{ "id": 396979, "title": "How To Get 32GB RAM - Avoid Memory Error!", "url": "/competitions/predict-student-performance-from-game-play/discussion/396979", "author_name": "Chris Deotte", "post_date": "2023-03-23T15:45:07.538000", "votes": 111, "comment_count": 32, "views": 0, "content": "

The most common discussion post in this competition is: \"Your notebook tried to allocate more memory than is available. It has restarted.\". The best way to avoid memory error is to use two notebooks in this competition.

\n\n

It is tricky to get 32GB RAM in this competition because if we create a new Kaggle notebook from INSIDE this competition, then Kaggle reduces notebook RAM to 8GB. The trick is to create a new Kaggle notebook OUTSIDE this competition from Kaggke's notebook page https://www.kaggle.com/code. Then attach the output from my notebook named https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram. This will give your notebook the competition data in parquet format and your notebook will have 32GB RAM ! (More info about two notebooks here and \"How to use one notebook\" here)

\n

Six Steps To Get 32GB RAM

\n

\"\"
\n\"\"
\n\"\"
\n\"\"
\n\"\"

", "messages": [ { "id": 2193933, "postDate": "2023-03-23T15:45:07.540Z", "content": "

The most common discussion post in this competition is: \"Your notebook tried to allocate more memory than is available. It has restarted.\". The best way to avoid memory error is to use two notebooks in this competition.

\n\n

It is tricky to get 32GB RAM in this competition because if we create a new Kaggle notebook from INSIDE this competition, then Kaggle reduces notebook RAM to 8GB. The trick is to create a new Kaggle notebook OUTSIDE this competition from Kaggke's notebook page https://www.kaggle.com/code. Then attach the output from my notebook named https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram. This will give your notebook the competition data in parquet format and your notebook will have 32GB RAM ! (More info about two notebooks here and \"How to use one notebook\" here)

\n

Six Steps To Get 32GB RAM

\n

\"\"
\n\"\"
\n\"\"
\n\"\"
\n\"\"

", "rawMarkdown": "The most common discussion post in this competition is: \"Your notebook tried to allocate more memory than is available. It has restarted.\". The best way to avoid memory error is to **use two notebooks** in this competition.\n* Train and save models with 32GB RAM notebook (CPU or GPU)\n* Infer from load models with 8GB RAM notebook (CPU)\n\nIt is tricky to get 32GB RAM in this competition because if we create a new Kaggle notebook from **INSIDE** this competition, then Kaggle reduces notebook RAM to 8GB. The trick is to create a new Kaggle notebook **OUTSIDE** this competition from Kaggke's notebook page `https://www.kaggle.com/code`. Then attach the output from my notebook named `https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram`. This will give your notebook the competition data in parquet format and your notebook will have 32GB RAM ! (More info about **two** notebooks [here][1] and \"How to use **one** notebook\" [here][2])\n\n# Six Steps To Get 32GB RAM\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m1.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m2.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m3.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m4.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m5.png)\n\n[1]: https://www.kaggle.com/competitions/predict-student-performance-from-game-play/discussion/386218\n[2]: https://www.kaggle.com/code/cdeotte/xgboost-baseline-0-680", "votes": 110 }, { "id": 2284504, "postDate": "2023-06-02T05:42:40.213Z", "content": "

Thanks for sharing @cdeotte . I got a problem that it turned back to 8Gb after I added the competition dataset. Is there anything I did wrong?

", "rawMarkdown": "Thanks for sharing @cdeotte . I got a problem that it turned back to 8Gb after I added the competition dataset. Is there anything I did wrong?", "votes": 1, "replies": [ { "id": 2285218, "postDate": "2023-06-02T14:33:43.553Z", "content": "

Yes @yemenng8888 . You should not add the competition dataset. You should add the output of my notebook which contains the competition data.

", "rawMarkdown": "Yes @yemenng8888 . You should not add the competition dataset. You should add the output of my notebook which contains the competition data.", "votes": 2 } ] }, { "id": 2267246, "postDate": "2023-05-20T18:10:33.910Z", "content": "

very helpful, much appreciated.

