![\"Scheme](\"https://storage.googleapis.com/kaggle-media/competitions/inclass/4520/media/comet_3d.gif\")
Information about position of straw tubes is kept in separate file wires.csv, it has 4482 lines with following information:
\r\nEach event is a 'snapshot' of detector, it consists of data taken from all 4482 wires: energy and relative time.
\r\nEvent is written only when hodoscope detects particle. Relative time is difference between time when straw tube detected energy and time of hodoscope.
\r\nData in train.csv is organized as following:
\r\nWire with zero energy deposition is not activated.
\r\ntest.csv has the same structure, but there is no label column (which you have to predict).
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Description", "content": "COMET is an experiment being constructed at the J-PARC proton beam laboratory in Japan. It will search for coherent neutrino-less conversion of a muon to an electron, \\\\(\\mu^- + N(A,Z) → e^- + N(A,Z)\\\\). This process breaks the law of lepton conservation. If detected, it will be a signal of new physics.
\r\nThe previous upper limit for this decay was set [5] by the SINDRUM II experiment in 2006. COMET is designed to have 10,000 times better sensitivity.
\r\nThe COMET experiment is looking for muon to electron conversion, \\\\(\\mu^- + N → e^- + N\\\\). COMET Phase-I will the Cylindrical Drift Chamber as the primary detector for physics measurements. Specifically, the momentum of resulting particles will be measured using the CyDet, which is a cylindrical wire array detector.
\r\nThe particles flying out of muon-stopping target and registered by the CyDet. Among those we are interested in tracks left by electrons with specific energy, which are produced by muon to electron conversion.
\r\nThe CyDet consists of 4482 sensitive wires organized in 18 layers. Each wire measures the energy deposited by a passing charged particle. Within each of the layers, the wires have same distance to the stopping target and stereometry angle.
\r\n
There is magnetic field in the detector, which causes electron moves in helical path as shown below. This electron deposits energy in the wires close to the flight path. The radius of helix is proportional to transverse momentum of the electron:
\r\n\\[ R = \\frac{p_t}{eB} \\]
\r\nwhere \\\\(p_t\\\\) is transverse momentum, \\\\(B\\\\) is strength of magnetic field, \\\\(e\\\\) is charge of the electron.
\r\n ![\"Trajectory](\"https://storage.googleapis.com/kaggle-media/competitions/inclass/4520/media/COMEThelixing.png\")
The energy deposited on each wire is measured at the end plate of the cylindrical detector. An example of the resulting signal event can be seen below, where blue dots are background hits and red are hits from signal electrons:
\r\n![\"Energy](\"https://storage.googleapis.com/kaggle-media/competitions/inclass/4520/media/COMET2dprojection.png\")
Check out starter kit to make first submission and explore the data!
\r\nDatasets available for this challenge are results of preliminary Monte Carlo simulation. They don't completely represent properties of COMET's detector and thus cannot be used to estimate final properties of tracking system, but are appropriate to test different approaches to tracking.
\r\nWe thank COMET collaboration (and specially Chen WU) for allowing us to use this dataset.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "Evaluation metric for this competition is area under ROC curve.
\r\nMeasured only quality of discrimination of signal hits vs background hits.
\r\nFor every activated wire in every event, submission files should contain two columns: global_id, prediction.
\r\nPrediction for signal hits should be higher than for background (to get high quality).
\r\nThe file should contain a header and have the following format:
\r\nglobal_id,prediction\r\n7619403,0.1\r\n
7619407,0.3\r\netc.\r\n
Pay attention, that only activated wires (with energy > 0) should be submitted.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "rules", "content": "\r\n\t\r\nYou cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.
\r\n\r\n Privately sharing code or data outside of teams is not permitted.\r\n It's okay to share code if made available to all participants on the forums.\r\n
\r\nTeam mergers are not allowed in this competition.
\r\n\r\nThe maximum size of a team is 2 participants.
\r\nYou may submit a maximum of 10 entries per day.
\r\nYou may select up to 2 final submissions for judging.
\r\nStart Date: 10/5/2016 6:28 PM UTC
\r\nMerger Deadline: None
\r\nEntry Deadline: None
\r\nEnd Date: 10/22/2016 11:59 PM UTC
\r\n\t\r\n'Data' means the Data or Datasets linked from the Competition Website for the purpose of use by Participants in the Competition. For the avoidance of doubt, Data is deemed for the purpose of these Competition Rules to include any prototype or executable code provided to Participants by Kaggle or Competition Organizer via the Website. Participants must use the Data only as permitted by these Competition Rules and any associated data use rules specified on the Competition Website.
\r\nUnless otherwise permitted by the terms of the Competition Website, Participants must use the Data solely for the purpose and duration of the Competition, including but not limited to reading and learning from the Data, analyzing the Data, modifying the Data and generally preparing your Submission and any underlying models and participating in forum discussions on the Website. Participants agree to use suitable measures to prevent persons who have not formally agreed to these Competition Rules from gaining access to the Data and agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Data to any party not participating in the Competition. Participants agree to notify Kaggle immediately upon learning of any possible unauthorized transmission or unauthorized access of the Data and agree to work with Kaggle to rectify any unauthorized transmission. Participants agree that participation in the Competition shall not be construed as having or being granted a license (expressly, by implication, estoppel, or otherwise) under, or any right of ownership in, any of the Data.
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }