Hi all - thanks for your patience! The host team has updated some of the test labels. We've also fixed an issue that impacted the scoring of image-level labels - thanks very much for all of your feedback. We're going to begin the process of updating the leaderboard now. Note that this will take some time, as we will be re-running all submissions.
\nUpdate: the leaderboard update should now be complete.
", "messages": [ { "id": 1349453, "postDate": "2021-06-14T19:26:08.517Z", "content": "Hi all - thanks for your patience! The host team has updated some of the test labels. We've also fixed an issue that impacted the scoring of image-level labels - thanks very much for all of your feedback. We're going to begin the process of updating the leaderboard now. Note that this will take some time, as we will be re-running all submissions.
\nUpdate: the leaderboard update should now be complete.
", "rawMarkdown": "Hi all - thanks for your patience! The host team has updated some of the test labels. We've also fixed an issue that impacted the scoring of image-level labels - thanks very much for all of your feedback. We're going to begin the process of updating the leaderboard now. Note that this will take some time, as we will be re-running all submissions.\n\nUpdate: the leaderboard update should now be complete.", "votes": 37 }, { "id": 1349561, "postDate": "2021-06-14T22:37:57.337Z", "content": "The host team has updated some of the test labels\nAre there any changes to the public test? I would like to know if we need to recreate the png/jpg data.
", "rawMarkdown": "```\nThe host team has updated some of the test labels\n```\nAre there any changes to the public test? I would like to know if we need to recreate the png/jpg data.\n", "votes": 6, "replies": [ { "id": 1350663, "postDate": "2021-06-15T16:26:26.433Z", "content": "Yes, label changes were made spanning the entire test set (public and private). The images, however were not changed. Just the labels corresponding to the existing population of images.
", "rawMarkdown": "Yes, label changes were made spanning the entire test set (public and private). The images, however were not changed. Just the labels corresponding to the existing population of images.", "votes": 3 }, { "id": 1353767, "postDate": "2021-06-17T08:31:16.983Z", "content": "Hi @juliaelliott can I just confirm please:
\nHence no need to re-download any dataset, is this correct?
", "rawMarkdown": "Hi @juliaelliott can I just confirm please:\n\n- No changes to any image data or other metadata in DICOM files\n- No changes to any training data labels or bounding boxes\n\nHence no need to re-download any dataset, is this correct?" }, { "id": 1354570, "postDate": "2021-06-17T17:04:13.460Z", "content": "@benvenutto Correct.
", "rawMarkdown": "@benvenutto Correct. ", "votes": 1 } ] }, { "id": 1349499, "postDate": "2021-06-14T20:34:39.370Z", "content": "Will there be any Deadline Extension? @philculliton
", "rawMarkdown": "Will there be any **Deadline Extension**? @philculliton ", "votes": 4, "replies": [ { "id": 1349546, "postDate": "2021-06-14T21:50:11.677Z", "content": "We still have two months to go. Everyone is working on study level analysis, so I don't think we need to extend it.
", "rawMarkdown": "We still have two months to go. Everyone is working on study level analysis, so I don't think we need to extend it.", "votes": 1 }, { "id": 1349594, "postDate": "2021-06-14T23:58:33.623Z", "content": "yes, I feel the same way.
", "rawMarkdown": "yes, I feel the same way.", "votes": 1 }, { "id": 1349596, "postDate": "2021-06-15T00:00:38.527Z", "content": "haha I got downvoted 😂
", "rawMarkdown": "haha I got downvoted 😂", "votes": 3 }, { "id": 1349674, "postDate": "2021-06-15T02:43:24.287Z", "content": "lol upvoted 😌
", "rawMarkdown": "lol upvoted 😌", "votes": 1 }, { "id": 1350654, "postDate": "2021-06-15T16:22:35.647Z", "content": "@awsaf49 No, there will not.
