nuNet is a convolutional neural network platform for reconstructing neutrino properties from radio emission produced in Askaryan events
.
├── ...
├── common # directory with tools, models, ...
│ ├── bandpass_filters
│ ├── models
│ │ ├── runD[n] #
│ │ ├── runE[n] # saved models from training sessions
│ │ └── runF[n] #
│ └── tools
│ └── start_training # tools for running training sessions
│
├── training # directory for training code
│ └── runs
│ ├── direction
│ │ ├── archive # directory for archived runs and bases
│ │ ├── base # directory for bases
│ │ └── runD[n] # n = 1, 2, 3, ...
│ ├── energy
│ │ ├── archive #
│ │ ├── base # ──── || ────
│ │ └── runE[n] #
│ └── flavour
│ ├── archive #
│ ├── base # ──── || ────
│ └── runF[n] #
└── ...
The work in this repository is done by Sigfrid Stjärnholm, as a part in the work as a research assistant at Uppsala University with Christian Glaser as a supervisor.