{ "id": 104338, "title": "How to study '2019 3rd ML month with KaKR' (커널 정리 및 추천)", "url": "/competitions/2019-3rd-ml-month-with-kakr/discussion/104338", "author_name": "", "post_date": "2019-08-16T07:12:24.693695300Z", "votes": 13, "comment_count": 3, "views": 0, "content": "

오늘 대회 마지막 날이네요.

\n\n

대회가 끝나고도 열심히 공부하실 분들을 위해 커널들을 카테고리별로 정리해보았습니다.\n현재까지 올라온 모든 커널들을 검토하여 커널의 완성도, 기여도, 인기도 등을 고려하여 선정하였습니다.

\n\n

마지막까지 화이팅입니다!

\n\n
    \n

  1. Baseline\n[3rd ML Month] Car Model Classification Baseline (Keras / TaeJin Kim / 75+) - official\nPytorch Baseline (updated 7.08) (Pytorch / saewonYang / 50+)\n[3rd ML Month] Xception, StratifiedKFold, Ensemble (Keras / Jang / 35+)\nXception Model in Keras for Car Classification (Keras / Chanran Kim / 24+)\n[ka.kr sillim] Fast.ai (High Level Framework) (Fast.ai / HoMin / 16+)\n[KaKR-2019-3rd]EDA, ImagePrep, Mixup, CV (Keras) (Keras / Bruce Kim / 7+)

    2. \n

  2. EDA\n2019 3rd ML month with KaKR - EDA (EDA / Hyeonho Lee / 23+)\nWant to see my Augmentations (pytorch) (EDA / saewonYang / 14+)\n[ka.kr sillim]Data_check (EDA / HoMin / 11+)\n[3rd ML Month] EDA (EDA / Taegwan Kim / 4+)

    3. \n

  3. Preprocessing\n[3rd ML Month] Car Image Cropping - updated 7.10 (Bounding box / Taemyung Heo / 45+)\nCutout Augmentation (on Keras EfficientNet) (Cutout Augmentation / Chanran Kim / 25+)\n[3rd ML Month] Car Image Segmentation Crop (Segmentation / Steve Jang / 14+)\n[ka.kr sillim] AutoAugmentation (AutoAugmentation / HoMin / 12+)

    4. \n

  4. Training\nHow to split folds and train (Cross Validation / saewonYang / 14+)\nKeras, How to use pretrained model? (Pretrained model / Daehun Gwak / 14+)\nKeras Semi-supervised Learning (Semi-supervised learning / Taegwan Kim / 9+)

    5. \n

  5. Etc\nA Race for 2019 3rd ML month with KaKR (Public score Trend / Taemyung Heo / 15+)

    6. \n
\n\n

*Kernel name (Description / Author / Upvote)

", "messages": [ { "id": "600490", "postDate": "08/16/2019 07:12:24", "content": "

오늘 대회 마지막 날이네요.

\n\n

대회가 끝나고도 열심히 공부하실 분들을 위해 커널들을 카테고리별로 정리해보았습니다.\n현재까지 올라온 모든 커널들을 검토하여 커널의 완성도, 기여도, 인기도 등을 고려하여 선정하였습니다.

\n\n

마지막까지 화이팅입니다!

\n\n
    \n

  1. Baseline\n[3rd ML Month] Car Model Classification Baseline (Keras / TaeJin Kim / 75+) - official\nPytorch Baseline (updated 7.08) (Pytorch / saewonYang / 50+)\n[3rd ML Month] Xception, StratifiedKFold, Ensemble (Keras / Jang / 35+)\nXception Model in Keras for Car Classification (Keras / Chanran Kim / 24+)\n[ka.kr sillim] Fast.ai (High Level Framework) (Fast.ai / HoMin / 16+)\n[KaKR-2019-3rd]EDA, ImagePrep, Mixup, CV (Keras) (Keras / Bruce Kim / 7+)

    2. \n

  2. EDA\n2019 3rd ML month with KaKR - EDA (EDA / Hyeonho Lee / 23+)\nWant to see my Augmentations (pytorch) (EDA / saewonYang / 14+)\n[ka.kr sillim]Data_check (EDA / HoMin / 11+)\n[3rd ML Month] EDA (EDA / Taegwan Kim / 4+)

    3. \n

  3. Preprocessing\n[3rd ML Month] Car Image Cropping - updated 7.10 (Bounding box / Taemyung Heo / 45+)\nCutout Augmentation (on Keras EfficientNet) (Cutout Augmentation / Chanran Kim / 25+)\n[3rd ML Month] Car Image Segmentation Crop (Segmentation / Steve Jang / 14+)\n[ka.kr sillim] AutoAugmentation (AutoAugmentation / HoMin / 12+)

    4. \n

  4. Training\nHow to split folds and train (Cross Validation / saewonYang / 14+)\nKeras, How to use pretrained model? (Pretrained model / Daehun Gwak / 14+)\nKeras Semi-supervised Learning (Semi-supervised learning / Taegwan Kim / 9+)

