{ "id": 452632, "title": "Enmo sensitivity (small measurements)", "url": "/competitions/child-mind-institute-detect-sleep-states/discussion/452632", "author_name": "", "post_date": "2023-11-02T22:04:07.640878500Z", "votes": 3, "comment_count": 1, "views": 0, "content": "

@gkiar07 there are some enmo readings that are extremely small but not zero, like in the pico and nano orders of magnitude. Maybe it's because of the device sensitivity. Do these measurements account for readings or can be simply interpreted as zero?

", "messages": [ { "id": "2510323", "postDate": "11/02/2023 22:04:07", "content": "

@gkiar07 there are some enmo readings that are extremely small but not zero, like in the pico and nano orders of magnitude. Maybe it's because of the device sensitivity. Do these measurements account for readings or can be simply interpreted as zero?

", "rawMarkdown": "gkiar07 there are some `enmo` readings that are extremely small but not zero, like in the pico and nano orders of magnitude. Maybe it's because of the device sensitivity. Do these measurements account for readings or can be simply interpreted as zero?", "votes": null }, { "id": "2514472", "postDate": "11/06/2023 10:34:38", "content": "

@alejopaullier : to tell the long story short, it is better to read such near-zero values for enmo as is. The higher values exceeding its 99th percentile should be clipped by a certain upper cap, in turn.

\n

You may want to review the respective EDA notebooks to see the long story told in visuals and calculations, per the links below

\n\n

I hope it is helpful.

", "rawMarkdown": "alejopaullier : to tell the long story short, it is better to read such near-zero values for *enmo* as is. The higher values exceeding its 99th percentile should be clipped by a certain upper cap, in turn.\n\nYou may want to review the respective EDA notebooks to see the long story told in visuals and calculations, per the links below\n- https://www.kaggle.com/code/veronikasarikaya/it-s-a-wonderful-wonderful-sleep-eda\n- https://www.kaggle.com/code/gvyshnya/to-sleep-or-not-to-sleep-deep-eda-dive\n\nI hope it is helpful.", "votes": null } ], "comments": [ { "id": 2514472, "author_name": "gvyshnya", "author_url": "", "post_date": "11/06/2023 10:34:38", "content": "

@alejopaullier : to tell the long story short, it is better to read such near-zero values for enmo as is. The higher values exceeding its 99th percentile should be clipped by a certain upper cap, in turn.

\n

You may want to review the respective EDA notebooks to see the long story told in visuals and calculations, per the links below

\n\n

I hope it is helpful.

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2510323": "gkiar07 there are some `enmo` readings that are extremely small but not zero, like in the pico and nano orders of magnitude. Maybe it's because of the device sensitivity. Do these measurements account for readings or can be simply interpreted as zero?", "2514472": "alejopaullier : to tell the long story short, it is better to read such near-zero values for *enmo* as is. The higher values exceeding its 99th percentile should be clipped by a certain upper cap, in turn.\n\nYou may want to review the respective EDA notebooks to see the long story told in visuals and calculations, per the links below\n- https://www.kaggle.com/code/veronikasarikaya/it-s-a-wonderful-wonderful-sleep-eda\n- https://www.kaggle.com/code/gvyshnya/to-sleep-or-not-to-sleep-deep-eda-dive\n\nI hope it is helpful." }, "source": "meta" }