|
The kernels are available at /tigress/ncaplar/Simulations/v01/Result/Fits. /tigress/ncaplar/Simulations/Scatter/v01/Npy contains the same files in numpy array form. In /tigress/ncaplar/Simulations/v01/ there is short readme directory describing usage of the files. The same directory also contains routines used to construct these kernels. The "v01" refers to the fact that I am using 0.1 version of the simulation code.
The fits files containing kernels are constructed as requested by cloomis (in 600 multifits files, 5x oversampled). The kernels are constructed from modeling real data at LAM - that is why only red camera data is available. Parameters which vary smoothly over the detector are kept the same as deduce in the real data (these parameters are Zernike coefficients describing wavefront aberrations, and description of the pupil). For parameters which are fiber dependent (e.g., FRD and other parameters describing the illumination of the exit pupil) numbers are randomly drawn from the normal distribution - parameters for these normal distributions have been deduced from the real data.
|