- No changes yet.
- Update test_footprint.py for latest tiles file with 5 (not 4) bright passes (update to main, not PR)
- Update high level desimodel sync scripts used by cron jobs (PR #171).
- Small fix to focal plane sync scripts (PR #172).
NOTE: non-backwards compatible change to :func:`~desimodel.io.loadtiles`!
- Change default file for :func:`~desimodel.io.loadtiles` (PR #168):
- Now reads from :envvar:`DESI_SURVEYOPS`
(/trunk/ops/tiles-main.ecsv)
- Also adds option to limit tiles to specified
PROGRAM
names. - Will now automatically load both .fits and .ecsv files.
- Maintains option to read from old :envvar:`DESIMODEL` location.
- Tests cover both :envvar:`DESI_SURVEYOPS` and :envvar:`DESIMODEL` cases.
- Addresses issue #167.
- Now reads from :envvar:`DESI_SURVEYOPS`
- Optionally fetch the valid date range from load_focalplane (PR #170).
- Update KPNO sync scripts to use the default software stack (PR #165).
- Add missing Python modules to
doc/api.rst
(PR #163). - Update the extra data download when installing with :command:`desiInstall` (PR #159).
- Fix :func:`~desimodel.footprint.tiles2pix` to allow
TILERA
,TILEDEC
(PR #156). - fastfiberacceptance use
.astype('=f8')
to force native endianness for scipy>=1.10.0 compatibility (PR #164).
Changes for syncing focalplane model with ICS database focalplane calibrations:
- On focal plane sync, update KPNO-wide default DESIMODEL/data checkout first. (PR #151).
- Handle swapped fibers 3429,3402 at locations 98,99 on PETAL_LOC=6 (PETAL_ID=11) (PR #152).
- Remove deprecated focal plane generation routine, update associated docs.
- Update
etc/desimodel_sync_kpno_cron.sh
for automatically syncing the focalplane model to the latest DB dump (PR #148). - Add fiberfrac to PSF seeing conversions (PR #149).
- Move from positioner exclusion yaml format to much faster json format (PR #150).
Code in GitHub:
- Use UTC time everywhere in the focalplane model (PR #147). This is backwards compatible with old files, but new FP models will not be readable by previous code tags.
Data in svn:
- DB sync 2021-04-03T23:53:23 appended to
desi-state-2021-03-17T23:20:01.ecsv
. - DB sync 2021-04-10T20:00:39 appended to
desi-state_2021-03-17T23:20:01.ecsv
. - DB sync 2021-04-13T20:00:30 appended to
desi-state_2021-03-17T23:20:01.ecsv
.
Data changes to svn, no code changes:
- Added LYA TSNR2 templates.
- Focalplane model updated 2021-03-17.
- Add fastfiberacceptance code originally in specsim (PR #145).
- Code (in GitHub):
- Data (in svn):
- added Template Signal-to-Noise (TSNR) ensembles
- added pre-calculated Noise Equivalent Area (NEA) from PSF model
- Corrected restricted reach focalplane model (@135002)
- Focal plan model with full reach for petal locs 0,2,4,5 (@135236)
- New tag of data+code since data had been erroneously pre-tagged 0.13.0
- Fix py3.8 syntax warnings (PR #140).
- Fix corner cases in generating and using focalplane models (PR #139).
- Use DESI-5501 (as built) instead of DESI-334 (design) for spectrograph throughput (PR #137).
- update platescale to as-built DESI-4037v5 (PR #136).
- update desi-focalplane model for limited phi range 20200306 (svn data).
- fix bug in generating focalplane model from old fiberpos files (PR #139).
- use >= not > when comparing runtime to focalplane model #139).
- Updated
data/footprint/desi-tiles.fits
anddesi-healpix-weights.fits
with new dither pattern; see DESI-0717. Layers 0=GRAY, 1-4=DARK instead of 0-3=DARK, 4=GRAY. (PR #135). - Update documentation for :mod:`desimodel.io`; use :func:`desimodel.io.findfile` consistently throughout the module (PR #133).
- Update README file and Travis tests (PR #132).
- Include S (curved focal surface arc length) vs. R (CS5 xy radius) table from DESI-0530 (PR #130 and #135).
- Pass multiple sets of exclusion polygons (PR #128).
- Propagate existing focalplane state to new focalplanes (PR #129).
- Improve focalplane creation code (PR #127).
- Workaround upstream bugs in positioner locations (PR #118).
- Added desimodel.focalplate.fieldrot.field_rotation_angle with field rotation CS5 vs. ICRS due to precession (PR #119).
- Add focalplane model documentation (PR #125).
- Store petal and gfa keepouts in the focalplane model (PR #112).
- When generating a focalplane, check for device locations assigned to the same slitblock and fiber (PR #113).
- Fix support for
datetime.isoformat()
in Python 3.5 (PR #114). - Update tests and documentation to be consistent with latest desiutil versions (PR #115).