", "rawMarkdown": "@awsaf49 No, there will not.", "votes": 2 }, { "id": 1350727, "postDate": "2021-06-15T17:25:35.700Z", "content": "thanks @juliaelliott for the clarification
", "rawMarkdown": "thanks @juliaelliott for the clarification" } ] }, { "id": 1350653, "postDate": "2021-06-15T16:22:13.937Z", "content": "Note to all that this is expected to take a number of days, due to the volume of submissions that need to be run through again. We will be updating this post when it is complete. Thank you for your patience.
\nAlso, since we've made the backend updates necessary, that means all net new submissions will have accurate scoring computed.
", "rawMarkdown": "Note to all that this is expected to take a number of days, due to the volume of submissions that need to be run through again. We will be updating this post when it is complete. Thank you for your patience.\n\nAlso, since we've made the backend updates necessary, that means all net new submissions will have accurate scoring computed.", "votes": 4 }, { "id": 1349629, "postDate": "2021-06-15T01:51:20.547Z", "content": "It's a timely update. One more question: When can the specific mAP calculation method of image-level and study-level samples be publicized?
", "rawMarkdown": "It's a timely update. One more question: When can the specific mAP calculation method of image-level and study-level samples be publicized? ", "votes": 4 }, { "id": 1359492, "postDate": "2021-06-21T10:20:35.573Z", "content": "I'm sorry if I'm wrong. Is the Submission Score calculation method back to the previous one now?
", "rawMarkdown": "I'm sorry if I'm wrong. Is the Submission Score calculation method back to the previous one now? ", "votes": 1 }, { "id": 1349481, "postDate": "2021-06-14T20:07:36.597Z", "content": "any change on training data ? @philculliton
", "rawMarkdown": "any change on **training** data ? @philculliton ", "votes": 2, "replies": [ { "id": 1349491, "postDate": "2021-06-14T20:22:41.553Z", "content": "Hi! No, the training data hasn't changed.
", "rawMarkdown": "Hi! No, the training data hasn't changed.", "votes": 1 }, { "id": 1349495, "postDate": "2021-06-14T20:27:05.300Z", "content": "according to this thread, I think there are some duplicates on the training data. So, does it mean those duplicates are part of the competition?
", "rawMarkdown": "according to this [thread](https://www.kaggle.com/c/siim-covid19-detection/discussion/240250#1322940), I think there are some duplicates on the **training** data. So, does it mean those duplicates are part of the competition?", "votes": 1 }, { "id": 1349498, "postDate": "2021-06-14T20:34:26.610Z", "content": "i think the main concern is the below
\nParasLakhaniCompetition Host • 20 days ago • Options • Report • Reply
\n\"We are now aware regarding some studies having multiple images (some are duplicates), some with bounding boxes, and others with not. We are investigating different approaches to remedy this issue. We will soon follow-up when we arrive at a solution and let you know.\"
", "rawMarkdown": "i think the main concern is the below\n\n---\n\nParasLakhaniCompetition Host • 20 days ago • Options • Report • Reply\n\n\n\"We are now aware regarding some studies having multiple images (some are duplicates), some with bounding boxes, and others with not. We are investigating different approaches to remedy this issue. We will soon follow-up when we arrive at a solution and let you know.\"", "votes": 6 }, { "id": 1350657, "postDate": "2021-06-15T16:24:02.440Z", "content": "The population of images have not changed. They are all remaining part of the competition (including duplicates), but their corresponding labels have been corrected.
", "rawMarkdown": "The population of images have not changed. They are all remaining part of the competition (including duplicates), but their corresponding labels have been corrected." }, { "id": 1350725, "postDate": "2021-06-15T17:25:15.530Z", "content": "@juliaelliott does it mean train labels have changed too ?
", "rawMarkdown": "@juliaelliott does it mean **train labels** have changed too ?", "votes": 2 }, { "id": 1353218, "postDate": "2021-06-17T01:52:16.927Z", "content": "just diff'ed old and current training labels. no difference…
", "rawMarkdown": "just diff'ed old and current training labels. no difference..." } ] }, { "id": 1356361, "postDate": "2021-06-19T00:51:17.727Z", "content": "Wanted to note here that the update should now be complete.
