Get Started
fitssets: The annoying thing about parsing is that you have to do the same operation on multiple fits files. In pyird, a set of fits is defined as a fitsset (a set of fits). In pyird, a set of fits is defined as a fitsset, and a single fit is also defined as a set of fits.
In pyird, a set of fits is defined by a fitsset (a set of fits). For example, bias, readout, unidimensionalization, wavelength assignment, etc. are one-way processes, and each time they are performed, iraf emits a large number of new fits files. This is the reason why stream is defined separately from the fits set. Basically, it is a one-way process, so if a process has been done once and saved in a directory, it will be skipped even if the same instruction is given again. There are two types of streams: one for two-dimensional data (the endpoint is one-dimensional), and one for one-dimensional data.
directory : Another annoying part of the analysis is directory management. If you try to put the data location and the large number of fits files generated by iraf in the same directory, just ls will be confusing. If you leave it for a week, you won’t know what’s going on. So, in pyird’s stream, the raw data directory (assumed to be a collection of soft links from other organized directories) and the analysis directory are handled separately. By keeping the output of iraf separate from the raw data, it is possible to run several different analyses in parallel, or to specify a separate empty analysis directory and reproduce the data processing from scratch by running the python code. In other words, simply put, it’s a tool for people with questionable memory.
Example of stream definition (42306,42308,…. Assigning the set of even (=YJ) fits to stream)
targets=irdstream.Stream2D("targets",datadir,anadir)
targets.fitsid=list(range(42306,42324,2))
When we make all the above odd (YJ+1=H)
targets.fitsid_increment()
Stream’s current location path
targets.path()
>[PosixPath('/media/kawahara/kingyo/IRD_TWINS/data/IRDA00042306.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/data/IRDA 00042308.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/data/IRDA00042310.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ data/IRDA00042312.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/data/IRDA00042314.fits'), PosixPath('/media/kawahara/kingyo/ IRD_TWINS/data/IRDA00042316.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/data/IRDA00042318.fits'), PosixPath('/media/kawahara /kingyo/IRD_TWINS/data/IRDA00042320.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/data/IRDA00042322.fits')]
Since this is the starting point of the stream, the path of the fits set in the data directory is returned. If we print the path again after the bias removal process, we get
targets.remove_bias("r")
targets.path()
>[PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA00042306_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA 00042308_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA00042310_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_ TWINS/ana/H/IRDA00042312_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA00042314_rb.fits'), PosixPath('/media/ kawahara/kingyo/IRD_TWINS/ana/H/IRDA00042316_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA00042318_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA00042320_rb.fits'), PosixPath('/media/kawahara/kingyo/IRD_TWINS/ana/H/IRDA 00042322_rb.fits')]
The path of the bias-removed fits is returned. Note that the path has been moved to the analysis directory, which is pathlib object by default. Extract the current location data (index=2,3) of Stream.
target.data([2,3])
Getting started
We recommend to look at, pyird/examples/python/IRD_stream.py for an example of H.
python IRD_stream.py
Please change the fits IDs when you try (or ask Kasagi for the data set). The tutorial and detailed explanation can be found IRD stream <https://secondearths.sakura.ne.jp/pyird/tutorials/IRD_stream.html>.
And, pyird/samples/IRD_TWINS/irdreductH.py for another example of H.
python irdreductH.py
Try it! The data set is here (about 1GB)
If you are a developer, please cut the feature/hogehoge from the develop branch and pull request when you merge it into develop. (Kawahara will also operate in this way after 4/7)
Example of what to prepare
(The followings are for irdreductH.py. If you follow the pipeline of IRD_stream.py, please read them only as a reference.)
fitsid.py – Define the stream. If you define everything here first, you won’t have to look at the observation log as often.
irdreductH.py, irdreductYJ.py – Single code for primary processing. aperture (created with iraf’s aparl). See below.
Example: remove bias in stream (“r” is rotated for H, not needed for YJ)
targets.remove_bias("r")
Example: image combine stream, output is a set of fits.
combined_flat=flat_mmf.imcombine("combined_flat")
Example: remove read noise from stream.
targets.rm_readnoise(maskfits)
Example: Define a set of fits (single file)
combined_flat=fitsset.FitsSet("combined_flat_rb",