In [1]:
%matplotlib inline
In [1]:
import numpy as np
import matplotlib.pyplot as plt
from IPython.display import Image, display, HTML
import pandas as pd
pd.set_option('display.max_colwidth', -1)
from grizli import utils, __version__
from grizli.aws import db as grizli_db
try:
from mastquery import overlaps
except:
overlaps = None
print('Grizli version: ', __version__)
utils.set_warnings()
In [2]:
# readonly DB access
config = {'hostname': 'grizdbinstance.c3prl6czsxrm.us-east-1.rds.amazonaws.com',
'username': '***',
'password': '***',
'database': '***',
'port': 5432}
# sqlalchemy engine for postgresql
engine = grizli_db.get_db_engine(config=config)
print('Tables: ', engine.table_names())
redshift_fit.status:
3 - The object should have an extracted beams.fits file in the S3 filesystem
5 - extractions and redshift fits that timed out in the 15 minute Lambda limit,
e.g., some deep pointings or unconstrained fits that find many peaks on the second pass
6 : completed redshift fits with full information stored in the redshift_fit table.
In [3]:
print(grizli_db.FLAGS)
In [4]:
for table in ['charge_fields', 'photometry_apcorr', 'redshift_fit']:
photom = pd.read_sql_query('select * from {0} where false;'.format(table), engine)
cols = list(photom.columns)
count = pd.read_sql_query('select count(*) as n from {0};'.format(table), engine)
print('### Table `{0}`: {1} rows x {2} columns ###\n' .format(table,
count['n'][0], len(photom.columns)))
r, c = 6, 10
for i in range(c):
print(' '+', '.join(['{0:10}'.format(ci) for ci in cols[i*r:(i+1)*r]]))
print(' ...\n')
Here z_specs are taken from a large variety of literature sources.
I've taken separate tables and matched them into the photometry_apcorr database, but
this would better be done with a separate z_spec table in the database.
The make_html_table function below automatically does a join on redshift_fit and photometry_apcorr based on the object root + id combinations.
If you don't set get_sql=True, it makes a sortable HTML table like the link below sends it to the S3 target specified in sync.
In [5]:
extra = "AND status > 4 AND z_spec > 0 AND z_spec_qual = 1" # AND z_spec_src NOT LIKE '%%Caminha%%'"
extra += "AND (t_g102 > 10 OR t_g141 > 10)"
#extra += "AND (t_g800l > 0)"
SQLtxt = grizli_db.make_html_table(engine=engine,
columns=['root','status','id','ra','dec','mag_auto','flux_radius',
't_g800l','t_g102','t_g141',
'z_spec','z_map','z_spec_src','bic_diff','chinu','log_pdf_max',
'zwidth1/(1+z_map) as zw1','(z_map-z_spec)/(1+z_spec) as dz','q_z'],
where=extra,
table_root='zspec_delta', sync='s3://grizli-v1/tables/',
png_ext=['R30', 'stack','full','line'], get_sql=True)
print(SQLtxt.replace('FROM','\n FROM').replace('WHERE','\n WHERE'))
# Below is like `pd.read_sql_query` but sends result
# to an astropy table rather than a `DataFrame`
res = grizli_db.from_sql(SQLtxt, engine)
print('N:',len(res))
In [6]:
# z_grism vs z_spec
from matplotlib.ticker import FixedLocator, AutoLocator, MaxNLocator
import matplotlib as mpl
# Pseudolog in log(1+z)
xti = xt = np.arange(0,6.6,0.5)
xti = list(xti)
for i in range(len(xti)):
if (xti[i] > 3) & (xti[i] != int(xti[i])):
xti[i] = ''
