![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8351195%2F0fc67a64144506e2c5d1d931a2cf8c25%2F107373477.png?generation=1675251099045712&alt=media\")
I separated the true and false data and plotted them. Given an event I find the following.
\nThe manual says, False is fully digitalised while True could be noise.
\nWhen I plotted using the sensor that were active for an particular event in a particular batch, I found they were all parallel to z axis. While the ones with true where scattered in the 3 dimension.
\nCan some one explain me more?
\nThanks
Event 1
Event 2 ![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8351195%2F85724b656982cb506f585e9f114a8044%2F42316582.png?generation=1675251112044820&alt=media\")
I separated the true and false data and plotted them. Given an event I find the following.
\nThe manual says, False is fully digitalised while True could be noise.
\nWhen I plotted using the sensor that were active for an particular event in a particular batch, I found they were all parallel to z axis. While the ones with true where scattered in the 3 dimension.
\nCan some one explain me more?
\nThanks
Event 1
Event 2
I think noisy events of auxilary==True can happen basically anywhere and randomly due to DOM detector noise or atmospheric muons etc.
\nOn the other hand, events of auxilary==True are signals from Cherenkov lights which happen locally.
The webinar record might be helpful to understand more:)
", "rawMarkdown": "I think noisy events of `auxilary==True` can happen basically anywhere and randomly due to DOM detector noise or atmospheric muons etc.\nOn the other hand, events of `auxilary==True` are signals from Cherenkov lights which happen locally.\n\n[The webinar record](https://www.kaggle.com/competitions/icecube-neutrinos-in-deep-ice/discussion/355653) might be helpful to understand more:)", "votes": null }, { "id": "2126974", "postDate": "02/02/2023 15:40:59", "content": "https://www.kaggle.com/competitions/icecube-neutrinos-in-deep-ice/discussion/379677
\nThe one thing to add to the above, is that after posting that quote in the discussion above about \"neighboring DOMs\", I've come to the conclusion that only z-axis sensors (aka DOMs) are considered neighbors. This makes sense because other strings of sensors are too far away (125m) compared with only 17m vertical distance between sensors on the same string.
\nBesides that, the events you plot are low signal (few rows), so that's why aux=False just looks purely vertical. With some of the big 200+ row events, it might look a bit different.
", "rawMarkdown": "https://www.kaggle.com/competitions/icecube-neutrinos-in-deep-ice/discussion/379677\n\nThe one thing to add to the above, is that after posting that quote in the discussion above about \"neighboring DOMs\", I've come to the conclusion that only z-axis sensors (aka DOMs) are considered neighbors. This makes sense because other strings of sensors are too far away (125m) compared with only 17m vertical distance between sensors on the same string. \n\nBesides that, the events you plot are low signal (few rows), so that's why aux=False just looks purely vertical. With some of the big 200+ row events, it might look a bit different.", "votes": null } ], "comments": [ { "id": 2126007, "author_name": "rmurata11", "author_url": "", "post_date": "02/02/2023 03:52:03", "content": "I think noisy events of auxilary==True can happen basically anywhere and randomly due to DOM detector noise or atmospheric muons etc.
\nOn the other hand, events of auxilary==True are signals from Cherenkov lights which happen locally.
The webinar record might be helpful to understand more:)
", "votes": null, "replies": [] }, { "id": 2126974, "author_name": "roberthatch", "author_url": "", "post_date": "02/02/2023 15:40:59", "content": "https://www.kaggle.com/competitions/icecube-neutrinos-in-deep-ice/discussion/379677
\nThe one thing to add to the above, is that after posting that quote in the discussion above about \"neighboring DOMs\", I've come to the conclusion that only z-axis sensors (aka DOMs) are considered neighbors. This makes sense because other strings of sensors are too far away (125m) compared with only 17m vertical distance between sensors on the same string.
\nBesides that, the events you plot are low signal (few rows), so that's why aux=False just looks purely vertical. With some of the big 200+ row events, it might look a bit different.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2125037": "I separated the true and false data and plotted them. Given an event I find the following. \n\nThe manual says, False is fully digitalised while True could be noise. \nWhen I plotted using the sensor that were active for an particular event in a particular batch, I found they were all parallel to z axis. While the ones with true where scattered in the 3 dimension.\nCan some one explain me more?\nThanks\n\nEvent 1\n\n\n\nEvent 2 ", "2126007": "I think noisy events of `auxilary==True` can happen basically anywhere and randomly due to DOM detector noise or atmospheric muons etc.\nOn the other hand, events of `auxilary==True` are signals from Cherenkov lights which happen locally.\n\n[The webinar record](https://www.kaggle.com/competitions/icecube-neutrinos-in-deep-ice/discussion/355653) might be helpful to understand more:)", "2126974": "https://www.kaggle.com/competitions/icecube-neutrinos-in-deep-ice/discussion/379677\n\nThe one thing to add to the above, is that after posting that quote in the discussion above about \"neighboring DOMs\", I've come to the conclusion that only z-axis sensors (aka DOMs) are considered neighbors. This makes sense because other strings of sensors are too far away (125m) compared with only 17m vertical distance between sensors on the same string. \n\nBesides that, the events you plot are low signal (few rows), so that's why aux=False just looks purely vertical. With some of the big 200+ row events, it might look a bit different." }, "source": "meta" }