", "rawMarkdown": "Wanted to note here that the update should now be complete." }, { "id": 1349649, "postDate": "2021-06-15T02:18:16.990Z", "content": "Thanks for the update @philculliton. Will we be notified once the rerun has finished?
", "rawMarkdown": "Thanks for the update @philculliton. Will we be notified once the rerun has finished?", "replies": [ { "id": 1349670, "postDate": "2021-06-15T02:38:08.347Z", "content": "All my score of submissions so far have been updated, so I don’t think there will be any notice.
", "rawMarkdown": "All my score of submissions so far have been updated, so I don’t think there will be any notice." } ] }, { "id": 1358949, "postDate": "2021-06-20T23:38:25.317Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 1349561, "author_name": "YujiAriyasu", "author_url": "", "post_date": "2021-06-14T22:37:57.337000", "content": "The host team has updated some of the test labels\nAre there any changes to the public test? I would like to know if we need to recreate the png/jpg data.
", "votes": 6, "replies": [ { "id": 1350663, "author_name": "Julia Elliott", "author_url": "", "post_date": "2021-06-15T16:26:26.433000", "content": "Yes, label changes were made spanning the entire test set (public and private). The images, however were not changed. Just the labels corresponding to the existing population of images.
", "votes": 3, "replies": [] }, { "id": 1353767, "author_name": "benvens", "author_url": "", "post_date": "2021-06-17T08:31:16.983000", "content": "Hi @juliaelliott can I just confirm please:
\nHence no need to re-download any dataset, is this correct?
", "votes": 0, "replies": [] }, { "id": 1354570, "author_name": "Julia Elliott", "author_url": "", "post_date": "2021-06-17T17:04:13.460000", "content": "@benvenutto Correct.
", "votes": 1, "replies": [] } ] }, { "id": 1349499, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-14T20:34:39.370000", "content": "Will there be any Deadline Extension? @philculliton
", "votes": 4, "replies": [ { "id": 1349546, "author_name": "YujiAriyasu", "author_url": "", "post_date": "2021-06-14T21:50:11.677000", "content": "We still have two months to go. Everyone is working on study level analysis, so I don't think we need to extend it.
", "votes": 1, "replies": [] }, { "id": 1349594, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-14T23:58:33.623000", "content": "yes, I feel the same way.
", "votes": 1, "replies": [] }, { "id": 1349596, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-15T00:00:38.527000", "content": "haha I got downvoted 😂
", "votes": 3, "replies": [] }, { "id": 1349674, "author_name": "YujiAriyasu", "author_url": "", "post_date": "2021-06-15T02:43:24.287000", "content": "lol upvoted 😌
", "votes": 1, "replies": [] }, { "id": 1350654, "author_name": "Julia Elliott", "author_url": "", "post_date": "2021-06-15T16:22:35.647000", "content": "@awsaf49 No, there will not.
", "votes": 2, "replies": [] }, { "id": 1350727, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-15T17:25:35.700000", "content": "thanks @juliaelliott for the clarification
", "votes": 0, "replies": [] } ] }, { "id": 1350653, "author_name": "Julia Elliott", "author_url": "", "post_date": "2021-06-15T16:22:13.937000", "content": "Note to all that this is expected to take a number of days, due to the volume of submissions that need to be run through again. We will be updating this post when it is complete. Thank you for your patience.
\nAlso, since we've made the backend updates necessary, that means all net new submissions will have accurate scoring computed.
", "votes": 4, "replies": [] }, { "id": 1349629, "author_name": "README", "author_url": "", "post_date": "2021-06-15T01:51:20.547000", "content": "It's a timely update. One more question: When can the specific mAP calculation method of image-level and study-level samples be publicized?
", "votes": 4, "replies": [] }, { "id": 1359492, "author_name": "Tanrei(nama)", "author_url": "", "post_date": "2021-06-21T10:20:35.573000", "content": "I'm sorry if I'm wrong. Is the Submission Score calculation method back to the previous one now?
