I'm excited to be announcing our newest TPU Stars, awarded to the most knowledgeable and helpful TPU experts in this competition.
\n✨ Chris Deotte was a much-touted source of TPU knowledge and support in this competition and a repeat TPU Star winner! He contributed widely-used TFRecords, and his Triple Stratified KFold with TFRecords was an outstanding starter for working with TFRecords on TPUs in this competition.
\n✨ Ertuğrul Demir whose well-organized Analysis of Melanoma Metadata and EffNet Ensemble notebook does a great job of stepping through an approach to this competition's dataset, from exploring the data to creating an EfficientNet ensemble.
\n✨ AgentAuers authored this fantastic TF + TPUs EfficientNet baseline, Incredible TPUs - finetune EffNetB0-B6 at once, which countless others leveraged in their own notebooks!
\n✨ Alexey Pronin wrote the EfficientNet BN+Tabular Features TF CV5 512x512 notebook that really promoted experimentation through an end-to-end detailed model development walk-through.
\n✨ Mobassir built upon other stars' contributions with his own qualitative expansion in his In-Depth Melanoma with modeling notebook.
All of these stars will be seeing their 20 hrs/wk of additional TPU quota for each of the next four weeks. Please join me in congratulating all of them!
\n✨ An extra special honorable mention to our Kaggle intern, Amy Jang who made tremendous notebook contributions during her few months with us, including this TensorFlow + Transfer Learning: Melanoma notebook, demonstrating how to approach this dataset using TensorFlow with transfer learning.
", "messages": [ { "id": 1023034, "postDate": "2020-09-22T22:39:24.310Z", "content": "I'm excited to be announcing our newest TPU Stars, awarded to the most knowledgeable and helpful TPU experts in this competition.
\n✨ Chris Deotte was a much-touted source of TPU knowledge and support in this competition and a repeat TPU Star winner! He contributed widely-used TFRecords, and his Triple Stratified KFold with TFRecords was an outstanding starter for working with TFRecords on TPUs in this competition.
\n✨ Ertuğrul Demir whose well-organized Analysis of Melanoma Metadata and EffNet Ensemble notebook does a great job of stepping through an approach to this competition's dataset, from exploring the data to creating an EfficientNet ensemble.
\n✨ AgentAuers authored this fantastic TF + TPUs EfficientNet baseline, Incredible TPUs - finetune EffNetB0-B6 at once, which countless others leveraged in their own notebooks!
\n✨ Alexey Pronin wrote the EfficientNet BN+Tabular Features TF CV5 512x512 notebook that really promoted experimentation through an end-to-end detailed model development walk-through.
\n✨ Mobassir built upon other stars' contributions with his own qualitative expansion in his In-Depth Melanoma with modeling notebook.
All of these stars will be seeing their 20 hrs/wk of additional TPU quota for each of the next four weeks. Please join me in congratulating all of them!
\n✨ An extra special honorable mention to our Kaggle intern, Amy Jang who made tremendous notebook contributions during her few months with us, including this TensorFlow + Transfer Learning: Melanoma notebook, demonstrating how to approach this dataset using TensorFlow with transfer learning.
