![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F610120%2F40a3002e4d3317f3d362c19a98f1dc36%2Fhakuna_matata.gif?generation=1580675997040846&alt=media\")
HI all !!
\n\nI would like to introduce Kornia: a computer vision library for PyTorch.\nPlease, visit our project website: www.kornia.org
\n\nWe recently released v0.2.0, introducing a new data augmentation API that mimics the best of the existing data augmentation frameworks such torchvision or albumentations all reimplemented assuming as input torch.Tensor data structures that will allowing to run the standard transformations (geometric and color) in batch mode in the GPU and backprop through it.
This is a good opportunity to make use of it in order to perform efficient data augmentation on videos. We provide an example kernel with a small example about it usage
\n\n[*] https://www.kaggle.com/edgarriba/kornia-data-augmentation
", "messages": [ { "id": "735294", "postDate": "02/02/2020 20:43:23", "content": "HI all !!
\n\nI would like to introduce Kornia: a computer vision library for PyTorch.\nPlease, visit our project website: www.kornia.org
\n\nWe recently released v0.2.0, introducing a new data augmentation API that mimics the best of the existing data augmentation frameworks such torchvision or albumentations all reimplemented assuming as input torch.Tensor data structures that will allowing to run the standard transformations (geometric and color) in batch mode in the GPU and backprop through it.
This is a good opportunity to make use of it in order to perform efficient data augmentation on videos. We provide an example kernel with a small example about it usage
\n\n[*] https://www.kaggle.com/edgarriba/kornia-data-augmentation
", "rawMarkdown": "HI all !!\n\nI would like to introduce Kornia: a computer vision library for PyTorch.\nPlease, visit our project website: www.kornia.org\n\nWe recently released v0.2.0, introducing a new data augmentation API that mimics the best of the existing data augmentation frameworks such `torchvision` or `albumentations` all reimplemented assuming as input torch.Tensor data structures that will allowing to run the standard transformations (geometric and color) in batch mode in the GPU and backprop through it.\n\n\n\nThis is a good opportunity to make use of it in order to perform efficient data augmentation on videos. We provide an example kernel with a small example about it usage\n\n[*] https://www.kaggle.com/edgarriba/kornia-data-augmentation", "votes": null }, { "id": "735345", "postDate": "02/02/2020 22:49:30", "content": "Fork of your example fails at the first line -
", "rawMarkdown": "Fork of your example fails at the first line -", "votes": null }, { "id": "735353", "postDate": "02/02/2020 22:54:05", "content": "Link to your web site returns to this page.
", "rawMarkdown": "Link to your web site returns to this page.", "votes": null }, { "id": "735618", "postDate": "02/03/2020 09:00:35", "content": "fixed
", "rawMarkdown": "fixed", "votes": null }, { "id": "735620", "postDate": "02/03/2020 09:01:00", "content": "will investigate that. And I'll do some improvements
", "rawMarkdown": "will investigate that. And I'll do some improvements", "votes": null }, { "id": "1010960", "postDate": "09/15/2020 06:52:47", "content": "Have you got any speed comparisons with albumentations?
", "rawMarkdown": "Have you got any speed comparisons with albumentations?", "votes": null } ], "comments": [ { "id": 1010960, "author_name": "sachin", "author_url": "", "post_date": "09/15/2020 06:52:47", "content": "Have you got any speed comparisons with albumentations?
", "votes": null, "replies": [] }, { "id": 735345, "author_name": "pcjimmmy", "author_url": "", "post_date": "02/02/2020 22:49:30", "content": "Fork of your example fails at the first line -
", "votes": null, "replies": [ { "id": 735620, "author_name": "edgarriba", "author_url": "", "post_date": "02/03/2020 09:01:00", "content": "will investigate that. And I'll do some improvements
", "votes": null, "replies": [] } ] }, { "id": 735353, "author_name": "pcjimmmy", "author_url": "", "post_date": "02/02/2020 22:54:05", "content": "Link to your web site returns to this page.
", "votes": null, "replies": [ { "id": 735618, "author_name": "edgarriba", "author_url": "", "post_date": "02/03/2020 09:00:35", "content": "fixed
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "735294": "HI all !!\n\nI would like to introduce Kornia: a computer vision library for PyTorch.\nPlease, visit our project website: www.kornia.org\n\nWe recently released v0.2.0, introducing a new data augmentation API that mimics the best of the existing data augmentation frameworks such `torchvision` or `albumentations` all reimplemented assuming as input torch.Tensor data structures that will allowing to run the standard transformations (geometric and color) in batch mode in the GPU and backprop through it.\n\n\n\nThis is a good opportunity to make use of it in order to perform efficient data augmentation on videos. We provide an example kernel with a small example about it usage\n\n[*] https://www.kaggle.com/edgarriba/kornia-data-augmentation", "735345": "Fork of your example fails at the first line -", "735353": "Link to your web site returns to this page.", "735618": "fixed", "735620": "will investigate that. And I'll do some improvements", "1010960": "Have you got any speed comparisons with albumentations?" }, "source": "meta" }