this ;-)", "votes": null } ], "comments": [ { "id": 984862, "author_name": "jpbremer", "author_url": "", "post_date": "08/25/2020 10:42:23", "content": "

https://www.kaggle.com/philculliton/kaggle-l5kit

", "votes": null, "replies": [ { "id": 984973, "author_name": "phalanx", "author_url": "", "post_date": "08/25/2020 12:04:42", "content": "

@jpbremer
\nPlease explain in detail how to use that notebook.
\nI can't understand.

", "votes": null, "replies": [] }, { "id": 984987, "author_name": "jpbremer", "author_url": "", "post_date": "08/25/2020 12:22:28", "content": "

For some Reason I cant add Images to my comment right now…
\nBut this is basicly how you want to do it: @phalanx

\n
    \n
  1. Click Add Data in your kernel
    2. \n
  2. -> Kernel Output Files
    3. \n
  3. search for kaggle_l5kit
    4. \n
  4. Add to your kernel
    5. \n
  5. Then at the top of your kernel add:
    6. \n
\n
import os\n\n## this script transports l5kit and dependencies\nos.system('pip uninstall typing -y')\nos.system('pip install --ignore-installed --target=/kaggle/working l5kit')\n
", "votes": null, "replies": [] }, { "id": 985005, "author_name": "phalanx", "author_url": "", "post_date": "08/25/2020 12:38:11", "content": "

Great, thanks!
\nI try it.

", "votes": null, "replies": [] }, { "id": 985212, "author_name": "pestipeti", "author_url": "", "post_date": "08/25/2020 15:10:39", "content": "

You don't need 5. After 1-4, you can simply
\nimport l5kit

\n

For #5, you would need an internet connection and it would re-install everything that is already in the kaggle_l5kit utility script.

\n

Unfortunately, there is a bug in the notebook's input/output explorer (it shows too many files and the browser freezes after a few seconds). see this comment So, as long as they don't fix this, we cannot submit anything useful.

", "votes": null, "replies": [] }, { "id": 985315, "author_name": "philculliton", "author_url": "", "post_date": "08/25/2020 16:34:19", "content": "

Hi all. Thanks for your feedback. Yep - there's currently a bug with displaying the utility script - it'll freeze until the utility script finishes loading all of its files in the explorer. I'm working on finding a workaround until it gets fixed.

", "votes": null, "replies": [] }, { "id": 985661, "author_name": "yujiariyasu", "author_url": "", "post_date": "08/25/2020 23:06:09", "content": "

@jpbremer
\nThank you, that's very helpful. Your icon reminds me of the ARC competition;) I wish you even more success!

", "votes": null, "replies": [] }, { "id": 985674, "author_name": "jpbremer", "author_url": "", "post_date": "08/25/2020 23:31:01", "content": "

@yujiariyasu
\nNow, I remember. I`ve just had a dejavu when I saw your profile pic :D
\nGood Luck

", "votes": null, "replies": [] } ] }, { "id": 1004135, "author_name": "jankins", "author_url": "", "post_date": "09/09/2020 13:41:31", "content": "

Thanks! Been hitting my head against the wall for this ;-)

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "984793": "The Data Description page says it can be used without doing anything, but it doesn't work. Maybe I'm mistaken, does anyone have any advice?", "984862": "https://www.kaggle.com/philculliton/kaggle-l5kit", "984973": "jpbremer \nPlease explain in detail how to use that notebook.\nI can't understand.", "984987": "For some Reason I cant add Images to my comment right now...\nBut this is basicly how you want to do it: @phalanx \n1. Click Add Data in your kernel\n2. -> Kernel Output Files\n3. search for kaggle_l5kit\n4. Add to your kernel\n5. Then at the top of your kernel add: \n ```\nimport os\n\n## this script transports l5kit and dependencies\nos.system('pip uninstall typing -y')\nos.system('pip install --ignore-installed --target=/kaggle/working l5kit')\n```", "985005": "Great, thanks!\nI try it.", "985212": "You don't need 5. After 1-4, you can simply\n`import l5kit`\n\nFor \\#5, you would need an internet connection and it would re-install everything that is already in the `kaggle_l5kit` utility script.\n\nUnfortunately, there is a bug in the notebook's input/output explorer (it shows too many files and the browser freezes after a few seconds). [see this comment](https://www.kaggle.com/c/lyft-motion-prediction-autonomous-vehicles/discussion/177125#984977) So, as long as they don't fix this, we cannot submit anything useful.", "985315": "Hi all. Thanks for your feedback. Yep - there's currently a bug with displaying the utility script - it'll freeze until the utility script finishes loading all of its files in the explorer. I'm working on finding a workaround until it gets fixed.", "985661": "jpbremer \nThank you, that's very helpful. Your icon reminds me of the ARC competition;) I wish you even more success!", "985674": "yujiariyasu \nNow, I remember. I`ve just had a dejavu when I saw your profile pic :D \nGood Luck", "1004135": "Thanks! Been hitting my head against the wall for this ;-)" }, "source": "meta" }