Data access

IceCube has a bunch of public datasets available at https://icecube.wisc.edu/science/data-releases/. icecube_tools provides an easy interface to this repository so that you can download and organise your data through python.

from icecube_tools.utils.data import IceCubeData

The IceCubeData class provides this functionality.

my_data = IceCubeData()

You can use the find method to pick out datasets you are interested in.

found_dataset = my_data.find("20181018")
found_dataset

['20181018']

The my_data object has been inititalised to store data in the package’s default location (“~/.icecube_data”). This is where other icecube_tools modules will look for stored data.

my_data.data_directory

'/home/runner/.icecube_data'

The fetch method will download this dataset to this default location for later use by icecube_tools. This method takes a list of names, so can also be used to download multiple datasets. fetch has a built in delay of a few seconds between HTTP requests to avoid spamming the website. fetch will not overwrite files by default, but this can be forced with overwrite=True.

my_data.fetch(found_dataset)

20181018: 11730609it [00:00, 56099319.94it/s]

You may not want to use icecube_tools for other stuff, so you can also fetch to a specificed location with the keyword write_to.

my_data.fetch(found_dataset, write_to="data", overwrite=True)

20181018: 11730609it [00:00, 52666286.58it/s]

For convenience, there is also the fetch_all_to method to download all the available data to a specified location. We comment this here as it can take a while to execute.

# my_data.fetch_all_to("data")