", "rawMarkdown": "very helpful, much appreciated.", "votes": 1 }, { "id": 2243782, "postDate": "2023-05-03T07:04:03.633Z", "content": "

@retroflake

", "rawMarkdown": "@retroflake ", "votes": 1 }, { "id": 2218272, "postDate": "2023-04-11T14:38:19.160Z", "content": "

Thanks for sharing, very helpful @cdeotte

", "rawMarkdown": "Thanks for sharing, very helpful @cdeotte ", "votes": 1 }, { "id": 2198297, "postDate": "2023-03-26T21:56:12.753Z", "content": "

This has been a great help. Thanks @cdeotte. Is there a decent workaround for sharing a notebook to the competition (not as a submission)? Or is sharing as a Topic in Discussion the workaround?

", "rawMarkdown": "This has been a great help. Thanks @cdeotte. Is there a decent workaround for sharing a notebook to the competition (not as a submission)? Or is sharing as a Topic in Discussion the workaround?", "votes": 1, "replies": [ { "id": 2202490, "postDate": "2023-03-30T05:16:43.667Z", "content": "

Every notebook is first a commit notebook and not a submit notebook. So you can create a notebook and click \"commit\". That will save it and then you can share it. You do not need to submit it (to the leaderboard) to share it.

", "rawMarkdown": "Every notebook is first a commit notebook and not a submit notebook. So you can create a notebook and click \"commit\". That will save it and then you can share it. You do not need to submit it (to the leaderboard) to share it.", "votes": 1 } ] }, { "id": 2197050, "postDate": "2023-03-25T20:36:18.377Z", "content": "

Thank you so much for the helpful information on how to get 32GB of RAM for the competition. Your advice on using two different notebooks and attaching the output from your notebook has been invaluable. I really appreciate your willingness to share your knowledge and help others in the community. Thanks again!

", "rawMarkdown": "Thank you so much for the helpful information on how to get 32GB of RAM for the competition. Your advice on using two different notebooks and attaching the output from your notebook has been invaluable. I really appreciate your willingness to share your knowledge and help others in the community. Thanks again!", "votes": 1 }, { "id": 2196654, "postDate": "2023-03-25T15:09:09.903Z", "content": "

Thanks, will try that out in my next project

", "rawMarkdown": "Thanks, will try that out in my next project", "votes": 1 }, { "id": 2193939, "postDate": "2023-03-23T15:51:27.300Z", "content": "

Valuable information. Thanks for Sharing @cdeotte

", "rawMarkdown": "Valuable information. Thanks for Sharing @cdeotte ", "votes": 1, "replies": [ { "id": 2195756, "postDate": "2023-03-24T21:56:03.860Z", "content": "

Thanks Pritom!

", "rawMarkdown": "Thanks Pritom!" } ] }, { "id": 2235020, "postDate": "2023-04-25T16:27:37.733Z", "content": "

Hi ， sorry that i am new to kaggle and I cant see ur picture for \"Six Steps To Get 32GB RAM\", but i tried to copy the code \"https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram\" , but seems the CPU (RAM) is still 13 GB max. do i misundestand anything in ur notebook ? Thanks advance.

", "rawMarkdown": "Hi ， sorry that i am new to kaggle and I cant see ur picture for \"Six Steps To Get 32GB RAM\", but i tried to copy the code \"https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram\" , but seems the CPU (RAM) is still 13 GB max. do i misundestand anything in ur notebook ? Thanks advance.", "votes": 2, "replies": [ { "id": 2267145, "postDate": "2023-05-20T16:18:37.477Z", "content": "

When I turn ou GPU, the max is 13GB. In Normal case, the max is 30GB.
\nCant we use 30GB with GPU?

