{ "id": 473973, "title": "Similar competitions important kernels and discussions for references ", "url": "/competitions/home-credit-credit-risk-model-stability/discussion/473973", "author_name": "Athar Sayed", "post_date": "2024-02-06T16:53:00.856000", "votes": 38, "comment_count": 7, "views": 0, "content": "

Two years ago, the 'Amex Default Prediction' competition on Kaggle sparked intense interest in predicting loan defaults. Now, with a new competition underway, we revisit the top ideas, discussions, and kernels from the previous challenge, exploring how innovative techniques continue to shape the field of loan default prediction

\n

Competition : American Express - Default Prediction
\nIts almost similar to this competition

\n

Important Discussions

\n
    \n

  1. Discussion on how to reduce dataset size can be found here , can come very handy in this competition when dealing with large set of features and models

    2. \n

  2. Discussion on Amazing tips and tricks for Tabular Data competition can be found here , it has amazing compilation of tips and tricks which can be very useful for this competition

    3. \n

  3. Amazing discussion on using gpu enabled xgboost and lgb to speed up model training can be found here

    4. \n
\n

Top Voted Kernels:

\n
    \n

  1. Amazing Xgboost starter kernel by Chris deotte can be found here

    2. \n

  2. Another kernel on EDA can be found here

    3. \n

  3. Kernel on building LGB with Dart CV can be found here

    4. \n

  4. Tensorflow based GRU starter can be found here

    5. \n
\n

Winning solutions:

\n
    \n
  1. 1st place solution with code
    2. \n
\n

2. 2nd place solution\n

\n
    \n

  1. 3rd place solution

    2. \n

  2. 10th place solution

    3. \n

  3. 11th place solution

    4. \n

  4. 13th place solution

    5. \n

  5. 14th place solution

    6. \n
\n

šŸ” As we reflect on the past competition and anticipate the outcomes of the current one, it's clear that the pursuit of accurate loan default prediction remains as relevant as ever. With each challenge, the community drives forward, pushing the boundaries of what's possible in financial forecasting. šŸ“ˆ

", "messages": [ { "id": 2639057, "postDate": "2024-02-06T16:53:00.857Z", "content": "

Two years ago, the 'Amex Default Prediction' competition on Kaggle sparked intense interest in predicting loan defaults. Now, with a new competition underway, we revisit the top ideas, discussions, and kernels from the previous challenge, exploring how innovative techniques continue to shape the field of loan default prediction

\n

Competition : American Express - Default Prediction
\nIts almost similar to this competition

\n

Important Discussions

\n
    \n

  1. Discussion on how to reduce dataset size can be found here , can come very handy in this competition when dealing with large set of features and models

    2. \n

  2. Discussion on Amazing tips and tricks for Tabular Data competition can be found here , it has amazing compilation of tips and tricks which can be very useful for this competition

    3. \n

  3. Amazing discussion on using gpu enabled xgboost and lgb to speed up model training can be found here

    4. \n
\n

Top Voted Kernels:

\n
    \n

  1. Amazing Xgboost starter kernel by Chris deotte can be found here

    2. \n

  2. Another kernel on EDA can be found here

    3. \n

  3. Kernel on building LGB with Dart CV can be found here

    4. \n

  4. Tensorflow based GRU starter can be found here

    5. \n
\n

Winning solutions:

\n
    \n
  1. 1st place solution with code
    2. \n
\n

2. 2nd place solution\n

\n
    \n

  1. 3rd place solution

    2. \n

  2. 10th place solution

    3. \n

  3. 11th place solution

    4. \n

  4. 13th place solution

    5. \n

  5. 14th place solution

    6. \n
\n

šŸ” As we reflect on the past competition and anticipate the outcomes of the current one, it's clear that the pursuit of accurate loan default prediction remains as relevant as ever. With each challenge, the community drives forward, pushing the boundaries of what's possible in financial forecasting. šŸ“ˆ