    5. \n

  5. Etc\nA Race for 2019 3rd ML month with KaKR (Public score Trend / Taemyung Heo / 15+)

    6. \n
\n\n

*Kernel name (Description / Author / Upvote)

", "rawMarkdown": "오늘 대회 마지막 날이네요.\n\n대회가 끝나고도 열심히 공부하실 분들을 위해 커널들을 카테고리별로 정리해보았습니다.\n현재까지 올라온 모든 커널들을 검토하여 커널의 완성도, 기여도, 인기도 등을 고려하여 선정하였습니다.\n\n마지막까지 화이팅입니다!\n\n1. Baseline\n[[3rd ML Month] Car Model Classification Baseline](https://www.kaggle.com/fulrose/3rd-ml-month-car-model-classification-baseline) (Keras / TaeJin Kim / 75+) - official\n[Pytorch Baseline (updated 7.08)](https://www.kaggle.com/yangsaewon/pytorch-baseline-updated-7-10) (Pytorch / saewonYang / 50+)\n[[3rd ML Month] Xception, StratifiedKFold, Ensemble](https://www.kaggle.com/janged/3rd-ml-month-xception-stratifiedkfold-ensemble) (Keras / Jang / 35+)\n[Xception Model in Keras for Car Classification](https://www.kaggle.com/seriousran/what-0-95121-model-failed) (Keras / Chanran Kim / 24+)\n[[ka.kr sillim] Fast.ai (High Level Framework)](https://www.kaggle.com/hominlee/ka-kr-sillim-fast-ai-high-level-framework) (Fast.ai / HoMin / 16+)\n[[KaKR-2019-3rd]EDA, ImagePrep, Mixup, CV (Keras)](https://www.kaggle.com/devbruce/kakr-2019-3rd-eda-imageprep-mixup-cv-keras) (Keras / Bruce Kim / 7+)\n\n2. EDA\n[2019 3rd ML month with KaKR - EDA](https://www.kaggle.com/hyeonho/2019-3rd-ml-month-with-kakr-eda) (EDA / Hyeonho Lee / 23+)\n[Want to see my Augmentations (pytorch)](https://www.kaggle.com/yangsaewon/want-to-see-my-augmentations-pytorch) (EDA / saewonYang / 14+)\n[[ka.kr sillim]Data_check](https://www.kaggle.com/hominlee/ka-kr-sillim-data-check) (EDA / HoMin / 11+)\n[[3rd ML Month] EDA](https://www.kaggle.com/kimtaegwan/3rd-ml-month-eda) (EDA / Taegwan Kim / 4+)\n\n3. Preprocessing\n[[3rd ML Month] Car Image Cropping - updated 7.10](https://www.kaggle.com/tmheo74/3rd-ml-month-car-image-cropping-updated-7-10) (Bounding box / Taemyung Heo / 45+)\n[Cutout Augmentation (on Keras EfficientNet)](https://www.kaggle.com/seriousran/cutout-augmentation-on-keras-efficientnet) (Cutout Augmentation / Chanran Kim / 25+)\n[[3rd ML Month] Car Image Segmentation Crop](https://www.kaggle.com/cruiserx/3rd-ml-month-car-image-segmentation-crop) (Segmentation / Steve Jang / 14+)\n[[ka.kr sillim] AutoAugmentation](https://www.kaggle.com/hominlee/ka-kr-sillim-autoaugmentation) (AutoAugmentation / HoMin / 12+)\n\n4. Training\n[How to split folds and train](https://www.kaggle.com/yangsaewon/how-to-split-folds-and-train) (Cross Validation / saewonYang / 14+)\n[Keras, How to use pretrained model?](https://www.kaggle.com/daehungwak/keras-how-to-use-pretrained-model) (Pretrained model / Daehun Gwak / 14+)\n[Keras Semi-supervised Learning](https://www.kaggle.com/kimtaegwan/keras-pseudo-labelling-0-941) (Semi-supervised learning / Taegwan Kim / 9+)\n\n5. Etc\n[A Race for 2019 3rd ML month with KaKR](https://www.kaggle.com/tmheo74/a-race-for-2019-3rd-ml-month-with-kakr) (Public score Trend / Taemyung Heo / 15+)\n\n**Kernel name (Description / Author / Upvote)*", "votes": null }, { "id": "600544", "postDate": "08/16/2019 08:47:53", "content": "

오~ 정성스런 리스트업 감사합니다.\n커널 찾기 훨씬 편하네요!

", "rawMarkdown": "오~ 정성스런 리스트업 감사합니다.\n커널 찾기 훨씬 편하네요!", "votes": null }, { "id": "600549", "postDate": "08/16/2019 09:03:18", "content": "

아직 실력이 많이 부족해서 할 수 있는 기여가 이정도네요..ㅎ

", "rawMarkdown": "아직 실력이 많이 부족해서 할 수 있는 기여가 이정도네요..ㅎ", "votes": null }, { "id": "614730", "postDate": "09/01/2019 01:09:07", "content": "

와! 감사합니다.