- Support for time-varying focal plane state (e.g. broken fibers) (PR #105).
- Documentation about CI weather versus model (PR #107).
- Fix :func:`~desimodel.footprint.find_points_radec` for scipy 1.3 (PR #109).
- Replace deprecated
yaml.load
withyaml.safe_load
(PR #110).
- Added data/footprint/ci-tiles-v7.fits, data/focalplane/ci-corners.ecsv to svn and docs to GitHub (PR #103).
io.load_tiles(tilesfile)
warns if local copy exists, but :envvar:`DESIMODEL` version wins (PR #98 and #101).- Update default tile radius (max radius, not typical outer pos radius) (PR #102).
- Change default healpy pixel overlap factor from 4 to 128 (PR #93).
- Implement :func:`~desimodel.weather.dome_close_fractions` to replay daily Mayall weather history (PR #92).
- Run tests using new svn branch test-0.9.8.
- Bug fix for GFA target selection when no targets overlap a GFA (PR #91).
- Create DESI-3977 in doc/tex/desi3977/ to track ELG SNR with changes to the DESI model.
- Add accompanying notebook doc/nb/ELG_SNR.ipynb.
- Update data and associated code to reflect changes in DESI-347-v13 (PR #89):
data/throughput/thru-[brz].fits
: new corrector coatings.data/throughput/DESI-0347_blur.ecsv
: new achromatic blurs.data/desi.yaml
: new read noise and dark currents.data/focalplane/gfa.ecsv
: replaceRADIUS_MM
withS
.data/throughput/DESI-0347_static_[123].fits
: replace random offset files (RMS=10.886um) with static offset files (RMS=8.0um).
- Use a new svn branch test-0.9.6 for travis tests (was test-0.9.3).
- Increase test coverage, especially for :mod:`desimodel.trim` (PR #82).
- Reorganize :mod:`desimodel.focalplane` and add more GFA selection code (PR #85).
- Allow an environment variable in the tilesfile filename (PR #87).
- Download script will create :envvar:`INSTALL_DIR` if it doesn't exist (PR #80).
- Fix some installation bugs, and update to latest versions on various dependencies (PR #77).
- Ensure that desimodel tests are compatible with Astropy 2 and 3, and with other DESI packages (PR #78).
- Add
footprint/desi-healpix-weights.fits
andthroughput/galsim-fiber-acceptance.fits
to the trimmed test data set (PR #79).
- Update LyA S/N calculation (PR #73).
- Optionally use an input pixel weight map in :func:`~desimodel.io.load_pixweight` (PR #74).
- Extracts wavelength coverage from specpsf files into params dictionary (PR #68).
- Added :func:`~desimodel.footprint.program2pass` and :func:`~desimodel.footprint.pass2program` to convert between tiling integer pass number and string program name (PR #67).
- Added desimodel.focalplane.radec2xy, which converts RA, Dec coordinates to x, y coordinates on the focal plane, which accepts vector inputs.
- Added desimodel.focalplane.on_gfa() and its respective helper functions to check if a target is on a GFA of arbitrary telescope pointing
- Added desimodel.focalplane.on_tile_gfa() to check return a list of indices of targets on a specific tile
- Added desimodel.focalplane.get_gfa_targets() to return a table with added columns GFA_LOC and TILEID that consists of all targets on any GFA on any tile satisfying a minimum flux in the r-band.
- Unittests for the desimodel.focalplane functions were updated accordingly.
- Added desimodel.footprint.find_points_in_tel_range() to return a list of indices withnin a radius of an arbitray telescope pointing, unaware of tiles (Added respective unittest)
- Adds desimodel.focalplane.fiber_area_arcsec2()
- Updates tests to work with trimmed data subset
- Add new weather module to specify assumed atmospheric seeing and transparency distributions at KPNO, with accompanying DESI-doc and jupyter notebook.
- Remove seeing module, which is superseded by new weather module.
- Added desimodel.footprint.pixweight() in :mod:`desimodel.footprint` to create an array of what fraction of every HEALPixel at a given nside overlaps the DESI footprint
- Also added desimodel.footprint.tiles2fracpix() to estimate which HEALPixels overlap the footprint edges
- Added desimodel.io.load_pixweight() in :mod:`desimodel.io` to load the array created by desimodel.footprint.pixweight() and resample it to any HEALPix nside
- Modified path to Lya SNR spectra files used in desi_quicklya.py, used in Lya Fisher forecast.
- Added desimodel.inputs.build_gfa_table and its helper functions to write a .ecsv file for GFA data
- Added desimodel.io.load_gfa to return the GFA data table
- Added desimodel.focalplane.xy2radec, which converts x,y coordinates on the focal plane to RA, Dec coordinates
- don't print warnings in desimodel.io if specter isn't installed
- Added desimodel.footprint.tiles2pix and .pix2tiles for mapping healpix to DESI tiles.