", "rawMarkdown": "When I turn ou GPU, the max is 13GB. In Normal case, the max is 30GB.\nCant we use 30GB with GPU?", "replies": [ { "id": 2267200, "postDate": "2023-05-20T17:04:07.790Z", "content": "

When we use GPU Kaggle notebook, then we get 16GB CPU RAM (and some is used up so maybe only 13GB available) and we get 16GB GPU RAM when using 1xP100 (so this is 32GB total) or we get 2x16GB GPU RAM when using 2xT4. So with 2xT4 Kaggle notebook, we have a total of 48GB RAM that we can use (when using a combination of RAPIDS and Pandas)

\n

When we use CPU Kaggle notebook then we get 32GB CPU RAM and we get 0GB GPU RAM.

", "rawMarkdown": "When we use GPU Kaggle notebook, then we get 16GB CPU RAM (and some is used up so maybe only 13GB available) **and** we get 16GB GPU RAM when using 1xP100 (so this is 32GB total) or we get 2x16GB GPU RAM when using 2xT4. So with 2xT4 Kaggle notebook, we have a total of 48GB RAM that we can use (when using a combination of RAPIDS and Pandas)\n\nWhen we use CPU Kaggle notebook then we get 32GB CPU RAM and we get 0GB GPU RAM.", "votes": 2 } ] } ] }, { "id": 2197645, "postDate": "2023-03-26T09:27:05.120Z", "content": "

Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.🙌

", "rawMarkdown": "Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.🙌" }, { "id": 2197135, "postDate": "2023-03-25T23:21:02.937Z", "content": "

Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.👍

", "rawMarkdown": "Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.👍" }, { "id": 2231253, "postDate": "2023-04-23T07:25:16.347Z", "content": "

This was very helpful, much appreciated.

", "rawMarkdown": "This was very helpful, much appreciated. " }, { "id": 2196382, "postDate": "2023-03-25T11:04:01.833Z", "rawMarkdown": "", "votes": -1, "isDeleted": true, "replies": [ { "id": 2196676, "postDate": "2023-03-25T15:32:53.360Z", "content": "

These are good suggestions. Thank you

", "rawMarkdown": "These are good suggestions. Thank you" }, { "id": 2197124, "postDate": "2023-03-25T22:57:49.857Z", "content": "

Did you use ChatGPT to write this? lol

", "rawMarkdown": "Did you use ChatGPT to write this? lol", "votes": 1, "replies": [ { "id": 2197206, "postDate": "2023-03-26T01:24:49.763Z", "content": "

yeah, and I wonder why Kaggle doesn't suspend nor ban chronic ChatGPT posters in Kaggle

", "rawMarkdown": "yeah, and I wonder why Kaggle doesn't suspend nor ban chronic ChatGPT posters in Kaggle", "votes": 2 } ] } ] }, { "id": 2219252, "postDate": "2023-04-12T12:41:57.423Z", "content": "

Thank you, I'll try this!

", "rawMarkdown": "Thank you, I'll try this!", "votes": 1 }, { "id": 2218154, "postDate": "2023-04-11T13:17:27.210Z", "content": "

Great info. Thanks

", "rawMarkdown": "Great info. Thanks", "votes": 1 }, { "id": 2214326, "postDate": "2023-04-08T11:17:08.547Z", "content": "

Thank you for your help 👍

", "rawMarkdown": "Thank you for your help 👍", "votes": 1 }, { "id": 2199393, "postDate": "2023-03-27T17:23:21.553Z", "content": "

thank you very much

", "rawMarkdown": "thank you very much", "votes": 1 }, { "id": 2198761, "postDate": "2023-03-27T09:33:09.090Z", "content": "

Great info. Thanks

", "rawMarkdown": "Great info. Thanks ", "votes": 1 }, { "id": 2196809, "postDate": "2023-03-25T17:08:08.137Z", "content": "

Thanks for the tip, super helpful!

", "rawMarkdown": "Thanks for the tip, super helpful!", "votes": 1 }, { "id": 2196339, "postDate": "2023-03-25T10:30:35.270Z", "content": "

Thanks for sharing.

", "rawMarkdown": "Thanks for sharing.", "votes": 1 }, { "id": 2195782, "postDate": "2023-03-24T22:15:43.763Z", "content": "

Thanks for the tip, super helpful!

