This repository contains pre-computed Fermi-LAT datasets, that can be used for analysis with gammapy. Currently the following datasets are included:

A more detailed description and listing of analysis parameters can be found in the corresponding sub-folders for the 2FHL and 3FHL datset.

To get the data just clone this repository to your local machine using git:

git clone https://github.com/gammapy/gammapy-fermi-lat-data

Now define the environment variable GAMMAPY_FERMI_LAT_DATA to point to the path where the repository is located on your machine:

export GAMMAPY_FERMI_LAT_DATA=path/to/gammapy-fermi-lat-data

In addition you have to download the latest galactic diffuse model directly by clicking here or using the command line:

wget https://fermi.gsfc.nasa.gov/ssc/data/analysis/software/aux/gll_iem_v06.fits path/to/fermi/diffuse/dir

Note: if you have an installation of the Fermi science tools, you already have the galactic diffuse model. It is contained in the sub folder path/to/fermi/science/tools/refdata/fermi/galdiffuse/gll_iem_v06.fits

Define the environment variable FERMI_DIFFUSE_DIR to point to the directory where the gll_iem_v06.fits file is contained:

export FERMI_DIFFUSE_DIR=path/to/fermi/diffuse/dir

Once you've copied the data and defined the environment variables to point to the corresponding directories, the data is ready to be used. Please check the examples provided in the docstring of the FermiLATDataset class from Gammapy or check out the tutorial Fermi-LAT data with Gammapy.

Every dataset includes a data preparation bash script, which runs the Fermi-LAT science tools to compute events, exposure, livetime and psf for given set of analysis parameters. Running the scripts to prepare the datasets in this repositiory requires the Fermi-LAT science tools to be installed. In addition the following environment variables have to be set:

* `FERMI_FT1_FILE`: Path to the event files list.
* `FERMI_FT2_FILE`: Path to the merged spacecraft file.

Normal users don't have to run the script but can just start their analyses from the datasets provided in this repository.