Skip to content

Releases: zoghbi-a/nicer-background

0.3.t1.35

06 Dec 18:29
Compare
Choose a tag to compare

This version uses the latest heasoft updates (as of August 2021). The model is optimizd to work in the energy range 0.5-10 keV (unlike the previous version). It uses 40 MKF parameters, including those from the geomagnetic data. It is a classification model that samples the MKF parameters every 1 second and classifies the background data into 35 basis spectra. The root-mean-squared performance in the background estimates is 0.21 counts/s vs 4.04 in the 3C50 model. If the 1% outlier backgorund observations are discarded, the performance is 0.15 counts/s (vs 1.25 for the 3C50 model).

This version requires a geomagnetic data files that can be downloaded from https://heasarc.gsfc.nasa.gov/FTP/caldb/data/gen/pcf/geomag (see discussion here).. Please download the 6 files: dst_kyoto.fits, f107_petincton.fits, geomag.tar.gz, kp_noaa.fits, kp_potsdam.fits, solarphi_oulu.fits into some directory and then provide the location of that directory as input to the nicerBgML.py script.

0.2.t4n20

15 Sep 01:49
Compare
Choose a tag to compare

This is an enhanced version of 0.1.t4n20, released on July 11, 2021. It is based on more MKF parameters, 27 in total, including the KP index from the geomagnetic data. It is also a classification model that samples the MKF parameters every 4 seconds that classifies the background data into 20 basis spectra. The root-mean-squared performance in the background estimates is 1.7 counts/s vs 3.2 in the 3C50 model. If the 1% outlier background observations are discarded, the performance is 0.87 counts/s (vs 2.3 for the 3C50 model).

Version 0.2.t4n20 requires a geomagnetic data file. The latest file can be downloaded from https://heasarc.gsfc.nasa.gov/FTP/caldb/data/gen/pcf/geomag/kp_noaa.fits. The model was trained using the noaa data file, so it is suggested that it is the one used and not the Potsdam one.

Basic Background Model t4n20

11 Jul 02:33
Compare
Choose a tag to compare

This is the first model presented at the NICER Observatory Science Working Group (OSWG) on June 30, 2021. It is based on a classification model that uses 15 parameters from the MKF file sampled every 4 seconds to classify the background data into 20 basis spectra.