loc = np.arange(0, 3.6, 0.1)
fig = plt.figure(figsize=[8,7])
ax = fig.add_subplot(111)
if False:
print('Color by zwidth')
cby = 'zw'
kwargs = {'marker':'.', 'vmin':-3.5, 'vmax': -0.5, 'cmap': 'plasma'}
c = np.log10(res['zw1'])
so = np.argsort(c)[::-1]
else:
print('Color by q_z')
cby = 'q_z'
cmap = plt.cm.plasma_r
bounds = [-2, -1.5, -1, -0.7, -0.3, 0]
norm = mpl.colors.BoundaryNorm(bounds, cmap.N)
kwargs = {'marker':'+', 'vmin':-1.5, 'vmax': 0, 'cmap': cmap, 'norm':norm}
c = res['q_z']
so = np.argsort(c)
ax.scatter(np.log(1+res['z_spec'][so]), np.log(1+res['z_map'][so]), alpha=0.2,
c=c[so], **kwargs)
# dummy replot for darker colorbar
sc = ax.scatter(np.log([1]), np.log([1]), alpha=0.8, c=[0], **kwargs)
cb = plt.colorbar(sc, shrink=0.6)
if cby == 'zw':
cb.set_label(r'$(z_{84}-z_{16})/(1+z_{50})$')
cb.set_ticks([-3,-2,-1])
cb.set_ticklabels([0.001, 0.01, 0.1])
elif cby == 'q_z':
cb.set_label(r'q_z')
cb.set_ticks([-1, -0.7, 0])
#cb.set_ticklabels([0.001, 0.01, 0.1])
xts = ax.set_xticks(np.log(1+xt))
xtl = ax.set_xticklabels(xti)
xts = ax.set_yticks(np.log(1+xt))
xtl = ax.set_yticklabels(xti)
ax.set_xlim(0, np.log(1+6.5))
ax.set_ylim(0, np.log(1+6.5))
ax.xaxis.set_minor_locator(FixedLocator(np.log(1+loc)))
ax.yaxis.set_minor_locator(FixedLocator(np.log(1+loc)))
ax.set_xlabel('z_spec')
ax.set_ylabel('z_MAP')
ax.set_aspect(1)
ax.grid()
ax.text(0.95, 0.1, r'$N={0}$'.format(len(res)), ha='right', va='bottom',
transform=ax.transAxes)
dz = (res['z_map'] - res['z_spec'])/(1+res['z_spec'])
#sel = (res['z_spec'] > 0) & (np.abs(dz) < 0.1)
#sel = (res['z_spec'] > 0) & (res['zw1'] < 0.05)
sel = (res['z_spec'] > 0) & (res['q_z'] > -1.0)
ax.text(0.95, 0.05, r'$\sigma_\mathrm{{{{nmad}}}}={0:.4f}$ ($N={1}$)'.format(utils.nmad(dz[sel]), sel.sum()),
ha='right', va='bottom', transform=ax.transAxes)
ax.plot(ax.get_xlim(), ax.get_xlim(), color='k',
alpha=0.2, linewidth=1, zorder=-10)
fig.tight_layout(pad=0.1)
In [7]:
# Count redshift sources
if 0:
print('z_spec references, in both imaging and spectroscopy')
SQL = """
SELECT * FROM (SELECT z_spec_src, count(z_spec_src) AS n FROM photometry_apcorr
WHERE z_spec_qual = 1 GROUP BY z_spec_src) foo ORDER BY n DESC;
"""
else:
# overlap with spectroscopy
print('z_spec references, overlap with spectroscopy')
SQL = """
SELECT * FROM (SELECT z_spec_src, count(z_spec_src) AS n FROM redshift_fit, photometry_apcorr
WHERE root = p_root AND id = p_id AND z_spec_qual = 1 GROUP BY z_spec_src) foo ORDER BY n DESC;
"""
counts = grizli_db.from_sql(SQL, engine)
print('N_src={0}, N_tot={1}\n'.format(len(counts), counts['n'].sum()))
nr = 40
msg = '{0:>26} {1:4} {2:>26} {3:4}'
print(msg.format('z_spec_src','n','z_spec_src','n'))
print('='*80)
for i in range(nr):
print(msg.format(counts['z_spec_src'][i], counts['n'][i],
counts['z_spec_src'][i+nr], counts['n'][i+nr]))
In [8]:
# Show where the sources are on the sky
fig = plt.figure(figsize=[10, 5])
ax = fig.add_subplot(111)
for src in counts['z_spec_src'][0:20]:
sub = res['z_spec_src'] == src
ax.scatter(res['ra'][sub], res['dec'][sub], marker='s', s=20,
label=src, alpha=0.5)
ax.set_aspect(1)
ax.set_xlim(-10, 370); ax.set_ylim(-80, 80)
ax.set_xlabel('RA'); ax.set_ylabel('Dec.')