", "rawMarkdown": "I'm excited to be announcing our newest [TPU Stars](https://www.kaggle.com/tpu-stars), awarded to the [most knowledgeable and helpful TPU experts](https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/156966) in this competition.\n\n✨ [Chris Deotte](https://www.kaggle.com/cdeotte) was a much-touted source of TPU knowledge and support in this competition and a repeat TPU Star winner! He contributed widely-used TFRecords, and his [Triple Stratified KFold with TFRecords](https://www.kaggle.com/cdeotte/triple-stratified-kfold-with-tfrecords) was an outstanding starter for working with TFRecords on TPUs in this competition.\n✨ [Ertuğrul Demir](https://www.kaggle.com/datafan07) whose well-organized [Analysis of Melanoma Metadata and EffNet Ensemble](https://www.kaggle.com/datafan07/analysis-of-melanoma-metadata-and-effnet-ensemble) notebook does a great job of stepping through an approach to this competition's dataset, from exploring the data to creating an EfficientNet ensemble.\n✨ [AgentAuers](https://www.kaggle.com/agentauers) authored this fantastic TF + TPUs EfficientNet baseline, [Incredible TPUs - finetune EffNetB0-B6 at once](https://www.kaggle.com/agentauers/incredible-tpus-finetune-effnetb0-b6-at-once), which countless others leveraged in their own notebooks!\n✨ [Alexey Pronin](https://www.kaggle.com/graf10a) wrote the [EfficientNet BN+Tabular Features TF CV5 512x512](https://www.kaggle.com/graf10a/efficientnet-bn-tabular-features-tf-cv5-512x512) notebook that really promoted experimentation through an end-to-end detailed model development walk-through.\n✨ [Mobassir](https://www.kaggle.com/mobassir) built upon other stars' contributions with his own qualitative expansion in his [In-Depth Melanoma with modeling](https://www.kaggle.com/mobassir/in-depth-melanoma-with-modeling) notebook.\n\nAll of these stars will be seeing their 20 hrs/wk of additional TPU quota for each of the next four weeks. Please join me in congratulating all of them!\n\n✨ An extra special honorable mention to our Kaggle intern, [Amy Jang](https://www.kaggle.com/amyjang) who made tremendous notebook contributions during her few months with us, including this [TensorFlow + Transfer Learning: Melanoma](https://www.kaggle.com/amyjang/tensorflow-transfer-learning-melanoma) notebook, demonstrating how to approach this dataset using TensorFlow with transfer learning.", "votes": 30 }, { "id": 1023095, "postDate": "2020-09-23T00:38:03.423Z", "content": "Congratulations @datafan07 @agentauers @graf10a @mobassir @amyjang !! Thanks for all your contributions.
", "rawMarkdown": "Congratulations @datafan07 @agentauers @graf10a @mobassir @amyjang !! Thanks for all your contributions.", "votes": 6, "replies": [ { "id": 1023938, "postDate": "2020-09-23T14:41:01.627Z", "content": "@cdeotte Congrats, learnt and yet to learn lot from you.
", "rawMarkdown": "@cdeotte Congrats, learnt and yet to learn lot from you.", "votes": 1 } ] }, { "id": 1024705, "postDate": "2020-09-24T04:38:42.960Z", "content": "Thank you @cdeotte @datafan07 @agentauers @graf10a @mobassir @amyjang for the awesome contributions!
", "rawMarkdown": "Thank you @cdeotte @datafan07 @agentauers @graf10a @mobassir @amyjang for the awesome contributions!", "votes": 4 }, { "id": 1023663, "postDate": "2020-09-23T11:17:11.657Z", "content": "Thanks for including me with these great kagglers and congratulations all!
", "rawMarkdown": "Thanks for including me with these great kagglers and congratulations all!", "votes": 1 }, { "id": 1024264, "postDate": "2020-09-23T18:12:21.493Z", "content": "great work !! Many congratulations to @datafan07 @agentauers @graf10a @mobassir @amyjang !
", "rawMarkdown": "great work !! Many congratulations to @datafan07 @agentauers @graf10a @mobassir @amyjang !", "votes": 2 }, { "id": 1023188, "postDate": "2020-09-23T02:46:37.807Z", "rawMarkdown": "", "votes": 2, "isDeleted": true } ], "comments": [ { "id": 1023095, "author_name": "Chris Deotte", "author_url": "", "post_date": "2020-09-23T00:38:03.423000", "content": "Congratulations @datafan07 @agentauers @graf10a @mobassir @amyjang !! Thanks for all your contributions.
", "votes": 6, "replies": [ { "id": 1023938, "author_name": "Rashmi Margani", "author_url": "", "post_date": "2020-09-23T14:41:01.627000", "content": "@cdeotte Congrats, learnt and yet to learn lot from you.
