GitLab

Add a carriage return after we have computed the last model or analysed the last object so that the information does not get erased.

Add the IRAS, SCUBA and VLA filters from Barbara.

Add some filters from Barbara.

The "galaxy_mass" parameter must not be set to 1.0 in the stellar
emission modules because when using CIGALE to generate SED without
normalising the 1.0 is wrong and misleading.

Instead, we compute the real galaxy mass (the integration of the SFH) in
the star formation history modules.

Now that we handle the luminosity and the individual component separately for speed reasons, we must not forget to scale the total luminosity appropriately when redshifting. Otherwise the fluxes in the filters will be overestimated by a factor 1+z.

Let's avoid using costly np.max() functions on ordered arrays. The max will always be the last element. Also use the fact that the arrays are sorted to select the last 100 elements to compute the average SFR.

Setters and getters are nice but they come with a significant overhead. Because we initialise class members to None anyway, they should not be needed at all. I leave the setter/getter for the sfh as it is actually doing something more. It now accounts for a non negligible fraction of the runtime though.

Rather than relying on deepcopy, let's build the new SED objects ourselves. As we know the structure perfectly we can do that much more efficiently. To do se we add a copy() member to the SED class. This function creates a new object and initialises its members with proper copies.

Call the copy() class member rather than the base numpy function. Quick testing shows it is faster: 1.1 μs vs 2.7 μs for an array of 2000 elements.

Now that we do not add an AGN component anymore when the AGN fraction is 0, the wavelength grid may change from one model to the other. To work around this issue we save the IGM component into a dictionary with the key being the size of the grid. This is not completely foolproof as two grids of same length may have different wavelengths, though that is quite unlikely given pcigale's architecture. It should be good enough for now.

If the AGN fraction is 0, then there is no need to add a component to the SED. It would have a value of 0 everywhere. It looks like we gain around 30% in computation time in this case.

Computing the luminosity each time we integrate in a filter (for instance) is not efficient. As the luminosity is in effect modified in only one place (add_contribution), let's just compute it manually each we add a component. This yields a ~10% improvement.

Update the Dale 2014 model with a finer sampling. We now use the same grid as the Draine models. Courtesy Daniel Dale.

In the Bruzual and Charlot (2003) SSP module, we add parameters to the
info dictionary containing the total star and gas masses for the galaxy.

Specify that encoding for pcigale.ini. Otherwise if a non-ASCII character is used it may work or crash depending on the computer. Now we encode everything in UTF-8, which will allow again to use these chracters in the description of the modules parameters.

Use the base numpy interp() function rather than scipy's interp1d(). The reason is that when interpolation only 1 array the former is much faster (probably because it is compiled and there is no overhead returning a function. We can only do that for the new component. For the old ones, there are too many for this to be worthwhile. Still, it nets a nice improvement.

The wavelength arrays are sorted. So there is no need to call min() and max() to find the extrema. We can directly get the first and last elements.

Small optimisation in the interpolation. There is no need to sort the arrays because they are already sorted.

Add errorbars to the SED plots. Because they tend to be small compared to the dynamic range, we use 3σ uncertainties rather than just 1σ.