ax.grid()
ax.legend(loc='lower center', fontsize=7, ncol=4)
Out[8]:
In [9]:
# Individual object: high-z quasar
where = "AND status > 4 AND z_spec > 6.0 AND z_spec_qual = 1 AND z_spec_src = 'Wang 2016 high-z QSO'"
SQLtxt = grizli_db.make_html_table(engine=engine,
columns=['status','root','id','ra','dec','mag_auto','flux_radius',
'z_spec','z_map','z_spec_src'],
where=where, get_sql=True)
print(SQLtxt.replace('FROM','\n FROM').replace('WHERE','\n WHERE'))
# `from_sql` is like `pd.read_sql_query` but sends result
# to an astropy table rather than a `DataFrame`
highz = grizli_db.from_sql(SQLtxt, engine)
In [10]:
highz
Out[10]:
In [11]:
HTML(grizli_db.render_for_notebook(highz['root','id','ra','dec','mag_auto','z_spec','z_map'],
image_extensions=['stack','full']))
Out[11]:
Every object in the redshift_fit table has 1D spectra stored in additional tables spec1d_g800l, spec1d_g102, and spec1d_g141.
The columns of those tables are like:
_flux: Extracted flux_err: Uncertainty_flat: Model spectrum for a flat spectrum with flux 1 erg/s/cm2/A used for flux calibration_contam: Contamination_cont: Best-fit template, continuum only_line: Best-fit templates, line plus continuumThe units of the spectra are the native detector electrons per second. To convert to cgs flux densities, divide by the _flat column.
In [12]:
# 1D spectra
tablename = 'spec1d_g102'
SQL = """SELECT root, id, z_map, sp.* from redshift_fit, {1} as sp
WHERE root = '{0}' AND id = {2} AND {1}_root = root AND {1}_id = id""".format(highz['root'][0],
tablename, highz['id'][0])
resp = pd.read_sql_query(SQL, engine)
# Wavelengths stored in a separate table
wave = pd.read_sql_query("SELECT * from {0}_wave".format(tablename), engine)[tablename+'_wave'].values
# Plot the spectra
data = {}
for c in ['flux','err','flat','contam','line','cont']:
data[c] = np.array(resp['{0}_{1}'.format(tablename, c)][0])
rest_wave = wave / (1+resp['z_map'][0])
fig = plt.figure(figsize=[8,4])
ax = fig.add_subplot(111)
er = ax.errorbar(rest_wave, data['flux']/data['flat'], data['err']/data['flat'],
marker='.', color='k', linestyle='None', alpha=0.5, label='Data', zorder=-1)
pl = ax.plot(rest_wave, data['contam']/data['flat'], color='orange', linestyle=':',
alpha=0.8, linewidth=2, label = 'Contamination')
pl = ax.plot(rest_wave, data['cont']/data['flat'], color='r',
alpha=0.8, linewidth=2, label = 'Continuum model')
ax.set_xlabel(r'$\lambda$ ($\mathrm{\AA}$)')
ax.set_ylabel(r'$f_\lambda$, cgs')
ax.legend(loc='upper right', fontsize=8)
ax.grid()
In [13]:
# Quasars
sel = (res['z_spec_src'] == 'sdss-dr15-QSO') | (res['z_spec_src'] == 'Veron 2006 QSO')
sel |= (res['z_spec_src'] == 'KODIAQ')
sel |= np.array(['lqac' in src for src in res['z_spec_src']])
sel |= np.array(['QSO' in src for src in res['z_spec_src']])