", "votes": 1, "replies": [] } ] }, { "id": 1024705, "author_name": "Ronaldo S.A. Batista", "author_url": "", "post_date": "2020-09-24T04:38:42.960000", "content": "Thank you @cdeotte @datafan07 @agentauers @graf10a @mobassir @amyjang for the awesome contributions!
", "votes": 4, "replies": [] }, { "id": 1023663, "author_name": "Ertuğrul Demir", "author_url": "", "post_date": "2020-09-23T11:17:11.657000", "content": "Thanks for including me with these great kagglers and congratulations all!
", "votes": 1, "replies": [] }, { "id": 1024264, "author_name": "IndrajitSingh", "author_url": "", "post_date": "2020-09-23T18:12:21.493000", "content": "great work !! Many congratulations to @datafan07 @agentauers @graf10a @mobassir @amyjang !
", "votes": 2, "replies": [] }, { "id": 1023188, "author_name": "", "author_url": "", "post_date": "2020-09-23T02:46:37.807000", "content": "", "votes": 2, "replies": [] } ], "raw_markdown_by_id": { "1023034": "I'm excited to be announcing our newest [TPU Stars](https://www.kaggle.com/tpu-stars), awarded to the [most knowledgeable and helpful TPU experts](https://www.kaggle.com/c/siim-isic-melanoma-classification/discussion/156966) in this competition.\n\n✨ [Chris Deotte](https://www.kaggle.com/cdeotte) was a much-touted source of TPU knowledge and support in this competition and a repeat TPU Star winner! He contributed widely-used TFRecords, and his [Triple Stratified KFold with TFRecords](https://www.kaggle.com/cdeotte/triple-stratified-kfold-with-tfrecords) was an outstanding starter for working with TFRecords on TPUs in this competition.\n✨ [Ertuğrul Demir](https://www.kaggle.com/datafan07) whose well-organized [Analysis of Melanoma Metadata and EffNet Ensemble](https://www.kaggle.com/datafan07/analysis-of-melanoma-metadata-and-effnet-ensemble) notebook does a great job of stepping through an approach to this competition's dataset, from exploring the data to creating an EfficientNet ensemble.\n✨ [AgentAuers](https://www.kaggle.com/agentauers) authored this fantastic TF + TPUs EfficientNet baseline, [Incredible TPUs - finetune EffNetB0-B6 at once](https://www.kaggle.com/agentauers/incredible-tpus-finetune-effnetb0-b6-at-once), which countless others leveraged in their own notebooks!\n✨ [Alexey Pronin](https://www.kaggle.com/graf10a) wrote the [EfficientNet BN+Tabular Features TF CV5 512x512](https://www.kaggle.com/graf10a/efficientnet-bn-tabular-features-tf-cv5-512x512) notebook that really promoted experimentation through an end-to-end detailed model development walk-through.\n✨ [Mobassir](https://www.kaggle.com/mobassir) built upon other stars' contributions with his own qualitative expansion in his [In-Depth Melanoma with modeling](https://www.kaggle.com/mobassir/in-depth-melanoma-with-modeling) notebook.\n\nAll of these stars will be seeing their 20 hrs/wk of additional TPU quota for each of the next four weeks. Please join me in congratulating all of them!\n\n✨ An extra special honorable mention to our Kaggle intern, [Amy Jang](https://www.kaggle.com/amyjang) who made tremendous notebook contributions during her few months with us, including this [TensorFlow + Transfer Learning: Melanoma](https://www.kaggle.com/amyjang/tensorflow-transfer-learning-melanoma) notebook, demonstrating how to approach this dataset using TensorFlow with transfer learning.", "1023095": "Congratulations @datafan07 @agentauers @graf10a @mobassir @amyjang !! Thanks for all your contributions.", "1024705": "Thank you @cdeotte @datafan07 @agentauers @graf10a @mobassir @amyjang for the awesome contributions!", "1023663": "Thanks for including me with these great kagglers and congratulations all!", "1024264": "great work !! Many congratulations to @datafan07 @agentauers @graf10a @mobassir @amyjang !", "1023188": "" } }