import numpy as np import matplotlib.pylab as plt import healpy as hp from lsst.sims.featureScheduler.modelObservatory import Model_observatory from lsst.sims.featureScheduler.schedulers import Core_scheduler, simple_filter_sched from lsst.sims.featureScheduler.utils import standard_goals, generate_goal_map, Footprint import lsst.sims.featureScheduler.basis_functions as bf from lsst.sims.featureScheduler.surveys import (Greedy_survey, generate_dd_surveys, Blob_survey) from lsst.sims.featureScheduler import sim_runner import lsst.sims.featureScheduler.detailers as detailers import sys import subprocess import os import argparse def wfd_half(target_map=None): """return Two maps that split the WFD in two dec bands """ if target_map is None: sg = standard_goals() target_map = sg['r'] + 0 wfd_pix = np.where(target_map == 1)[0] wfd_map = target_map*0 wfd_map[wfd_pix] = 1 wfd_halves = slice_wfd_area(2, {'r': wfd_map}, scale_down_factor=0) result = [-wfd_halves[0]['r'], -wfd_halves[1]['r']] return result def slice_wfd_area(nslice, target_map, scale_down_factor=0.2): """ Slice the WFD area into even dec bands """ # Make it so things still sum to one. scale_up_factor = nslice - scale_down_factor*(nslice-1) wfd = target_map['r'] * 0 wfd_indices = np.where(target_map['r'] == 1)[0] wfd[wfd_indices] = 1 wfd_accum = np.cumsum(wfd) split_wfd_indices = np.floor(np.max(wfd_accum)/nslice*(np.arange(nslice)+1)).astype(int) split_wfd_indices = split_wfd_indices.tolist() split_wfd_indices = [0] + split_wfd_indices all_scaled_down = {} for filtername in target_map: all_scaled_down[filtername] = target_map[filtername]+0 all_scaled_down[filtername][wfd_indices] *= scale_down_factor scaled_maps = [] for i in range(len(split_wfd_indices)-1): new_map = {} indices = wfd_indices[split_wfd_indices[i]:split_wfd_indices[i+1]] for filtername in all_scaled_down: new_map[filtername] = all_scaled_down[filtername] + 0 new_map[filtername][indices] = target_map[filtername][indices]*scale_up_factor scaled_maps.append(new_map) return scaled_maps def gen_greedy_surveys(nside=32, nexp=1, exptime=30., filters=['i', 'z', 'y'], camera_rot_limits=[-80., 80.], shadow_minutes=60., max_alt=76., moon_distance=30., ignore_obs='DD', m5_weight=3., footprint_weight=0.3, slewtime_weight=3., stayfilter_weight=3., footprints=None, roll_weight=3.): """ Make a quick set of greedy surveys This is a convienence function to generate a list of survey objects that can be used with lsst.sims.featureScheduler.schedulers.Core_scheduler. To ensure we are robust against changes in the sims_featureScheduler codebase, all kwargs are explicitly set. Parameters ---------- nside : int (32) The HEALpix nside to use nexp : int (1) The number of exposures to use in a visit. exptime : float (30.) The exposure time to use per visit (seconds) filters : list of str (['r', 'i', 'z', 'y']) Which filters to generate surveys for. camera_rot_limits : list of float ([-80., 80.]) The limits to impose when rotationally dithering the camera (degrees). shadow_minutes : float (60.) Used to mask regions around zenith (minutes) max_alt : float (76. The maximium altitude to use when masking zenith (degrees) moon_distance : float (30.) The mask radius to apply around the moon (degrees) ignore_obs : str or list of str ('DD') Ignore observations by surveys that include the given substring(s). m5_weight : float (3.) The weight for the 5-sigma depth difference basis function footprint_weight : float (0.3) The weight on the survey footprint basis function. slewtime_weight : float (3.) The weight on the slewtime basis function stayfilter_weight : float (3.) The weight on basis function that tries to stay avoid filter changes. """ # Define the extra parameters that are used in the greedy survey. I # think these are fairly set, so no need to promote to utility func kwargs greed_survey_params = {'block_size': 1, 'smoothing_kernel': None, 'seed': 42, 'camera': 'LSST', 'dither': True, 