print('{0} Quasars from:\n {1}'.format(sel.sum(), '\n '.join(np.unique(res['z_spec_src'][sel]))))
# Rigby et al. "MegaSaura" Arc
sel = (res['z_spec_src'] == 'Rigby 2018')
print('\n{0} Megasaura arc images'.format(sel.sum()))
In [14]:
HTML(grizli_db.render_for_notebook(res[sel]['root','id','ra','dec','mag_auto','z_spec','z_map'],
image_extensions=['stack','full','line']))
Out[14]:
In [15]:
# G102 + G141
query = '%%-grism%%'
# G800L
#query = '%%-g800l%%'
SQLtxt = """
SELECT status,root,id,p_ra as ra,p_dec as dec,mag_auto,flux_radius,
t_g800l,t_g141,t_g102,z_spec,z_map,z_spec_src,z_spec_qual,
bic_diff,zwidth1/(1+z_map) as zw1,
zwidth2/(1+z_map) as zw2,
z02,z16,z50,z84,z97
FROM photometry_apcorr, redshift_fit
WHERE phot_root = p_root AND id = p_id AND status > 4
AND root LIKE '{0}';
""".format(query)
print(SQLtxt)
tab = grizli_db.from_sql(SQLtxt, engine)
fields = np.unique(tab['root'])
print('N total: {0} \nN fields: {1}\n'.format(len(tab), len(fields)))
print('\n'.join(fields))
In [16]:
# mag limit
mag_lim = 24
msel = tab['mag_auto'] < mag_lim
# Full photometric catalog
SQLphot = """
SELECT p_root as root, p_ra as ra, p_dec as dec, mag_auto
FROM photometry_apcorr
WHERE p_root LIKE '{0}' AND mag_auto < {1} ORDER BY mag_auto;
"""
extra = {}
for field in fields:
extra[field] = grizli_db.from_sql(SQLphot.format(field.replace('-g800l','-grism'), mag_lim), engine)
print(field, len(extra[field]))
In [17]:
ct = 'g141'
if ct == 'g800l':
print('Color on G800l exposure time')
c = (np.log10(tab['t_g800l']),3,5)
elif ct == 'g141':
print('Color on G141 exposure time')
c = (np.log10(tab['t_g141']),3,5)
elif ct == 'g102':
print('Color on G102 exposure time')
c = (np.log10(tab['t_g102']),3,5)
else:
print('Color on redshift fit status (5=timeout, 6=complete)')
c = (tab['status'],4.5,6.2)
In [18]:
# Show the field layouts
for field in fields:
fsel = (tab['root'] == field)
f_i = tab[fsel & msel]
so = np.argsort(tab['status'][fsel & msel])
f_i = f_i[so]
# All sources from parent photometry database
ex = extra[field]
sk = len(ex)//5000 # skip
sk = 1 if sk == 0 else sk
msg = 'field: {0}, n_grism, (< {3:.0f}) = {1:4} ({2:4}), n_full = {4:4}'
print(msg.format(field, fsel.sum(), (fsel & msel).sum(), mag_lim, len(ex)))
dy = (ex['dec'].max()-ex['dec'].min())*60
dx = (ex['ra'].max()-ex['ra'].min())*60*np.cos(ex['dec'][0]/180*np.pi)
s = np.clip(5*(np.maximum(dx, dy)/20), 5, 10)
fig = plt.figure(figsize=[s,s])
ax = fig.add_subplot(111)
ax.scatter(ex['ra'][::sk], ex['dec'][::sk], marker='.', color='k',
alpha=0.05, zorder=-1, label='photometry_apcorr')
ax.scatter(f_i['ra'], f_i['dec'], c=c[0][fsel & msel][so], alpha=0.2,
marker='s', s=20, vmin=c[1], vmax=c[2], cmap='magma',
label='`redshift_fit.{0}`'.format(c[0].name))
ax.set_aspect(1./np.cos(f_i['dec'][0]/180*np.pi))
ax.set_xlim(ax.get_xlim()[::-1])
ax.set_xlabel('RA'); ax.set_ylabel('Dec.')