", "rawMarkdown": "Thanks for the tip, super helpful!", "votes": 1 }, { "id": 2194933, "postDate": "2023-03-24T09:15:08.363Z", "content": "

Thank you for sharing! Helpful

", "rawMarkdown": "Thank you for sharing! Helpful", "votes": 1 }, { "id": 2194203, "postDate": "2023-03-23T19:57:53.027Z", "content": "

Thank you for doing this!

", "rawMarkdown": "Thank you for doing this!", "votes": 1 }, { "id": 2196747, "postDate": "2023-03-25T16:29:49.040Z", "content": "

Thanks, I'll try it in my next project!

", "rawMarkdown": "Thanks, I'll try it in my next project!" } ], "comments": [ { "id": 2284504, "author_name": "YeMenng8888", "author_url": "", "post_date": "2023-06-02T05:42:40.213000", "content": "

Thanks for sharing @cdeotte . I got a problem that it turned back to 8Gb after I added the competition dataset. Is there anything I did wrong?

", "votes": 1, "replies": [ { "id": 2285218, "author_name": "Chris Deotte", "author_url": "", "post_date": "2023-06-02T14:33:43.553000", "content": "

Yes @yemenng8888 . You should not add the competition dataset. You should add the output of my notebook which contains the competition data.

", "votes": 2, "replies": [] } ] }, { "id": 2267246, "author_name": "", "author_url": "", "post_date": "2023-05-20T18:10:33.910000", "content": "

very helpful, much appreciated.

", "votes": 1, "replies": [] }, { "id": 2243782, "author_name": "rajatdogra", "author_url": "", "post_date": "2023-05-03T07:04:03.633000", "content": "

@retroflake

", "votes": 1, "replies": [] }, { "id": 2218272, "author_name": "‎ Srihari", "author_url": "", "post_date": "2023-04-11T14:38:19.160000", "content": "

Thanks for sharing, very helpful @cdeotte

", "votes": 1, "replies": [] }, { "id": 2198297, "author_name": "Paul Johnson", "author_url": "", "post_date": "2023-03-26T21:56:12.753000", "content": "

This has been a great help. Thanks @cdeotte. Is there a decent workaround for sharing a notebook to the competition (not as a submission)? Or is sharing as a Topic in Discussion the workaround?

", "votes": 1, "replies": [ { "id": 2202490, "author_name": "Chris Deotte", "author_url": "", "post_date": "2023-03-30T05:16:43.667000", "content": "

Every notebook is first a commit notebook and not a submit notebook. So you can create a notebook and click \"commit\". That will save it and then you can share it. You do not need to submit it (to the leaderboard) to share it.

", "votes": 1, "replies": [] } ] }, { "id": 2197050, "author_name": "Ayaan Sharif", "author_url": "", "post_date": "2023-03-25T20:36:18.377000", "content": "

Thank you so much for the helpful information on how to get 32GB of RAM for the competition. Your advice on using two different notebooks and attaching the output from your notebook has been invaluable. I really appreciate your willingness to share your knowledge and help others in the community. Thanks again!

", "votes": 1, "replies": [] }, { "id": 2196654, "author_name": "Kenneth Lo", "author_url": "", "post_date": "2023-03-25T15:09:09.903000", "content": "

Thanks, will try that out in my next project

", "votes": 1, "replies": [] }, { "id": 2193939, "author_name": "Pritom Saha", "author_url": "", "post_date": "2023-03-23T15:51:27.300000", "content": "

Valuable information. Thanks for Sharing @cdeotte

", "votes": 1, "replies": [ { "id": 2195756, "author_name": "Chris Deotte", "author_url": "", "post_date": "2023-03-24T21:56:03.860000", "content": "

Thanks Pritom!

", "votes": 0, "replies": [] } ] }, { "id": 2235020, "author_name": "John F J Liu", "author_url": "", "post_date": "2023-04-25T16:27:37.733000", "content": "

Hi ， sorry that i am new to kaggle and I cant see ur picture for \"Six Steps To Get 32GB RAM\", but i tried to copy the code \"https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram\" , but seems the CPU (RAM) is still 13 GB max. do i misundestand anything in ur notebook ? Thanks advance.

