{ "id": 289519, "title": "[Compilation] Good starter Notebooks- G-Research Crypto Forecasting ₿💵", "url": "/competitions/g-research-crypto-forecasting/discussion/289519", "author_name": "", "post_date": "2021-11-20T16:50:59.075052200Z", "votes": 43, "comment_count": 6, "views": 0, "content": "

Hi all.
\nAs we move over Halloween and Thanksgiving, and into the Christmas season, May your hot chocolate always be warm enough to drink, and may your socks always be cozy.

\n

To get your brain cells moving, here are a few notebooks:

\n
    \n

  1. Reinforcement learning starter by GM @yamqwe . Other works from them are also great in this competition.

    2. \n

  2. If you like cubes, you'll like this notebook on cube roots by @swaralipibose . The proof is the best part!

    3. \n

  3. New to time series? This notebook has your back by @iamleonie

    4. \n

  4. Official guide does belong here.

    5. \n

  5. An interactive dashboard

    6. \n

  6. Intro to ACF and PACF by @iamleonie

    7. \n

  7. Simple LGBM baseline by @junjitakeshima

    8. \n

  8. Two asset model + FE by @sergears

    9. \n

  9. Catboost regression by @yamqwe

    10. \n

  10. N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting by @yamqwe . I loved learning about this new method. His series is wonderful

    11. \n

  11. Starter LGBM by @saumilagrawal10

    12. \n
\n

This is a very leaky solution, but its worth a read by @julian3833

\n

All due credits to original authors. If you like someone's work, do leave a comment and an upvote

\n

A note to the systematic downvote gang, thanks for stopping by. I hope you're being paid for your service by your masters.

", "messages": [ { "id": "1589866", "postDate": "11/20/2021 16:50:59", "content": "

Hi all.
\nAs we move over Halloween and Thanksgiving, and into the Christmas season, May your hot chocolate always be warm enough to drink, and may your socks always be cozy.

\n

To get your brain cells moving, here are a few notebooks:

\n
    \n

  1. Reinforcement learning starter by GM @yamqwe . Other works from them are also great in this competition.

    2. \n

  2. If you like cubes, you'll like this notebook on cube roots by @swaralipibose . The proof is the best part!

    3. \n

  3. New to time series? This notebook has your back by @iamleonie

    4. \n

  4. Official guide does belong here.

    5. \n

  5. An interactive dashboard

    6. \n

  6. Intro to ACF and PACF by @iamleonie

    7. \n

  7. Simple LGBM baseline by @junjitakeshima

    8. \n

  8. Two asset model + FE by @sergears

    9. \n

  9. Catboost regression by @yamqwe

    10. \n

  10. N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting by @yamqwe . I loved learning about this new method. His series is wonderful

    11. \n

  11. Starter LGBM by @saumilagrawal10

    12. \n
\n

This is a very leaky solution, but its worth a read by @julian3833

\n

All due credits to original authors. If you like someone's work, do leave a comment and an upvote

\n

A note to the systematic downvote gang, thanks for stopping by. I hope you're being paid for your service by your masters.

", "rawMarkdown": "Hi all.\nAs we move over Halloween and Thanksgiving, and into the Christmas season, May your hot chocolate always be warm enough to drink, and may your socks always be cozy.\n\nTo get your brain cells moving, here are a few notebooks:\n1. [Reinforcement learning starter](https://www.kaggle.com/yamqwe/g-research-reinforcement-learning-starter) by GM @yamqwe . Other works from them are also great in this competition.\n\n2. [If you like cubes, you'll like this notebook on cube roots](https://www.kaggle.com/swaralipibose/why-not-use-cube-root-for-outliers-improved-score) by @swaralipibose . The proof is the best part!\n\n3. [New to time series? This notebook has your back](https://www.kaggle.com/iamleonie/to-the-moon-g-research-crypto-forecasting-eda) by @iamleonie \n\n4. [Official guide](https://www.kaggle.com/cstein06/tutorial-to-the-g-research-crypto-competition) does belong here.\n\n5. [An interactive dashboard ](https://www.kaggle.com/toomuchsauce/interactive-crypto-dashboard-indicators-g-crypto)\n\n6. [Intro to ACF and PACF](https://www.kaggle.com/iamleonie/time-series-interpreting-acf-and-pacf) by @iamleonie \n\n7. [Simple LGBM baseline](https://www.kaggle.com/junjitakeshima/crypto-beginner-s-try-for-simple-lgbm-en-jp) by @junjitakeshima \n\n8. [Two asset model + FE](https://www.kaggle.com/sergears/simplified-two-asset-model-feature-engineering) by @sergears\n\n9. [Catboost regression](https://www.kaggle.com/yamqwe/crypto-prediction-catboost-regressor) by @yamqwe \n\n10. [N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting](https://www.kaggle.com/yamqwe/time-series-modeling-n-beats) by @yamqwe . I loved learning about this new method. His series is wonderful\n\n11. [Starter LGBM](https://www.kaggle.com/saumilagrawal10/g-research-starter-lgbm-pipeline-lb-high) by @saumilagrawal10 \n\n[This is a very leaky solution, but its worth a read](https://www.kaggle.com/julian3833/g-research-using-the-overlap-fully-lb-0-99) by @julian3833\n\n**All due credits to original authors. If you like someone's work, do leave a comment and an upvote**\n\nA note to the systematic downvote gang, thanks for stopping by. I hope you're being paid for your service by your masters.", "votes": null }, { "id": "1593990", "postDate": "11/24/2021 13:02:25", "content": "

Thanks for sharing @kritidoneria. Really helpful.

