{ "id": 357533, "title": "Public artefacts and references from the previous edition", "url": "/competitions/g2net-detecting-continuous-gravitational-waves/discussion/357533", "author_name": "Ravi Ramakrishnan", "post_date": "2022-10-04T15:45:58.026000", "votes": 34, "comment_count": 5, "views": 0, "content": "

Hello participants,

\n

Please find the top approaches and solutions from the previous edition of the competition as below-

\n
    \n
  1. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275507
    \nhttps://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275476 -- these are the winning solutions using data generation. This is very well explained for all to peruse and fathom.
    2. \n
  2. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275617 -- this approach is explained very well with a detailed ensemble of 1-d and 2-d constituents. This approach yielded the 3rd place
    3. \n
  3. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275433 -- the data engineering component is very well elucidated, the approach led to an 8th place finish
    4. \n
  4. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275405 -- this is a terse overview of the 11th place approach
    5. \n
  5. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275841 -- this is a brief overview of the 13th place approach, this seems to be a heavy ensemble
    6. \n
  6. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275440 -- this is an overview of the 19th place solution with the blending done with a public notebook as below-
    \nhttps://www.kaggle.com/itsuki9180/g2net-oof-weight-optimizer
    7. \n
\n

Please find below the top 5 most popular public kernels from the past edition-

\n
    \n
  1. https://www.kaggle.com/code/ihelon/g2net-eda-and-modeling
    2. \n
  2. https://www.kaggle.com/code/yasufuminakama/g2net-efficientnet-b7-baseline-training
    3. \n
  3. https://www.kaggle.com/code/headsortails/when-stars-collide-g2net-eda
    4. \n
  4. https://www.kaggle.com/code/andradaolteanu/g2net-searching-the-sky-pytorch-effnet-w-meta
    5. \n
  5. https://www.kaggle.com/code/allunia/signal-where-are-you
    6. \n
\n

Kindly find the top 3 best scoring public kernels for perusal-

\n
    \n
  1. https://www.kaggle.com/code/hijest/1-g2net-stupid-ensembling
    2. \n
  2. https://www.kaggle.com/code/jbomitchell/effb7-normalised-18-epochs
    3. \n
  3. https://www.kaggle.com/code/vamsikrishnab/gw-detection-blend
    4. \n
\n

Many thanks to the winners of the last edition and the creators of these public artefacts!
\nHope these resources help participants with the current assignment too!

\n

All the best!

", "messages": [ { "id": 1971427, "postDate": "2022-10-04T15:45:58.027Z", "content": "

Hello participants,

\n

Please find the top approaches and solutions from the previous edition of the competition as below-

\n
    \n
  1. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275507
    \nhttps://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275476 -- these are the winning solutions using data generation. This is very well explained for all to peruse and fathom.
    2. \n
  2. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275617 -- this approach is explained very well with a detailed ensemble of 1-d and 2-d constituents. This approach yielded the 3rd place
    3. \n
  3. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275433 -- the data engineering component is very well elucidated, the approach led to an 8th place finish
    4. \n
  4. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275405 -- this is a terse overview of the 11th place approach
    5. \n
  5. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275841 -- this is a brief overview of the 13th place approach, this seems to be a heavy ensemble
    6. \n
  6. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275440 -- this is an overview of the 19th place solution with the blending done with a public notebook as below-
    \nhttps://www.kaggle.com/itsuki9180/g2net-oof-weight-optimizer
    7. \n
\n

Please find below the top 5 most popular public kernels from the past edition-

\n
    \n
  1. https://www.kaggle.com/code/ihelon/g2net-eda-and-modeling
    2. \n
  2. https://www.kaggle.com/code/yasufuminakama/g2net-efficientnet-b7-baseline-training
    3. \n
  3. https://www.kaggle.com/code/headsortails/when-stars-collide-g2net-eda
    4. \n
  4. https://www.kaggle.com/code/andradaolteanu/g2net-searching-the-sky-pytorch-effnet-w-meta
    5. \n
  5. https://www.kaggle.com/code/allunia/signal-where-are-you
    6. \n
\n

Kindly find the top 3 best scoring public kernels for perusal-

\n
    \n
  1. https://www.kaggle.com/code/hijest/1-g2net-stupid-ensembling
    2. \n
  2. https://www.kaggle.com/code/jbomitchell/effb7-normalised-18-epochs
    3. \n
  3. https://www.kaggle.com/code/vamsikrishnab/gw-detection-blend
    4. \n
\n

Many thanks to the winners of the last edition and the creators of these public artefacts!
\nHope these resources help participants with the current assignment too!

\n

All the best!

", "rawMarkdown": "Hello participants,\n\nPlease find the top approaches and solutions from the previous edition of the competition as below-\n1. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275507 \nhttps://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275476 -- these are the winning solutions using data generation. This is very well explained for all to peruse and fathom. \n2. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275617 -- this approach is explained very well with a detailed ensemble of 1-d and 2-d constituents. This approach yielded the 3rd place\n3. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275433 -- the data engineering component is very well elucidated, the approach led to an 8th place finish\n4. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275405 -- this is a terse overview of the 11th place approach\n5. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275841 -- this is a brief overview of the 13th place approach, this seems to be a heavy ensemble\n6. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275440 -- this is an overview of the 19th place solution with the blending done with a public notebook as below-\nhttps://www.kaggle.com/itsuki9180/g2net-oof-weight-optimizer\n\nPlease find below the top 5 most popular public kernels from the past edition- \n1. https://www.kaggle.com/code/ihelon/g2net-eda-and-modeling\n2. https://www.kaggle.com/code/yasufuminakama/g2net-efficientnet-b7-baseline-training\n3. https://www.kaggle.com/code/headsortails/when-stars-collide-g2net-eda\n4. https://www.kaggle.com/code/andradaolteanu/g2net-searching-the-sky-pytorch-effnet-w-meta\n5. https://www.kaggle.com/code/allunia/signal-where-are-you\n\nKindly find the top 3 best scoring public kernels for perusal- \n1. https://www.kaggle.com/code/hijest/1-g2net-stupid-ensembling\n2. https://www.kaggle.com/code/jbomitchell/effb7-normalised-18-epochs\n3. https://www.kaggle.com/code/vamsikrishnab/gw-detection-blend\n\nMany thanks to the winners of the last edition and the creators of these public artefacts! \nHope these resources help participants with the current assignment too!\n\nAll the best!\n", "votes": 34 }, { "id": 2002609, "postDate": "2022-10-25T00:20:17.217Z", "content": "

