snowstorm_dataset

Snowstorm dataset module hosted on the IceCube Collaboration servers.

class graphnet.datasets.snowstorm_dataset.SnowStormDataset(run_ids, graph_definition, download_dir, truth, features, train_dataloader_kwargs, validation_dataloader_kwargs, test_dataloader_kwargs)[source]

Bases: IceCubeHostedDataset

IceCube SnowStorm Monte Carlo simulation dataset.

This module provides access to certain RunIDs of the SnowStorm simulation data set. It prepares the data for the training and evaluation of deep learning models in GraphNeT by parsing it into the CuratedDataset format.

The data is organized by SnowStorm RunIDs containing pulsemaps input features along with event-level truth information.

The access requires an IceCube Collaboration account.

References: SnowStorm documentation: https://wiki.icecube.wisc.edu/index.php/SnowStorm_MC#File_Locations SnowStorm paper: arXiv:1909.01530

Construct SnowStormDataset.

Parameters:

  • run_ids (List[int]) – List of RunIDs to include.

  • graph_definition (GraphDefinition) – Method that defines the data representation.

  • download_dir (str) – Directory to download dataset to.

  • truth (Optional) – List of event-level truth to include. Will include all available information if not given.

  • features (Optional) – List of input features from pulsemap to use. If not given, all available features will be used.

  • train_dataloader_kwargs (Optional) – Arguments for the training DataLoader. Default None.

  • validation_dataloader_kwargs (Optional) – Arguments for the validation DataLoader, Default None.

  • test_dataloader_kwargs (Optional) – Arguments for the test DataLoader. Default None.