", "rawMarkdown": "Two years ago, the 'Amex Default Prediction' competition on Kaggle sparked intense interest in predicting loan defaults. Now, with a new competition underway, we revisit the top ideas, discussions, and kernels from the previous challenge, exploring how innovative techniques continue to shape the field of loan default prediction\n\nCompetition : [American Express - Default Prediction](https://www.kaggle.com/competitions/amex-default-prediction/overview)\nIts almost similar to this competition\n\nImportant Discussions\n1. Discussion on how to reduce dataset size can be found [here](https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054) , can come very handy in this competition when dealing with large set of features and models\n\n2. Discussion on Amazing tips and tricks for Tabular Data competition can be found [here](https://www.kaggle.com/competitions/amex-default-prediction/discussion/335892) , it has amazing compilation of tips and tricks which can be very useful for this competition\n\n3. Amazing discussion on using gpu enabled xgboost and lgb to speed up model training can be found [here](https://www.kaggle.com/competitions/amex-default-prediction/discussion/328606)\n\nTop Voted Kernels:\n1. Amazing Xgboost starter kernel by Chris deotte can be found [here](https://www.kaggle.com/code/cdeotte/xgboost-starter-0-793)\n\n2. Another kernel on EDA can be found [here](https://www.kaggle.com/code/ambrosm/amex-eda-which-makes-sense)\n\n3. Kernel on building LGB with Dart CV can be found [here](https://www.kaggle.com/code/ragnar123/amex-lgbm-dart-cv-0-7977)\n\n4. Tensorflow based GRU starter can be found [here ](https://www.kaggle.com/code/cdeotte/tensorflow-gru-starter-0-790)\n\nWinning solutions:\n1. [1st place solution with code](https://www.kaggle.com/competitions/amex-default-prediction/discussion/348111)\n\n2.[ 2nd place solution\n](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347637)\n\n3. [3rd place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/349741)\n\n4. [10th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347668)\n\n5. [11th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347786)\n\n6. [13th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/348014)\n\n7. [14th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347641)\n\nšŸ” As we reflect on the past competition and anticipate the outcomes of the current one, it's clear that the pursuit of accurate loan default prediction remains as relevant as ever. With each challenge, the community drives forward, pushing the boundaries of what's possible in financial forecasting. šŸ“ˆ", "votes": 36 }, { "id": 2642267, "postDate": "2024-02-08T04:12:46.093Z", "content": "

thanks for the for references, can't wait to start on this

", "rawMarkdown": "thanks for the for references, can't wait to start on this", "votes": 1 }, { "id": 2640678, "postDate": "2024-02-07T03:37:38.717Z", "content": "

Thank you very much @sayedathar11 ! I am planning to join this competition and this is really helpful!

", "rawMarkdown": "Thank you very much @sayedathar11 ! I am planning to join this competition and this is really helpful!", "votes": 1 }, { "id": 2639068, "postDate": "2024-02-06T16:57:27.640Z", "content": "

Hi, @sayedathar11! Helpful post. Reduced my search time for similar solutions. Thank you!

", "rawMarkdown": "Hi, @sayedathar11! Helpful post. Reduced my search time for similar solutions. Thank you!", "votes": 1, "replies": [ { "id": 2639077, "postDate": "2024-02-06T16:59:42.103Z", "content": "

@bratkovskyevgeny you are most welcome happy kaggling :) , do share useful resources if you come across too

", "rawMarkdown": "@bratkovskyevgeny you are most welcome happy kaggling :) , do share useful resources if you come across too", "votes": 1, "replies": [ { "id": 2639116, "postDate": "2024-02-06T17:14:19.727Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2639128, "postDate": "2024-02-06T17:21:41.853Z", "content": "

@sayedathar11 , I found a similar competition on Kaggle: https://www.kaggle.com/competitions/home-credit-default-risk. You may find some useful ideas there.

", "rawMarkdown": "@sayedathar11 , I found a similar competition on Kaggle: https://www.kaggle.com/competitions/home-credit-default-risk. You may find some useful ideas there." } ] } ] }, { "id": 2639214, "postDate": "2024-02-06T18:16:14.753Z", "content": "

Thanks for the post. It was useful!

", "rawMarkdown": "Thanks for the post. It was useful!", "votes": 1 } ], "comments": [ { "id": 2642267, "author_name": "Tommy Chan", "author_url": "", "post_date": "2024-02-08T04:12:46.093000", "content": "

thanks for the for references, can't wait to start on this

", "votes": 1, "replies": [] }, { "id": 2640678, "author_name": "Azrai Mahadan", "author_url": "", "post_date": "2024-02-07T03:37:38.717000", "content": "

Thank you very much @sayedathar11 ! I am planning to join this competition and this is really helpful!