", "votes": 2, "replies": [ { "id": 2267145, "author_name": "wakaka", "author_url": "", "post_date": "2023-05-20T16:18:37.477000", "content": "

When I turn ou GPU, the max is 13GB. In Normal case, the max is 30GB.
\nCant we use 30GB with GPU?

", "votes": 0, "replies": [ { "id": 2267200, "author_name": "Chris Deotte", "author_url": "", "post_date": "2023-05-20T17:04:07.790000", "content": "

When we use GPU Kaggle notebook, then we get 16GB CPU RAM (and some is used up so maybe only 13GB available) and we get 16GB GPU RAM when using 1xP100 (so this is 32GB total) or we get 2x16GB GPU RAM when using 2xT4. So with 2xT4 Kaggle notebook, we have a total of 48GB RAM that we can use (when using a combination of RAPIDS and Pandas)

\n

When we use CPU Kaggle notebook then we get 32GB CPU RAM and we get 0GB GPU RAM.

", "votes": 2, "replies": [] } ] } ] }, { "id": 2197645, "author_name": "Ahmed Rasheed", "author_url": "", "post_date": "2023-03-26T09:27:05.120000", "content": "

Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.🙌

", "votes": 0, "replies": [] }, { "id": 2197135, "author_name": "Tariq Mahmood", "author_url": "", "post_date": "2023-03-25T23:21:02.937000", "content": "

Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.👍

", "votes": 0, "replies": [] }, { "id": 2231253, "author_name": "Bassem Gouty", "author_url": "", "post_date": "2023-04-23T07:25:16.347000", "content": "

This was very helpful, much appreciated.

", "votes": 0, "replies": [] }, { "id": 2196382, "author_name": "", "author_url": "", "post_date": "2023-03-25T11:04:01.833000", "content": "", "votes": -1, "replies": [ { "id": 2196676, "author_name": "Chris Deotte", "author_url": "", "post_date": "2023-03-25T15:32:53.360000", "content": "

These are good suggestions. Thank you

", "votes": 0, "replies": [] }, { "id": 2197124, "author_name": "ktyser", "author_url": "", "post_date": "2023-03-25T22:57:49.857000", "content": "

Did you use ChatGPT to write this? lol

", "votes": 1, "replies": [ { "id": 2197206, "author_name": "BwandoWando", "author_url": "", "post_date": "2023-03-26T01:24:49.763000", "content": "

yeah, and I wonder why Kaggle doesn't suspend nor ban chronic ChatGPT posters in Kaggle

", "votes": 2, "replies": [] } ] } ] }, { "id": 2219252, "author_name": "waticson", "author_url": "", "post_date": "2023-04-12T12:41:57.423000", "content": "

Thank you, I'll try this!

", "votes": 1, "replies": [] }, { "id": 2218154, "author_name": "Byungeun Hwang", "author_url": "", "post_date": "2023-04-11T13:17:27.210000", "content": "

Great info. Thanks

", "votes": 1, "replies": [] }, { "id": 2214326, "author_name": "MM (MachineLearningMindset)", "author_url": "", "post_date": "2023-04-08T11:17:08.547000", "content": "

Thank you for your help 👍

", "votes": 1, "replies": [] }, { "id": 2199393, "author_name": "Franciscus Valentinus Ongkosianbhadra", "author_url": "", "post_date": "2023-03-27T17:23:21.553000", "content": "

thank you very much

", "votes": 1, "replies": [] }, { "id": 2198761, "author_name": "Elvid Muslim", "author_url": "", "post_date": "2023-03-27T09:33:09.090000", "content": "

Great info. Thanks

", "votes": 1, "replies": [] }, { "id": 2196809, "author_name": "AndrewNgo", "author_url": "", "post_date": "2023-03-25T17:08:08.137000", "content": "

Thanks for the tip, super helpful!

", "votes": 1, "replies": [] }, { "id": 2196339, "author_name": "Philip8113", "author_url": "", "post_date": "2023-03-25T10:30:35.270000", "content": "

Thanks for sharing.