", "votes": 1, "replies": [] }, { "id": 1349481, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-14T20:07:36.597000", "content": "any change on training data ? @philculliton
", "votes": 2, "replies": [ { "id": 1349491, "author_name": "Phil Culliton", "author_url": "", "post_date": "2021-06-14T20:22:41.553000", "content": "Hi! No, the training data hasn't changed.
", "votes": 1, "replies": [] }, { "id": 1349495, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-14T20:27:05.300000", "content": "according to this thread, I think there are some duplicates on the training data. So, does it mean those duplicates are part of the competition?
", "votes": 1, "replies": [] }, { "id": 1349498, "author_name": "hengck23", "author_url": "", "post_date": "2021-06-14T20:34:26.610000", "content": "i think the main concern is the below
\nParasLakhaniCompetition Host • 20 days ago • Options • Report • Reply
\n\"We are now aware regarding some studies having multiple images (some are duplicates), some with bounding boxes, and others with not. We are investigating different approaches to remedy this issue. We will soon follow-up when we arrive at a solution and let you know.\"
", "votes": 6, "replies": [] }, { "id": 1350657, "author_name": "Julia Elliott", "author_url": "", "post_date": "2021-06-15T16:24:02.440000", "content": "The population of images have not changed. They are all remaining part of the competition (including duplicates), but their corresponding labels have been corrected.
", "votes": 0, "replies": [] }, { "id": 1350725, "author_name": "Awsaf", "author_url": "", "post_date": "2021-06-15T17:25:15.530000", "content": "@juliaelliott does it mean train labels have changed too ?
", "votes": 2, "replies": [] }, { "id": 1353218, "author_name": "human intelligence", "author_url": "", "post_date": "2021-06-17T01:52:16.927000", "content": "just diff'ed old and current training labels. no difference…
", "votes": 0, "replies": [] } ] }, { "id": 1356361, "author_name": "Phil Culliton", "author_url": "", "post_date": "2021-06-19T00:51:17.727000", "content": "Wanted to note here that the update should now be complete.
", "votes": 0, "replies": [] }, { "id": 1349649, "author_name": "Stephen Hudson", "author_url": "", "post_date": "2021-06-15T02:18:16.990000", "content": "Thanks for the update @philculliton. Will we be notified once the rerun has finished?
", "votes": 0, "replies": [ { "id": 1349670, "author_name": "README", "author_url": "", "post_date": "2021-06-15T02:38:08.347000", "content": "All my score of submissions so far have been updated, so I don’t think there will be any notice.
", "votes": 0, "replies": [] } ] }, { "id": 1358949, "author_name": "", "author_url": "", "post_date": "2021-06-20T23:38:25.317000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1349453": "Hi all - thanks for your patience! The host team has updated some of the test labels. We've also fixed an issue that impacted the scoring of image-level labels - thanks very much for all of your feedback. We're going to begin the process of updating the leaderboard now. Note that this will take some time, as we will be re-running all submissions.\n\nUpdate: the leaderboard update should now be complete.", "1349561": "```\nThe host team has updated some of the test labels\n```\nAre there any changes to the public test? I would like to know if we need to recreate the png/jpg data.\n", "1349499": "Will there be any **Deadline Extension**? @philculliton ", "1350653": "Note to all that this is expected to take a number of days, due to the volume of submissions that need to be run through again. We will be updating this post when it is complete. Thank you for your patience.\n\nAlso, since we've made the backend updates necessary, that means all net new submissions will have accurate scoring computed.", "1349629": "It's a timely update. One more question: When can the specific mAP calculation method of image-level and study-level samples be publicized? ", "1359492": "I'm sorry if I'm wrong. Is the Submission Score calculation method back to the previous one now? ", "1349481": "any change on **training** data ? @philculliton ", "1356361": "Wanted to note here that the update should now be complete.", "1349649": "Thanks for the update @philculliton. Will we be notified once the rerun has finished?", "1358949": "" } }