Thanks for all your feedback, everybody, it was extremely helpful.
\n\nThe API should work now for download competition data - you just need to update your kaggle library as described [here] (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109520#latest-630128). Let us know if you have any further problems!
Sorry, Phil, my message was too cryptic.\nAs a workaround of the Kaggle API giving me 429s, I used the \"Download All\" link , with all the cookies in a curl call, and it only downloads stage_1_train_images.zip
\n\nIt is not urgent just to let you know.
", "rawMarkdown": "Sorry, Phil, my message was too cryptic.\nAs a workaround of the Kaggle API giving me 429s, I used the \"Download All\" link , with all the cookies in a curl call, and it only downloads stage_1_train_images.zip\n\nIt is not urgent just to let you know.\n" }, { "id": 629290, "postDate": "2019-09-18T15:58:48.317Z", "content": "Thanks for the reports! The Kaggle API is currently not working correctly with our downloads. Please use the Download All button to grab the dataset.
", "rawMarkdown": "Thanks for the reports! The Kaggle API is currently not working correctly with our downloads. Please use the Download All button to grab the dataset.", "replies": [ { "id": 629323, "postDate": "2019-09-18T16:25:01.130Z", "content": "I used download all to grab the dataset (which is frustrating to be unable to do on a remote machine even copying cookies and trying to use wget with those cookies). Unzipping the 139GB file (149GiB), there seems to be just be 674,258 dicoms -- which seem to only correspond to the training set. Is the test set available as a zip somewhere as well?
\n\nAlso it would be nice if it would be possible to download both through the kaggle API for training on headless servers (without having to reupload).
", "rawMarkdown": "I used download all to grab the dataset (which is frustrating to be unable to do on a remote machine even copying cookies and trying to use wget with those cookies). Unzipping the 139GB file (149GiB), there seems to be just be 674,258 dicoms -- which seem to only correspond to the training set. Is the test set available as a zip somewhere as well?\n\nAlso it would be nice if it would be possible to download both through the kaggle API for training on headless servers (without having to reupload)." } ] }, { "id": 629255, "postDate": "2019-09-18T15:11:29.333Z", "content": "Same here using kaggle API
", "rawMarkdown": "Same here using kaggle API" } ], "comments": [ { "id": 630139, "author_name": "Phil Culliton", "author_url": "", "post_date": "2019-09-19T19:30:26.170000", "content": "Thanks for all your feedback, everybody, it was extremely helpful.
\n\nThe API should work now for download competition data - you just need to update your kaggle library as described [here] (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109520#latest-630128). Let us know if you have any further problems!
Sorry, Phil, my message was too cryptic.\nAs a workaround of the Kaggle API giving me 429s, I used the \"Download All\" link , with all the cookies in a curl call, and it only downloads stage_1_train_images.zip
\n\nIt is not urgent just to let you know.
", "votes": 0, "replies": [] }, { "id": 629290, "author_name": "Phil Culliton", "author_url": "", "post_date": "2019-09-18T15:58:48.317000", "content": "Thanks for the reports! The Kaggle API is currently not working correctly with our downloads. Please use the Download All button to grab the dataset.
", "votes": 0, "replies": [ { "id": 629323, "author_name": "jledoux", "author_url": "", "post_date": "2019-09-18T16:25:01.130000", "content": "I used download all to grab the dataset (which is frustrating to be unable to do on a remote machine even copying cookies and trying to use wget with those cookies). Unzipping the 139GB file (149GiB), there seems to be just be 674,258 dicoms -- which seem to only correspond to the training set. Is the test set available as a zip somewhere as well?
\n\nAlso it would be nice if it would be possible to download both through the kaggle API for training on headless servers (without having to reupload).
", "votes": 0, "replies": [] } ] }, { "id": 629255, "author_name": "Marc Mauri", "author_url": "", "post_date": "2019-09-18T15:11:29.333000", "content": "Same here using kaggle API
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "629079": "", "630139": "Thanks for all your feedback, everybody, it was extremely helpful.\n\nThe API should work now for download competition data - you just need to update your `kaggle` library as described [here] (https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109520#latest-630128). Let us know if you have any further problems!", "629359": "Sorry, Phil, my message was too cryptic.\nAs a workaround of the Kaggle API giving me 429s, I used the \"Download All\" link , with all the cookies in a curl call, and it only downloads stage_1_train_images.zip\n\nIt is not urgent just to let you know.\n", "629290": "Thanks for the reports! The Kaggle API is currently not working correctly with our downloads. Please use the Download All button to grab the dataset.", "629255": "Same here using kaggle API" } }