", "rawMarkdown": "Thanks for sharing @kritidoneria. Really helpful.", "votes": null }, { "id": "1594288", "postDate": "11/24/2021 17:09:38", "content": "

Thanks for sharing, upvoted

", "rawMarkdown": "Thanks for sharing, upvoted", "votes": null }, { "id": "1595870", "postDate": "11/26/2021 04:02:31", "content": "

Thanks for sharing, reviewing a few of them tonight.

", "rawMarkdown": "Thanks for sharing, reviewing a few of them tonight.", "votes": null }, { "id": "1598382", "postDate": "11/28/2021 12:58:46", "content": "

Thanks for sharing!

", "rawMarkdown": "Thanks for sharing!", "votes": null }, { "id": "1598705", "postDate": "11/28/2021 17:54:47", "content": "

Thanks for sharing @kritidoneria. This is very useful!

", "rawMarkdown": "Thanks for sharing @kritidoneria. This is very useful!", "votes": null }, { "id": "1608952", "postDate": "12/06/2021 19:54:43", "content": "

Thanks for your sharing. This is very beautiful for me.

", "rawMarkdown": "Thanks for your sharing. This is very beautiful for me.", "votes": null } ], "comments": [ { "id": 1593990, "author_name": "kartik2khandelwal", "author_url": "", "post_date": "11/24/2021 13:02:25", "content": "

Thanks for sharing @kritidoneria. Really helpful.

", "votes": null, "replies": [] }, { "id": 1594288, "author_name": "dhinaharp", "author_url": "", "post_date": "11/24/2021 17:09:38", "content": "

Thanks for sharing, upvoted

", "votes": null, "replies": [] }, { "id": 1595870, "author_name": "cv13j0", "author_url": "", "post_date": "11/26/2021 04:02:31", "content": "

Thanks for sharing, reviewing a few of them tonight.

", "votes": null, "replies": [] }, { "id": 1598382, "author_name": "carriech", "author_url": "", "post_date": "11/28/2021 12:58:46", "content": "

Thanks for sharing!

", "votes": null, "replies": [] }, { "id": 1598705, "author_name": "saurabhbagchi", "author_url": "", "post_date": "11/28/2021 17:54:47", "content": "

Thanks for sharing @kritidoneria. This is very useful!

", "votes": null, "replies": [] }, { "id": 1608952, "author_name": "alexjorson", "author_url": "", "post_date": "12/06/2021 19:54:43", "content": "

Thanks for your sharing. This is very beautiful for me.

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1589866": "Hi all.\nAs we move over Halloween and Thanksgiving, and into the Christmas season, May your hot chocolate always be warm enough to drink, and may your socks always be cozy.\n\nTo get your brain cells moving, here are a few notebooks:\n1. [Reinforcement learning starter](https://www.kaggle.com/yamqwe/g-research-reinforcement-learning-starter) by GM @yamqwe . Other works from them are also great in this competition.\n\n2. [If you like cubes, you'll like this notebook on cube roots](https://www.kaggle.com/swaralipibose/why-not-use-cube-root-for-outliers-improved-score) by @swaralipibose . The proof is the best part!\n\n3. [New to time series? This notebook has your back](https://www.kaggle.com/iamleonie/to-the-moon-g-research-crypto-forecasting-eda) by @iamleonie \n\n4. [Official guide](https://www.kaggle.com/cstein06/tutorial-to-the-g-research-crypto-competition) does belong here.\n\n5. [An interactive dashboard ](https://www.kaggle.com/toomuchsauce/interactive-crypto-dashboard-indicators-g-crypto)\n\n6. [Intro to ACF and PACF](https://www.kaggle.com/iamleonie/time-series-interpreting-acf-and-pacf) by @iamleonie \n\n7. [Simple LGBM baseline](https://www.kaggle.com/junjitakeshima/crypto-beginner-s-try-for-simple-lgbm-en-jp) by @junjitakeshima \n\n8. [Two asset model + FE](https://www.kaggle.com/sergears/simplified-two-asset-model-feature-engineering) by @sergears\n\n9. [Catboost regression](https://www.kaggle.com/yamqwe/crypto-prediction-catboost-regressor) by @yamqwe \n\n10. [N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting](https://www.kaggle.com/yamqwe/time-series-modeling-n-beats) by @yamqwe . I loved learning about this new method. His series is wonderful\n\n11. [Starter LGBM](https://www.kaggle.com/saumilagrawal10/g-research-starter-lgbm-pipeline-lb-high) by @saumilagrawal10 \n\n[This is a very leaky solution, but its worth a read](https://www.kaggle.com/julian3833/g-research-using-the-overlap-fully-lb-0-99) by @julian3833\n\n**All due credits to original authors. If you like someone's work, do leave a comment and an upvote**\n\nA note to the systematic downvote gang, thanks for stopping by. I hope you're being paid for your service by your masters.", "1593990": "Thanks for sharing @kritidoneria. Really helpful.", "1594288": "Thanks for sharing, upvoted", "1595870": "Thanks for sharing, reviewing a few of them tonight.", "1598382": "Thanks for sharing!", "1598705": "Thanks for sharing @kritidoneria. This is very useful!", "1608952": "Thanks for your sharing. This is very beautiful for me." }, "source": "meta" }