As this competition comes to a close, I wanted to remind you of the \"TPU Star\" prizes that will be issued alongside the standard leaderboard prizes.
\nSubmit this form by February 28, 2021 to self-nominate for consideration. At a minimum, you must have shared a public notebook which uses Kaggle's TPU integration on this competition's dataset.
", "messages": [ { "id": "1207275", "postDate": "02/17/2021 19:21:02", "content": "As this competition comes to a close, I wanted to remind you of the \"TPU Star\" prizes that will be issued alongside the standard leaderboard prizes.
\nSubmit this form by February 28, 2021 to self-nominate for consideration. At a minimum, you must have shared a public notebook which uses Kaggle's TPU integration on this competition's dataset.
", "rawMarkdown": "As this competition comes to a close, I wanted to remind you of the [\"TPU Star\" prizes](https://www.kaggle.com/c/cassava-leaf-disease-classification/overview/prizes) that will be issued alongside the standard leaderboard prizes.\n\n[Submit this form](https://www.kaggle.com/tpu-star-submission-form) by **February 28, 2021** to self-nominate for consideration. At a minimum, you must have shared a public notebook which uses Kaggle's TPU integration on this competition's dataset.", "votes": null }, { "id": "1209835", "postDate": "02/19/2021 04:08:33", "content": "@tanlikesmath Should definitely be receiving the same as he always has a TPU kernel for the comp which is very generic, well written and easily understood as well. Plus the efforts are being made not just for this comp, but right from when pytorch-xla was there (wasn't stable or was supposed to work with PyTorch kernels at that point of time, but we opened issues at their github repo and got the help needed and a lot of different ideas as well to try out and build a better approach to leverage them in general) and kaggle introduced TPU's. He is one of the first curious person to play with them at length. (right from this comp)
\nOther works from the same person,
\nAnd there are more!
", "rawMarkdown": "tanlikesmath Should definitely be receiving the same as he always has a TPU kernel for the comp which is very generic, well written and easily understood as well. Plus the efforts are being made not just for this comp, but right from when pytorch-xla was there (wasn't stable or was supposed to work with PyTorch kernels at that point of time, but we opened issues at their github repo and got the help needed and a lot of different ideas as well to try out and build a better approach to leverage them in general) and kaggle introduced TPU's. He is one of the first curious person to play with them at length. (right from [this](https://www.kaggle.com/c/flower-classification-with-tpus/) comp)\n\nOther works from the same person,\n\n- https://www.kaggle.com/c/cassava-leaf-disease-classification/discussion/210537\n- https://www.kaggle.com/c/cassava-leaf-disease-classification/discussion/220628\n- https://www.kaggle.com/c/jigsaw-multilingual-toxic-comment-classification/discussion/143005\n- https://www.kaggle.com/c/jigsaw-multilingual-toxic-comment-classification/discussion/138511\n\nAnd there are more!", "votes": null } ], "comments": [ { "id": 1209835, "author_name": "adityaecdrid", "author_url": "", "post_date": "02/19/2021 04:08:33", "content": "@tanlikesmath Should definitely be receiving the same as he always has a TPU kernel for the comp which is very generic, well written and easily understood as well. Plus the efforts are being made not just for this comp, but right from when pytorch-xla was there (wasn't stable or was supposed to work with PyTorch kernels at that point of time, but we opened issues at their github repo and got the help needed and a lot of different ideas as well to try out and build a better approach to leverage them in general) and kaggle introduced TPU's. He is one of the first curious person to play with them at length. (right from this comp)
\nOther works from the same person,
\nAnd there are more!
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1207275": "As this competition comes to a close, I wanted to remind you of the [\"TPU Star\" prizes](https://www.kaggle.com/c/cassava-leaf-disease-classification/overview/prizes) that will be issued alongside the standard leaderboard prizes.\n\n[Submit this form](https://www.kaggle.com/tpu-star-submission-form) by **February 28, 2021** to self-nominate for consideration. At a minimum, you must have shared a public notebook which uses Kaggle's TPU integration on this competition's dataset.", "1209835": "tanlikesmath Should definitely be receiving the same as he always has a TPU kernel for the comp which is very generic, well written and easily understood as well. Plus the efforts are being made not just for this comp, but right from when pytorch-xla was there (wasn't stable or was supposed to work with PyTorch kernels at that point of time, but we opened issues at their github repo and got the help needed and a lot of different ideas as well to try out and build a better approach to leverage them in general) and kaggle introduced TPU's. He is one of the first curious person to play with them at length. (right from [this](https://www.kaggle.com/c/flower-classification-with-tpus/) comp)\n\nOther works from the same person,\n\n- https://www.kaggle.com/c/cassava-leaf-disease-classification/discussion/210537\n- https://www.kaggle.com/c/cassava-leaf-disease-classification/discussion/220628\n- https://www.kaggle.com/c/jigsaw-multilingual-toxic-comment-classification/discussion/143005\n- https://www.kaggle.com/c/jigsaw-multilingual-toxic-comment-classification/discussion/138511\n\nAnd there are more!" }, "source": "meta" }