Working with sources¶

The Source abstraction¶

The main abstraction of Agilepy is the Source class. It is described by several parameters, some of which can be free to vary, and they are changed by the mle() analysis.

The set of the parameters describing the source can vary, depending on the spectrum and spatial model types of the source.

The different types of sources are described here.

How to load or add new sources¶

In order to perform a scientific analysis with Agilpy, at least one Source model must be loaded. There are several ways to do that.

The loadSourcesFromCatalog(catalogName, rangeDist=0, inf, show=False) allows to load a source catalog, while filtering the sources by their distance (degree) from the l,b position provided within the configuration file.

sources = ag.loadSourcesFromCatalog('2AGL', rangeDist=(0, 10))

The loadSourcesFromFile(sourcesFilepath, rangeDist=0, inf, show=False) loads the sources, reading their model from a file.

The addSource(sourceName, sourceDict) method allows the user to define on the fly a source model with a python dictionary. Check the tutorial notebooks for an example.

How to let the source’s parameters to vary¶

In order to free or fix a sources’ parameter, the user can rely on the freeSources(selection, parameterName, free, show=False) method. The sources can be selected via the “selection” argument, supporting either lambda functions and boolean expression strings. Please, check the selectSources(selection, show=False) method documentation for more informations.

aganalysis.freeSources(lambda name, dist, flux : Name == "2AGLJ2021+4029" AND dist > 0 AND flux > 0, "flux", True)

ag.freeSources('name == "2AGLJ1513-0905"', "index", True, show=True)

Check the api documentation or the tutorial notebooks for additional examples.

How to inspect a Source object¶

The common way to inspect a Source object is to print it:

 source = ag.selectSources('name == "2AGLJ2254+1609"', show=False).pop()
 print(source)
 -----------------------------------------------------------
 Source name: 2AGLJ2254+1609 (PointSource)
  * Free parameters: flux
  * Initial source parameters: (PowerLaw)
     - flux(ph/cm2s): 7.50937e-07
     - index: 2.20942
     - Source position: (86.1236, -38.1824) (l,b)
     - Distance from map center: 0.011 deg
-----------------------------------------------------------

The user can call selectSources (with show=True) to show the source description

ag.selectSources('name == "PKS1510-089"', show=True)

ag.selectSources('flux > 0', show=True)

ag.selectSources(lambda name, sqrtTS: name == "2AGLJ2021+4029" AND sqrtTS> 0, show=True)

A getter method is available for “spectrum” and “multi” parameters:

print(source.multi.get("multiFitLikelihood1"))

print(source.spectrum.get("index1"))

How to check which source’s parameters are free to vary¶

The user can obtain this information by printing the Source object or calling the getFreeParams() method of the Source object.

print(source.getFreeParams())
['flux']

The “multi” description of a Source object¶

If the user performs an mle analysis, the Source object will contain also the analysis results.

 print(source)
 -----------------------------------------------------------
 Source name: 2AGLJ2254+1609 (PointSource) => sqrt(ts): 10.2226
  * Free parameters: flux index
  * Initial source parameters: (PowerLaw)
  - flux(ph/cm2s): 7.50937e-07
  - index: 2.20942
  - Source position: (86.1236, -38.1824) (l,b)
  - Distance from map center: 0.011 deg
  * Last MLE analysis:
  - flux(ph/cm2s): 8.68363e-06 +/- 1.62474e-06
  - index: 2.51001 +/- 0.173795
  - upper limit(ph/cm2s): 1.13967e-05
  - ergLog(erg/cm2s): 1.39217e-09 +/- 2.6048e-10
  - galCoeff: [0.7, 0.7, 0.7, 0.7, 0.7]
  - isoCoeff: [5.24757, 3.14662, 0.953512, 1.59944e-10, 0.557554] +/- [0.519283, 0.209362, 0.051478, 0.000108147, 0.000720814]
  - exposure(cm2s): 13672200.0
  - exp-ratio: 0.0
  - L_peak: 86.1236
  - B_peak: -38.1824
  - Distance from start pos: 0.0
  - position:
      - L: -1.0
      - B: -1.0
      - Distance from start pos: -1.0
      - radius of circle: -1.0
      - ellipse:
      - a: -1.0
      - b: -1.0
      - phi: -1.0
-----------------------------------------------------------

The values L_peak and B_peak set to the initial values in the source location is fixed. If it is allowed to vary then they are set to the position for which the TS is maximized. If a confidence contour was found, the parameters of the “ellipse” section describe the best-fit ellipse of the contour, described in detail below. The counts and fluxes are provided, as well as their symmetric, positive, and negative errors if the flux is allowed to vary. For convenience, the exposure of the source, used to calculate the source counts from the flux, is also provided. Finally, the spectral index and its error, or the other spectral parameters, if applicable, are provided.

How to manually change a source object’s attributes¶

In order to change the spectrum parameters values, the user can call the set() method of the Spectrum object within the Source object. Example:

source.spectrum.set("index2", 1.34774)

Another way is to call the setAttributes method of a Source attribute. This allow to change the following attributes: value, free, scale, min, max, locationLimit. Example:

source.spectrum.cutoffEnergy.setAttributes(min=3000, max=5000)

In order to change the position of a source, the user can rely on the updateSourcePosition(sourceName, glon, glat) method.