'survey_name': 'greedy'} surveys = [] detailer = detailers.Camera_rot_detailer(min_rot=np.min(camera_rot_limits), max_rot=np.max(camera_rot_limits)) wfd_halves = wfd_half() for filtername in filters: bfs = [] bfs.append((bf.M5_diff_basis_function(filtername=filtername, nside=nside), m5_weight)) bfs.append((bf.Footprint_basis_function(filtername=filtername, footprint=footprints, out_of_bounds_val=np.nan, nside=nside), footprint_weight)) bfs.append((bf.Slewtime_basis_function(filtername=filtername, nside=nside), slewtime_weight)) bfs.append((bf.Strict_filter_basis_function(filtername=filtername), stayfilter_weight)) bfs.append((bf.Map_modulo_basis_function(wfd_halves), roll_weight)) # Masks, give these 0 weight bfs.append((bf.Zenith_shadow_mask_basis_function(nside=nside, shadow_minutes=shadow_minutes, max_alt=max_alt), 0)) bfs.append((bf.Moon_avoidance_basis_function(nside=nside, moon_distance=moon_distance), 0)) bfs.append((bf.Filter_loaded_basis_function(filternames=filtername), 0)) bfs.append((bf.Planet_mask_basis_function(nside=nside), 0)) weights = [val[1] for val in bfs] basis_functions = [val[0] for val in bfs] surveys.append(Greedy_survey(basis_functions, weights, exptime=exptime, filtername=filtername, nside=nside, ignore_obs=ignore_obs, nexp=nexp, detailers=[detailer], **greed_survey_params)) return surveys def generate_blobs(nside, nexp=1, exptime=30., filter1s=['u', 'u', 'g', 'r', 'i', 'z', 'y'], filter2s=['g', 'r', 'r', 'i', 'z', 'y', 'y'], pair_time=22., camera_rot_limits=[-80., 80.], n_obs_template=3, season=300., season_start_hour=-4., season_end_hour=2., shadow_minutes=60., max_alt=76., moon_distance=30., ignore_obs='DD', m5_weight=6., footprint_weight=0.6, slewtime_weight=3., stayfilter_weight=3., template_weight=12., footprints=None, roll_weight=3.): """ Generate surveys that take observations in blobs. Parameters ---------- nside : int (32) The HEALpix nside to use nexp : int (1) The number of exposures to use in a visit. exptime : float (30.) The exposure time to use per visit (seconds) filter1s : list of str The filternames for the first set filter2s : list of str The filter names for the second in the pair (None if unpaired) pair_time : float (22) The ideal time between pairs (minutes) camera_rot_limits : list of float ([-80., 80.]) The limits to impose when rotationally dithering the camera (degrees). n_obs_template : int (3) The number of observations to take every season in each filter season : float (300) The length of season (i.e., how long before templates expire) (days) season_start_hour : float (-4.) For weighting how strongly a template image needs to be observed (hours) sesason_end_hour : float (2.) For weighting how strongly a template image needs to be observed (hours) shadow_minutes : float (60.) Used to mask regions around zenith (minutes) max_alt : float (76. The maximium altitude to use when masking zenith (degrees) moon_distance : float (30.) The mask radius to apply around the moon (degrees) ignore_obs : str or list of str ('DD') Ignore observations by surveys that include the given substring(s). m5_weight : float (3.) The weight for the 5-sigma depth difference basis function footprint_weight : float (0.3) The weight on the survey footprint basis function. slewtime_weight : float (3.) The weight on the slewtime basis function stayfilter_weight : float (3.) The weight on basis function that tries to stay avoid filter changes. template_weight : float (12.) The weight to place on getting image templates every season """ blob_survey_params = {'slew_approx': 7.5, 'filter_change_approx': 140., 'read_approx': 2., 'min_pair_time': 15., 'search_radius': 30., 'alt_max': 85., 'az_range': 90., 'flush_time': 30., 'smoothing_kernel': None, 'nside': nside, 'seed': 42, 'dither': True, 'twilight_scale': True} surveys = [] times_needed = [pair_time, pair_time*2] wfd_halves = wfd_half() # For the m5 depth, use r m5 for r and redder filter_translate = {'u': 'u', 