", "rawMarkdown": "와! 감사합니다.", "votes": null } ], "comments": [ { "id": 600544, "author_name": "tmheo74", "author_url": "", "post_date": "08/16/2019 08:47:53", "content": "

오~ 정성스런 리스트업 감사합니다.\n커널 찾기 훨씬 편하네요!

", "votes": null, "replies": [ { "id": 600549, "author_name": "kimjihoo", "author_url": "", "post_date": "08/16/2019 09:03:18", "content": "

아직 실력이 많이 부족해서 할 수 있는 기여가 이정도네요..ㅎ

", "votes": null, "replies": [] } ] }, { "id": 614730, "author_name": "yslee04", "author_url": "", "post_date": "09/01/2019 01:09:07", "content": "

와! 감사합니다.

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "600490": "오늘 대회 마지막 날이네요.\n\n대회가 끝나고도 열심히 공부하실 분들을 위해 커널들을 카테고리별로 정리해보았습니다.\n현재까지 올라온 모든 커널들을 검토하여 커널의 완성도, 기여도, 인기도 등을 고려하여 선정하였습니다.\n\n마지막까지 화이팅입니다!\n\n1. Baseline\n[[3rd ML Month] Car Model Classification Baseline](https://www.kaggle.com/fulrose/3rd-ml-month-car-model-classification-baseline) (Keras / TaeJin Kim / 75+) - official\n[Pytorch Baseline (updated 7.08)](https://www.kaggle.com/yangsaewon/pytorch-baseline-updated-7-10) (Pytorch / saewonYang / 50+)\n[[3rd ML Month] Xception, StratifiedKFold, Ensemble](https://www.kaggle.com/janged/3rd-ml-month-xception-stratifiedkfold-ensemble) (Keras / Jang / 35+)\n[Xception Model in Keras for Car Classification](https://www.kaggle.com/seriousran/what-0-95121-model-failed) (Keras / Chanran Kim / 24+)\n[[ka.kr sillim] Fast.ai (High Level Framework)](https://www.kaggle.com/hominlee/ka-kr-sillim-fast-ai-high-level-framework) (Fast.ai / HoMin / 16+)\n[[KaKR-2019-3rd]EDA, ImagePrep, Mixup, CV (Keras)](https://www.kaggle.com/devbruce/kakr-2019-3rd-eda-imageprep-mixup-cv-keras) (Keras / Bruce Kim / 7+)\n\n2. EDA\n[2019 3rd ML month with KaKR - EDA](https://www.kaggle.com/hyeonho/2019-3rd-ml-month-with-kakr-eda) (EDA / Hyeonho Lee / 23+)\n[Want to see my Augmentations (pytorch)](https://www.kaggle.com/yangsaewon/want-to-see-my-augmentations-pytorch) (EDA / saewonYang / 14+)\n[[ka.kr sillim]Data_check](https://www.kaggle.com/hominlee/ka-kr-sillim-data-check) (EDA / HoMin / 11+)\n[[3rd ML Month] EDA](https://www.kaggle.com/kimtaegwan/3rd-ml-month-eda) (EDA / Taegwan Kim / 4+)\n\n3. Preprocessing\n[[3rd ML Month] Car Image Cropping - updated 7.10](https://www.kaggle.com/tmheo74/3rd-ml-month-car-image-cropping-updated-7-10) (Bounding box / Taemyung Heo / 45+)\n[Cutout Augmentation (on Keras EfficientNet)](https://www.kaggle.com/seriousran/cutout-augmentation-on-keras-efficientnet) (Cutout Augmentation / Chanran Kim / 25+)\n[[3rd ML Month] Car Image Segmentation Crop](https://www.kaggle.com/cruiserx/3rd-ml-month-car-image-segmentation-crop) (Segmentation / Steve Jang / 14+)\n[[ka.kr sillim] AutoAugmentation](https://www.kaggle.com/hominlee/ka-kr-sillim-autoaugmentation) (AutoAugmentation / HoMin / 12+)\n\n4. Training\n[How to split folds and train](https://www.kaggle.com/yangsaewon/how-to-split-folds-and-train) (Cross Validation / saewonYang / 14+)\n[Keras, How to use pretrained model?](https://www.kaggle.com/daehungwak/keras-how-to-use-pretrained-model) (Pretrained model / Daehun Gwak / 14+)\n[Keras Semi-supervised Learning](https://www.kaggle.com/kimtaegwan/keras-pseudo-labelling-0-941) (Semi-supervised learning / Taegwan Kim / 9+)\n\n5. Etc\n[A Race for 2019 3rd ML month with KaKR](https://www.kaggle.com/tmheo74/a-race-for-2019-3rd-ml-month-with-kakr) (Public score Trend / Taemyung Heo / 15+)\n\n**Kernel name (Description / Author / Upvote)*", "600544": "오~ 정성스런 리스트업 감사합니다.\n커널 찾기 훨씬 편하네요!", "600549": "아직 실력이 많이 부족해서 할 수 있는 기여가 이정도네요..ㅎ", "614730": "와! 감사합니다." }, "source": "meta" }