- fixed psf-quicksim.fits units to be astropy-friendly
- added desimodel.io.load_target_info()
- Add desimodel.seeing module with functions that model the expected DESI zenith seeing at 6355A, with an accompanying jupyter notebook.
- Altered xy offset RMS calculation in focalplane.py to scale the distribution RMS rather than the sample standard deviation.
- Update focal plane to positioner mapping
- z-channel 250 um CCD instead of 500 um CCD
- Update DocDB -> desimodel update method for fiberpos and throughput
- By default, desimodel.io.load_tiles now excludes PROGRAM=EXTRA layers
- Adds desi-tiles.* tests
- Moved test of focalplane code into the actual test suite.
- Preparing for Python 3.
- Changed default svn version to trunk and added error handling to :command:`install_desimodel_data`.
- Update template module file to reflect DESI+Anaconda infrastructure.
- Add code to generate random centroid offsets in :mod:`desimodel.focalplane`.
- Add jupyter notebook documenting new throughput files of PR#29.
- Use Astropy-recommended method of reading FITS data tables.
- Remove reference to Travis scripts in MANIFEST.in.
- Fixed a minor bug that made the help message for :command:`install_desimodel_data` garbled.
- Add additional files to lightweight test data to work with quickgen
- Allow :command:`desiInstall` to download and install the data from svn.
- No changes to data in svn.
- "First" post-separation tag.
- Added :func:`desimodel.trim.trim_data` for trimming a data directory into a lightweight version for testing.
- svn data includes targets.dat: preliminary numbers for MWS and BGS densities (Still waiting upon supporting technote).
- Improved svn download instructions in the README file.
- Changes to data on svn side
- updated desi.yaml with dark vs. bright exptime
- updated targets.dat to include MWS placeholders
- :func:`desimodel.io.load_desiparams` adds 'exptime' -> 'exptime_dark' key for temporary backwards compatibility
- Removed deprecated fibers module
- Use ci-helpers to handle most of the dirty work of Travis build scripts.
- Make specter import errors more verbose.
- Last tag prior to separating desimodel into code (GitHub) and data (svn) repositories.
- pip install support (BAW).
- Replace fitsio dependency with astropy.io.fits.
- Added tile file for the bright time survey.
- Adds python io library (SJB).
- Tag to support dogwood production (SJB).
- Adds :meth:`desimodel.focalplane.FocalPlane.xy2radec` from Jaime (SJB).
- data/targets/targets.dat
- added fractions for sky and stdstar fibers (SJB).
- py/desimodel/focalplane.py
- bug fixes for transformations (SJB).
- Fix a simple import error (BAW).
- Fix a simple version error (BAW).
- Change how version is set (BAW).
- Updated target numbers.
- Also updated quicksim sn-spec* file output, using IDL version which is slightly more optimistic than the python version (diff is dark current?) (SJB).
- Updated throughput files for real.
- Added initial "compare_versions.py" script to make it easier to visualize differences in versions. This script should grow as various parameters change; right now it only makes a thoughput difference plot (SJB).
- Updated throughput files from 0334v3 (spectro) and 0347v5 (system throughput) Correction: thoughput files didn't make it into that change (SJB, 2014-07-08).
- Updated psf-b.fits and psf-quicksim.fits to match new npix_y for blue STA/ITL CCDs (SJB, 2014-07-08).
- Added ELG spectrum with continuum and multiple emission lines
Python quicksim
- add readnoise contributions in quadrature during the downsampling
- Refactor for speed, results now named ndarray, updated plots
- Allow different base directories
- Put sky back to dimmer UVES sky model
- Extended fiberloss range from 3500-10000 instead of 3600-10000
- Added data/throughput/fiberloss-qso.dat (same as fiberloss-star.dat)
- Updated data/focalplane/platescale.txt with latest from DESI-0329v14. This includes a new "theta" column.
- Updated desi.yaml from DESI-0347v4. This removes the FWHM and wavemin/max params which are not derived quantities associated with the PSFs.
- Updated throughput files with new numbers from DESI-0347v4.
- Updated spectrograph throughput files with new numbers from DESI-0334v2.
- Updated py/fiberloss.py -> bin/fiberloss.py . Biggest change is ELG half light radius 0.35" -> 0.45" which drops us below 7-sigma.
- Updated data/throughput/fiberloss-*.dat files with calculation based upon fiberloss.py
- bin/psf2quicksim.py extracted PSF parameters needed for quicksim.
- pro/desi_quicksim.pro updated, but it still treats FWHM as constant rather than wavelength dependent.
- python quicksim will be broken until it is updated to use new inputs.
- Reorganized data/inputs/throughput/
- spots2psf.py: leftover spot mirroring bug removed, PSFs updated
- Updated throughputs to not double count central obscuration.
- Updated PSF files to remove throughputs to avoid possible inconsistency.
- Added wavemin_all, wavemax_all to desi.yaml with min/max wavelength seen by all spectra
- Updated CCD pixel dimensions and regenerated PSFs to match.