![\"Short](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F292812%2F014a1bcdd7712d0c1a60a7244ea4735b%2Fevents_dur.png?generation=1682522408247520&alt=media\")
First, I would like to thank the organizers for putting together this interesting competition.
\nI have a question about the data. I noticed that many events are quite short, even less than 1 second, with periods of no events in between. What do these events represent? Is it some data collection artifact, or are these real events, and what would that look like in real life? E.g., a person was turning normally, but then there was a half-second period of pathological turning and then normal again? Also, how accurate are the time stamps on these events? Was someone trying to pinpoint events to a tenth of a second, or is it more approximate?
", "messages": [ { "id": "2185141", "postDate": "03/16/2023 20:56:07", "content": "First, I would like to thank the organizers for putting together this interesting competition.
\nI have a question about the data. I noticed that many events are quite short, even less than 1 second, with periods of no events in between. What do these events represent? Is it some data collection artifact, or are these real events, and what would that look like in real life? E.g., a person was turning normally, but then there was a half-second period of pathological turning and then normal again? Also, how accurate are the time stamps on these events? Was someone trying to pinpoint events to a tenth of a second, or is it more approximate?
", "rawMarkdown": "First, I would like to thank the organizers for putting together this interesting competition. \n\nI have a question about the data. I noticed that many events are quite short, even less than 1 second, with periods of no events in between. What do these events represent? Is it some data collection artifact, or are these real events, and what would that look like in real life? E.g., a person was turning normally, but then there was a half-second period of pathological turning and then normal again? Also, how accurate are the time stamps on these events? Was someone trying to pinpoint events to a tenth of a second, or is it more approximate?", "votes": null }, { "id": "2200172", "postDate": "03/28/2023 10:46:12", "content": "I have the same question. Most of the short events are from defog.
", "rawMarkdown": "I have the same question. Most of the short events are from defog.", "votes": null }, { "id": "2234309", "postDate": "04/25/2023 05:22:08", "content": "@jeffhausdorff @ryanholbrook @erangazit
\nI have a similar question. In the event.csv file, there are events with very short durations. For example, there are 32 events in event.csv that last less than 0.1 seconds. Are these considered label noise? Or have they been accurately labeled using some method?
", "rawMarkdown": "jeffhausdorff @ryanholbrook @erangazit \n\nI have a similar question. In the event.csv file, there are events with very short durations. For example, there are 32 events in event.csv that last less than 0.1 seconds. Are these considered label noise? Or have they been accurately labeled using some method?", "votes": null }, { "id": "2236154", "postDate": "04/26/2023 15:22:08", "content": "@jeffhausdorff @ryanholbrook @erangazit
\nI have a similar question. Where is the objective minimum duration of FOG-events?
\n
![\"Short](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F292812%2F80e96dce8ddc3ac1c94add754b8d97b1%2Fevents_dur1.png?generation=1682522467277295&alt=media\")
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F292812%2Fdb16bb07879a0ea026e6f4fc29f0f958%2Fevents_dur2.png?generation=1682523711576193&alt=media\")
I have the same question. Most of the short events are from defog.
", "votes": null, "replies": [] }, { "id": 2234309, "author_name": "nomorevotch", "author_url": "", "post_date": "04/25/2023 05:22:08", "content": "@jeffhausdorff @ryanholbrook @erangazit
\nI have a similar question. In the event.csv file, there are events with very short durations. For example, there are 32 events in event.csv that last less than 0.1 seconds. Are these considered label noise? Or have they been accurately labeled using some method?
", "votes": null, "replies": [ { "id": 2236154, "author_name": "maatkara", "author_url": "", "post_date": "04/26/2023 15:22:08", "content": "@jeffhausdorff @ryanholbrook @erangazit
\nI have a similar question. Where is the objective minimum duration of FOG-events?
\n