'g': 'g', 'r': 'r', 'i': 'r', 'z': 'r', 'y': 'r'} for filtername, filtername2 in zip(filter1s, filter2s): detailer_list = [] detailer_list.append(detailers.Camera_rot_detailer(min_rot=np.min(camera_rot_limits), max_rot=np.max(camera_rot_limits))) detailer_list.append(detailers.Close_alt_detailer()) # List to hold tuples of (basis_function_object, weight) bfs = [] if filtername2 is not None: bfs.append((bf.M5_diff_basis_function(filtername=filter_translate[filtername], nside=nside), m5_weight/2.)) bfs.append((bf.M5_diff_basis_function(filtername=filter_translate[filtername2], nside=nside), m5_weight/2.)) else: bfs.append((bf.M5_diff_basis_function(filtername=filter_translate[filtername], nside=nside), m5_weight)) if filtername2 is not None: bfs.append((bf.Footprint_basis_function(filtername=filtername, footprint=footprints, out_of_bounds_val=np.nan, nside=nside), footprint_weight/2.)) bfs.append((bf.Footprint_basis_function(filtername=filtername2, footprint=footprints, out_of_bounds_val=np.nan, nside=nside), footprint_weight/2.)) else: bfs.append((bf.Footprint_basis_function(filtername=filtername, footprint=footprints, out_of_bounds_val=np.nan, nside=nside), footprint_weight)) bfs.append((bf.Slewtime_basis_function(filtername=filtername, nside=nside), slewtime_weight)) bfs.append((bf.Strict_filter_basis_function(filtername=filtername), stayfilter_weight)) if filtername2 is not None: bfs.append((bf.N_obs_per_year_basis_function(filtername=filtername, nside=nside, footprint=footprints.get_footprint(filtername), n_obs=n_obs_template, season=season, season_start_hour=season_start_hour, season_end_hour=season_end_hour), template_weight/2.)) bfs.append((bf.N_obs_per_year_basis_function(filtername=filtername2, nside=nside, footprint=footprints.get_footprint(filtername2), n_obs=n_obs_template, season=season, season_start_hour=season_start_hour, season_end_hour=season_end_hour), template_weight/2.)) else: bfs.append((bf.N_obs_per_year_basis_function(filtername=filtername, nside=nside, footprint=footprints.get_footprint(filtername), n_obs=n_obs_template, season=season, season_start_hour=season_start_hour, season_end_hour=season_end_hour), template_weight)) bfs.append((bf.Map_modulo_basis_function(wfd_halves), roll_weight)) # Masks, give these 0 weight bfs.append((bf.Zenith_shadow_mask_basis_function(nside=nside, shadow_minutes=shadow_minutes, max_alt=max_alt, penalty=np.nan, site='LSST'), 0.)) bfs.append((bf.Moon_avoidance_basis_function(nside=nside, moon_distance=moon_distance), 0.)) filternames = [fn for fn in [filtername, filtername2] if fn is not None] bfs.append((bf.Filter_loaded_basis_function(filternames=filternames), 0)) if filtername2 is None: time_needed = times_needed[0] else: time_needed = times_needed[1] bfs.append((bf.Time_to_twilight_basis_function(time_needed=time_needed), 0.)) bfs.append((bf.Not_twilight_basis_function(), 0.)) bfs.append((bf.Planet_mask_basis_function(nside=nside), 0.)) # unpack the basis functions and weights weights = [val[1] for val in bfs] basis_functions = [val[0] for val in bfs] if filtername2 is None: survey_name = 'blob, %s' % filtername else: survey_name = 'blob, %s%s' % (filtername, filtername2) if filtername2 is not None: detailer_list.append(detailers.Take_as_pairs_detailer(filtername=filtername2)) surveys.append(Blob_survey(basis_functions, weights, filtername1=filtername, filtername2=filtername2, exptime=exptime, ideal_pair_time=pair_time, survey_note=survey_name, ignore_obs=ignore_obs, nexp=nexp, detailers=detailer_list, **blob_survey_params)) return surveys def nes_light_footprints(nside=None): """ A quick function to generate the "standard" goal maps. This is the traditional WFD/mini survey footprint. """ NES_scaledown = 2. SCP_scaledown = 1.5 result = {} result['u'] = generate_goal_map(nside=nside, NES_fraction=0./NES_scaledown, WFD_fraction=0.31, SCP_fraction=0.15/SCP_scaledown, GP_fraction=0.15, wfd_dec_min=-62.5, wfd_dec_max=3.6) result['g'] = generate_goal_map(nside=nside, NES_fraction=0.2/NES_scaledown, WFD_fraction=0.44, SCP_fraction=0.15/SCP_scaledown, GP_fraction=0.15, wfd_dec_min=-62.5, wfd_dec_max=3.6) result['r'] = generate_goal_map(nside=nside, NES_fraction=0.46/NES_scaledown, WFD_fraction=1.0, SCP_fraction=0.15/SCP_scaledown, GP_fraction=0.15, wfd_dec_min=-62.5, wfd_dec_max=3.6) result['i'] = generate_goal_map(nside=nside, NES_fraction=0.46/NES_scaledown, WFD_fraction=1.0, SCP_fraction=0.15/SCP_scaledown, GP_fraction=0.15, wfd_dec_min=-62.5, wfd_dec_max=3.6) result['z'] = generate_goal_map(nside=nside, NES_fraction=0.4/NES_scaledown, WFD_fraction=0.9, SCP_fraction=0.15/SCP_scaledown, GP_fraction=0.15, wfd_dec_min=-62.5, wfd_dec_max=3.6) result['y'] = generate_goal_map(nside=nside, NES_fraction=0./NES_scaledown, WFD_fraction=0.9, SCP_fraction=0.15/SCP_scaledown, GP_fraction=0.15, wfd_dec_min=-62.5, wfd_dec_max=3.6) return result def run_sched(surveys, survey_length=365.25, nside=32, fileroot='baseline_', verbose=False, extra_info=None, illum_limit=40.): years = np.round(survey_length/365.25) scheduler = Core_scheduler(surveys, nside=nside) n_visit_limit = None filter_sched = simple_filter_sched(illum_limit=illum_limit) observatory = Model_observatory(nside=nside) observatory, scheduler, observations = sim_runner(observatory, scheduler, survey_length=survey_length, filename=fileroot+'%iyrs.db' % years, delete_past=True, n_visit_limit=n_visit_limit, verbose=verbose, extra_info=extra_info, filter_scheduler=filter_sched) if __name__ == "__main__": parser = argparse.ArgumentParser() parser.add_argument("--verbose", dest='verbose', action='store_true') parser.set_defaults(verbose=False) parser.add_argument("--survey_length", type=float, default=365.25*10) parser.add_argument("--outDir", type=str, default="") parser.add_argument("--maxDither", type=float, default=0.7, help="Dither size for DDFs (deg)") parser.add_argument("--moon_illum_limit", type=float, default=40., help="illumination limit to remove u-band") parser.add_argument("--nexp", type=int, default=1) parser.add_argument("--scale_down", dest='scale_down', action='store_true') parser.set_defaults(scale_down=False) args = parser.parse_args() survey_length = args.survey_length # Days outDir = args.outDir verbose = args.verbose max_dither = args.maxDither illum_limit = args.moon_illum_limit nexp = args.nexp scale_down = args.scale_down nside = 32 per_night = True # Dither DDF per night camera_ddf_rot_limit = 75. extra_info = {} exec_command = '' for arg in sys.argv: exec_command += ' ' + arg extra_info['exec command'] = exec_command try: extra_info['git hash'] = subprocess.check_output(['git', 'rev-parse', 'HEAD']) except subprocess.CalledProcessError: extra_info['git hash'] = 'Not in git repo' extra_info['file executed'] = os.path.realpath(__file__) fileroot = 'even_filters_alt' file_end = 'v1.6_' if scale_down: footprints_hp = nes_light_footprints(nside=nside) fileroot = fileroot +'scaleddown_' else: footprints_hp = standard_goals(nside=nside) observatory = Model_observatory(nside=nside) conditions = observatory.return_conditions() footprints = Footprint(conditions.mjd_start, sun_RA_start=conditions.sun_RA_start, nside=nside) for i, key in enumerate(footprints_hp): footprints.footprints[i, :] = footprints_hp[key] # Set up the DDF surveys to dither dither_detailer = detailers.Dither_detailer(per_night=per_night, max_dither=max_dither) details = [detailers.Camera_rot_detailer(min_rot=-camera_ddf_rot_limit, max_rot=camera_ddf_rot_limit), dither_detailer] ddfs = generate_dd_surveys(nside=nside, nexp=nexp, detailers=details) greedy = gen_greedy_surveys(nside, nexp=nexp, footprints=footprints) blobs = generate_blobs(nside, nexp=nexp, footprints=footprints) surveys = [ddfs, blobs, greedy] run_sched(surveys, survey_length=survey_length, verbose=verbose, fileroot=os.path.join(outDir, fileroot+file_end), extra_info=extra_info, nside=nside, illum_limit=illum_limit)