Thanks for the info. It helps me a lot! I have a question. Does this competition as the same as the previous one. The goal for the both competitions is detecting the appearance of the signal. Does that mean the approach of the previous competition also work for the current one?

", "rawMarkdown": "Thanks for the info. It helps me a lot! I have a question. Does this competition as the same as the previous one. The goal for the both competitions is detecting the appearance of the signal. Does that mean the approach of the previous competition also work for the current one?", "votes": 1 }, { "id": 1972731, "postDate": "2022-10-05T09:50:46.323Z", "content": "

Informative

", "rawMarkdown": "Informative", "votes": 1, "replies": [ { "id": 1972783, "postDate": "2022-10-05T10:22:34.350Z", "content": "

I am happy if it helps @faisaljanjua0555

", "rawMarkdown": "I am happy if it helps @faisaljanjua0555 " } ] }, { "id": 1971531, "postDate": "2022-10-04T16:56:04.650Z", "content": "

Thanks for the infodump!

", "rawMarkdown": "Thanks for the infodump!", "votes": 1, "replies": [ { "id": 1971624, "postDate": "2022-10-04T17:41:47.907Z", "content": "

Welcome, hope this helps @chazzer

", "rawMarkdown": "Welcome, hope this helps @chazzer " } ] } ], "comments": [ { "id": 2002609, "author_name": "Chenjie", "author_url": "", "post_date": "2022-10-25T00:20:17.217000", "content": "

Thanks for the info. It helps me a lot! I have a question. Does this competition as the same as the previous one. The goal for the both competitions is detecting the appearance of the signal. Does that mean the approach of the previous competition also work for the current one?

", "votes": 1, "replies": [] }, { "id": 1972731, "author_name": "Muhammad Faisal Ali", "author_url": "", "post_date": "2022-10-05T09:50:46.323000", "content": "

Informative

", "votes": 1, "replies": [ { "id": 1972783, "author_name": "Ravi Ramakrishnan", "author_url": "", "post_date": "2022-10-05T10:22:34.350000", "content": "

I am happy if it helps @faisaljanjua0555

", "votes": 0, "replies": [] } ] }, { "id": 1971531, "author_name": "Marco Ciavarella", "author_url": "", "post_date": "2022-10-04T16:56:04.650000", "content": "

Thanks for the infodump!

", "votes": 1, "replies": [ { "id": 1971624, "author_name": "Ravi Ramakrishnan", "author_url": "", "post_date": "2022-10-04T17:41:47.907000", "content": "

Welcome, hope this helps @chazzer

", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "1971427": "Hello participants,\n\nPlease find the top approaches and solutions from the previous edition of the competition as below-\n1. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275507 \nhttps://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275476 -- these are the winning solutions using data generation. This is very well explained for all to peruse and fathom. \n2. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275617 -- this approach is explained very well with a detailed ensemble of 1-d and 2-d constituents. This approach yielded the 3rd place\n3. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275433 -- the data engineering component is very well elucidated, the approach led to an 8th place finish\n4. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275405 -- this is a terse overview of the 11th place approach\n5. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275841 -- this is a brief overview of the 13th place approach, this seems to be a heavy ensemble\n6. https://www.kaggle.com/competitions/g2net-gravitational-wave-detection/discussion/275440 -- this is an overview of the 19th place solution with the blending done with a public notebook as below-\nhttps://www.kaggle.com/itsuki9180/g2net-oof-weight-optimizer\n\nPlease find below the top 5 most popular public kernels from the past edition- \n1. https://www.kaggle.com/code/ihelon/g2net-eda-and-modeling\n2. https://www.kaggle.com/code/yasufuminakama/g2net-efficientnet-b7-baseline-training\n3. https://www.kaggle.com/code/headsortails/when-stars-collide-g2net-eda\n4. https://www.kaggle.com/code/andradaolteanu/g2net-searching-the-sky-pytorch-effnet-w-meta\n5. https://www.kaggle.com/code/allunia/signal-where-are-you\n\nKindly find the top 3 best scoring public kernels for perusal- \n1. https://www.kaggle.com/code/hijest/1-g2net-stupid-ensembling\n2. https://www.kaggle.com/code/jbomitchell/effb7-normalised-18-epochs\n3. https://www.kaggle.com/code/vamsikrishnab/gw-detection-blend\n\nMany thanks to the winners of the last edition and the creators of these public artefacts! \nHope these resources help participants with the current assignment too!\n\nAll the best!\n", "2002609": "Thanks for the info. It helps me a lot! I have a question. Does this competition as the same as the previous one. The goal for the both competitions is detecting the appearance of the signal. Does that mean the approach of the previous competition also work for the current one?", "1972731": "Informative", "1971531": "Thanks for the infodump!" } }