", "votes": 1, "replies": [] }, { "id": 2639068, "author_name": "Evgeny Bratkovsky", "author_url": "", "post_date": "2024-02-06T16:57:27.640000", "content": "

Hi, @sayedathar11! Helpful post. Reduced my search time for similar solutions. Thank you!

", "votes": 1, "replies": [ { "id": 2639077, "author_name": "Athar Sayed", "author_url": "", "post_date": "2024-02-06T16:59:42.103000", "content": "

@bratkovskyevgeny you are most welcome happy kaggling :) , do share useful resources if you come across too

", "votes": 1, "replies": [ { "id": 2639116, "author_name": "", "author_url": "", "post_date": "2024-02-06T17:14:19.727000", "content": "", "votes": 0, "replies": [] }, { "id": 2639128, "author_name": "Evgeny Bratkovsky", "author_url": "", "post_date": "2024-02-06T17:21:41.853000", "content": "

@sayedathar11 , I found a similar competition on Kaggle: https://www.kaggle.com/competitions/home-credit-default-risk. You may find some useful ideas there.

", "votes": 0, "replies": [] } ] } ] }, { "id": 2639214, "author_name": "MrSimple", "author_url": "", "post_date": "2024-02-06T18:16:14.753000", "content": "

Thanks for the post. It was useful!

", "votes": 1, "replies": [] } ], "raw_markdown_by_id": { "2639057": "Two years ago, the 'Amex Default Prediction' competition on Kaggle sparked intense interest in predicting loan defaults. Now, with a new competition underway, we revisit the top ideas, discussions, and kernels from the previous challenge, exploring how innovative techniques continue to shape the field of loan default prediction\n\nCompetition : [American Express - Default Prediction](https://www.kaggle.com/competitions/amex-default-prediction/overview)\nIts almost similar to this competition\n\nImportant Discussions\n1. Discussion on how to reduce dataset size can be found [here](https://www.kaggle.com/competitions/amex-default-prediction/discussion/328054) , can come very handy in this competition when dealing with large set of features and models\n\n2. Discussion on Amazing tips and tricks for Tabular Data competition can be found [here](https://www.kaggle.com/competitions/amex-default-prediction/discussion/335892) , it has amazing compilation of tips and tricks which can be very useful for this competition\n\n3. Amazing discussion on using gpu enabled xgboost and lgb to speed up model training can be found [here](https://www.kaggle.com/competitions/amex-default-prediction/discussion/328606)\n\nTop Voted Kernels:\n1. Amazing Xgboost starter kernel by Chris deotte can be found [here](https://www.kaggle.com/code/cdeotte/xgboost-starter-0-793)\n\n2. Another kernel on EDA can be found [here](https://www.kaggle.com/code/ambrosm/amex-eda-which-makes-sense)\n\n3. Kernel on building LGB with Dart CV can be found [here](https://www.kaggle.com/code/ragnar123/amex-lgbm-dart-cv-0-7977)\n\n4. Tensorflow based GRU starter can be found [here ](https://www.kaggle.com/code/cdeotte/tensorflow-gru-starter-0-790)\n\nWinning solutions:\n1. [1st place solution with code](https://www.kaggle.com/competitions/amex-default-prediction/discussion/348111)\n\n2.[ 2nd place solution\n](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347637)\n\n3. [3rd place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/349741)\n\n4. [10th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347668)\n\n5. [11th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347786)\n\n6. [13th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/348014)\n\n7. [14th place solution](https://www.kaggle.com/competitions/amex-default-prediction/discussion/347641)\n\nšŸ” As we reflect on the past competition and anticipate the outcomes of the current one, it's clear that the pursuit of accurate loan default prediction remains as relevant as ever. With each challenge, the community drives forward, pushing the boundaries of what's possible in financial forecasting. šŸ“ˆ", "2642267": "thanks for the for references, can't wait to start on this", "2640678": "Thank you very much @sayedathar11 ! I am planning to join this competition and this is really helpful!", "2639068": "Hi, @sayedathar11! Helpful post. Reduced my search time for similar solutions. Thank you!", "2639214": "Thanks for the post. It was useful!" } }