![](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8386305%2Fac84f49ec42ae13c0951fbca43aa8284%2F2024-01-14%2019.50.13.png?generation=1705229597945476&alt=media\")
\nThank you all.
I am struggling considerably to interpret the data.
\n『The expert annotators reviewed 50 second long EEG samples plus matched spectrograms covering 10 a minute window centered at the same time and labeled the central 10 seconds.』
\nCould you please tell me the meaning of the above text?
Does it mean that when the annotator is observing spectrogram data for annotation, they are only using the central 10 seconds (295 sec to 305 sec) out of the 10 minutes window of data (Label_offset_second + 600 seconds)? (like below)
\n
\nThank you all.
I am struggling considerably to interpret the data.
\n『The expert annotators reviewed 50 second long EEG samples plus matched spectrograms covering 10 a minute window centered at the same time and labeled the central 10 seconds.』
\nCould you please tell me the meaning of the above text?
Does it mean that when the annotator is observing spectrogram data for annotation, they are only using the central 10 seconds (295 sec to 305 sec) out of the 10 minutes window of data (Label_offset_second + 600 seconds)? (like below)
\n
\nThank you all.
\n\nDoes it mean that when the annotator is observing spectrogram data for annotation, they are only using the central 10 seconds (295 sec to 305 sec) out of the 10 minutes window of data (Label_offset_second + 600 seconds)? (like below)
\n
Yes. Call the middle timestamp in seconds time=T. Then
Thank you for your reply.
\nSo [T-5:T+5],
\nDoes this mean that not only spectrogram data but also EEG data are involved?
Yes the labelers use both data. And both data are the same. A Spectrogram is just a visual representation of EEG data.
", "rawMarkdown": "Yes the labelers use both data. And both data are the same. A Spectrogram is just a visual representation of EEG data.", "votes": null }, { "id": "2655153", "postDate": "02/16/2024 17:44:46", "content": "Is this also true for the test data? Thank you.
", "rawMarkdown": "Is this also true for the test data? Thank you.", "votes": null }, { "id": "2655176", "postDate": "02/16/2024 17:52:31", "content": "Yes, this is also true for the test data. Each test row that we need to classify has an associated 50 second EEG and 10 minute Spectrogram. Our prediction should be the classification of the middle 10 seconds of each.
", "rawMarkdown": "Yes, this is also true for the test data. Each test row that we need to classify has an associated 50 second EEG and 10 minute Spectrogram. Our prediction should be the classification of the middle 10 seconds of each.", "votes": null }, { "id": "2661449", "postDate": "02/21/2024 08:54:24", "content": "Thank you.
", "rawMarkdown": "Thank you.", "votes": null } ], "comments": [ { "id": 2601604, "author_name": "cdeotte", "author_url": "", "post_date": "01/14/2024 14:39:32", "content": "\n\nDoes it mean that when the annotator is observing spectrogram data for annotation, they are only using the central 10 seconds (295 sec to 305 sec) out of the 10 minutes window of data (Label_offset_second + 600 seconds)? (like below)
\n
Yes. Call the middle timestamp in seconds time=T. Then
Thank you for your reply.
\nSo [T-5:T+5],
\nDoes this mean that not only spectrogram data but also EEG data are involved?
Yes the labelers use both data. And both data are the same. A Spectrogram is just a visual representation of EEG data.
", "votes": null, "replies": [] } ] }, { "id": 2655153, "author_name": "lester1027", "author_url": "", "post_date": "02/16/2024 17:44:46", "content": "Is this also true for the test data? Thank you.
", "votes": null, "replies": [ { "id": 2655176, "author_name": "cdeotte", "author_url": "", "post_date": "02/16/2024 17:52:31", "content": "Yes, this is also true for the test data. Each test row that we need to classify has an associated 50 second EEG and 10 minute Spectrogram. Our prediction should be the classification of the middle 10 seconds of each.
", "votes": null, "replies": [ { "id": 2661449, "author_name": "lester1027", "author_url": "", "post_date": "02/21/2024 08:54:24", "content": "Thank you.
", "votes": null, "replies": [] } ] } ] } ] } ], "raw_markdown_by_id": { "2601308": "I am struggling considerably to interpret the data.\n\n『The expert annotators reviewed 50 second long EEG samples plus matched spectrograms covering 10 a minute window centered at the same time and labeled the central 10 seconds.』\nCould you please tell me the meaning of the above text?\n\nDoes it mean that when the annotator is observing spectrogram data for annotation, they are only using the central 10 seconds (295 sec to 305 sec) out of the 10 minutes window of data (Label_offset_second + 600 seconds)? (like below)\n\n\nThank you all.", "2601604": ">Does it mean that when the annotator is observing spectrogram data for annotation, they are only using the central 10 seconds (295 sec to 305 sec) out of the 10 minutes window of data (Label_offset_second + 600 seconds)? (like below)\n\nYes. Call the middle timestamp in seconds `time=T`. Then\n* spectrogram is [T-300:T+300]\n* eeg is [T-25:T+25]\n* we classify [T-5:T+5]", "2602107": "Thank you for your reply.\nSo [T-5:T+5], \nDoes this mean that not only spectrogram data but also EEG data are involved?", "2602130": "Yes the labelers use both data. And both data are the same. A Spectrogram is just a visual representation of EEG data.", "2655153": "Is this also true for the test data? Thank you.", "2655176": "Yes, this is also true for the test data. Each test row that we need to classify has an associated 50 second EEG and 10 minute Spectrogram. Our prediction should be the classification of the middle 10 seconds of each.", "2661449": "Thank you." }, "source": "meta" }