", "votes": 1, "replies": [] }, { "id": 2195782, "author_name": "Abhinav Naikawadi", "author_url": "", "post_date": "2023-03-24T22:15:43.763000", "content": "

Thanks for the tip, super helpful!

", "votes": 1, "replies": [] }, { "id": 2194933, "author_name": "Elshan Kazim", "author_url": "", "post_date": "2023-03-24T09:15:08.363000", "content": "

Thank you for sharing! Helpful

", "votes": 1, "replies": [] }, { "id": 2194203, "author_name": "Richard Cao", "author_url": "", "post_date": "2023-03-23T19:57:53.027000", "content": "

Thank you for doing this!

", "votes": 1, "replies": [] }, { "id": 2196747, "author_name": "Ericka42", "author_url": "", "post_date": "2023-03-25T16:29:49.040000", "content": "

Thanks, I'll try it in my next project!

", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "2193933": "The most common discussion post in this competition is: \"Your notebook tried to allocate more memory than is available. It has restarted.\". The best way to avoid memory error is to **use two notebooks** in this competition.\n* Train and save models with 32GB RAM notebook (CPU or GPU)\n* Infer from load models with 8GB RAM notebook (CPU)\n\nIt is tricky to get 32GB RAM in this competition because if we create a new Kaggle notebook from **INSIDE** this competition, then Kaggle reduces notebook RAM to 8GB. The trick is to create a new Kaggle notebook **OUTSIDE** this competition from Kaggke's notebook page `https://www.kaggle.com/code`. Then attach the output from my notebook named `https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram`. This will give your notebook the competition data in parquet format and your notebook will have 32GB RAM ! (More info about **two** notebooks [here][1] and \"How to use **one** notebook\" [here][2])\n\n# Six Steps To Get 32GB RAM\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m1.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m2.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m3.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m4.png)\n![](https://raw.githubusercontent.com/cdeotte/Kaggle_Images/main/Mar-2023/m5.png)\n\n[1]: https://www.kaggle.com/competitions/predict-student-performance-from-game-play/discussion/386218\n[2]: https://www.kaggle.com/code/cdeotte/xgboost-baseline-0-680", "2284504": "Thanks for sharing @cdeotte . I got a problem that it turned back to 8Gb after I added the competition dataset. Is there anything I did wrong?", "2267246": "very helpful, much appreciated.", "2243782": "@retroflake ", "2218272": "Thanks for sharing, very helpful @cdeotte ", "2198297": "This has been a great help. Thanks @cdeotte. Is there a decent workaround for sharing a notebook to the competition (not as a submission)? Or is sharing as a Topic in Discussion the workaround?", "2197050": "Thank you so much for the helpful information on how to get 32GB of RAM for the competition. Your advice on using two different notebooks and attaching the output from your notebook has been invaluable. I really appreciate your willingness to share your knowledge and help others in the community. Thanks again!", "2196654": "Thanks, will try that out in my next project", "2193939": "Valuable information. Thanks for Sharing @cdeotte ", "2235020": "Hi ， sorry that i am new to kaggle and I cant see ur picture for \"Six Steps To Get 32GB RAM\", but i tried to copy the code \"https://www.kaggle.com/code/cdeotte/how-to-get-32gb-ram\" , but seems the CPU (RAM) is still 13 GB max. do i misundestand anything in ur notebook ? Thanks advance.", "2197645": "Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.🙌", "2197135": "Thanks @cdeotte for providing a fabulous solution to a real problem of facing a shortage of memory. It is full of knowledgeable insights and a wonderful explanation. It will provide the great help for young as well as experienced Kagglers.👍", "2231253": "This was very helpful, much appreciated. ", "2196382": "", "2219252": "Thank you, I'll try this!", "2218154": "Great info. Thanks", "2214326": "Thank you for your help 👍", "2199393": "thank you very much", "2198761": "Great info. Thanks ", "2196809": "Thanks for the tip, super helpful!", "2196339": "Thanks for sharing.", "2195782": "Thanks for the tip, super helpful!", "2194933": "Thank you for sharing! Helpful", "2194203": "Thank you for doing this!", "2196747": "Thanks, I'll try it in my next project!" } }