fig.tight_layout(pad=0.2)
if overlaps is not None:
overlaps.draw_axis_labels(ax=ax, nlabel=3, format='latex')
leg = ax.legend(title=field, loc='lower left', fontsize=8)
# ax.text(0.03, 0.03, field, ha='left', va='bottom',
# transform=ax.transAxes, fontsize=9,
# bbox={'facecolor':'w', 'alpha':0.5, 'edgecolor':'None'})
ax.grid()
In [19]:
# Get field information from the charge database
pi_names = ['Bielby']
#pi_names = ['Lotz'] # GLASS on FFs
field_query = 'AND (' + 'OR '.join(["proposal_pi LIKE '%%{0}%%'".format(pi) for pi in pi_names]) + ')'
## other test query
# field_query = "AND (field_t_g141 > 4000 AND field_t_g102 > 4000 AND field_t_g141 < 8000"
# field_query += "AND comment = 'WSP' AND gaia5 < 10)"
SQLtxt = """
SELECT root,target,proposal_pi,count(root) as n
FROM photometry_apcorr, redshift_fit, charge_fields
WHERE phot_root = p_root AND id = p_id AND field_root = p_root AND status > 4 {field_query}
GROUP BY root,target,proposal_pi;
""".format(field_query=field_query)
print(SQLtxt)
parent = grizli_db.from_sql(SQLtxt, engine)
print('N total: {0} \nN fields: {1}'.format(len(parent), len(np.unique(parent['root']))))
parent
Out[19]:
In [20]:
# Select histogram on grism redshifts, otherwise try to use z_spec
GRISM_HISTOGRAM = False
In [21]:
########## Make field layout plots with optional histograms
SQLphot = """
SELECT p_root as root, p_ra as ra, p_dec as dec, mag_auto, z_spec, z_spec_src
FROM photometry_apcorr
WHERE p_root LIKE '{0}' AND mag_auto < 24 ORDER BY mag_auto;
"""
SQLroot = """
SELECT status,phot_root,root,id,p_ra as ra,p_dec as dec,mag_auto,flux_radius,
t_g141,t_g102,z_spec,z_map,z_spec_src,z_spec_qual,
bic_diff,zwidth1/(1+z_map) as zw1,
zwidth2/(1+z_map) as zw2,
z02,z16,z50,z84,z97
FROM photometry_apcorr, redshift_fit
WHERE phot_root = p_root AND id = p_id AND status > 4 AND (root = '{0}') AND q_z > -1.0;"""
fields = np.unique(parent['root'])
# bins for histograms
bins = utils.log_zgrid([0.01, 3.5], 0.005)
# Coarse binning for maxima
spbins = utils.log_zgrid([0.01, 3.5], 0.03)
URL = 'https://s3.amazonaws.com/grizli-v1/Pipeline/{0}/Extractions/{0}.html?&z_map_min={1:.04f}&z_map_max={2:.04f}'
URL += '&bic_diff_min=10&chinu_max=2'
links = []
for field in fields[:10]:
tab = grizli_db.from_sql(SQLroot.format(field), engine)
msel = tab['mag_auto'] < 25
#qso = tab['z_spec_src'].filled('---') == 'sdss-dr15-QSO'
try:
qso = (tab['z_spec_qual'].filled(0) == 1) #== 'sdss-dr15-QSO'
qso &= ~utils.column_string_operation(tab['z_spec_src'].filled('---'), 'erb', 'startswith')
except:
qso = (tab['z_spec_qual'] == 1) #== 'sdss-dr15-QSO'
qso &= ~utils.column_string_operation(tab['z_spec_src'], 'erb', 'startswith')
# Color by status
c = (tab['status'], 4.5, 6.2)
# Color by grism exposure time
#c = (np.log10(tab['t_g800l'].filled(1) + tab['t_g141'].filled(1) + tab['t_g102'].filled(1)), 3, 5)
fsel = (tab['root'].filled() == field)
fq = fsel & qso
f_i = tab[fsel & msel]
so = np.argsort(tab['status'][fsel & msel])
f_i = f_i[so]
# All sources from parent photometry database
ex = grizli_db.from_sql(SQLphot.format(tab['phot_root'][0]), engine)
sk = len(ex)//5000 # skip
sk = 1 if sk == 0 else sk
# Figure size
dy = (ex['dec'].max()-ex['dec'].min())*60
dx = (ex['ra'].max()-ex['ra'].min())*60*np.cos(ex['dec'][0]/180*np.pi)
s = np.clip(8*(np.maximum(dx, dy)/20), 7, 14)
fig = plt.figure(figsize=[s,s])
ax = fig.add_subplot(111)
# All sources
ax.scatter(ex['ra'][::sk], ex['dec'][::sk], marker='.', color='k',
alpha=0.05, zorder=-1, label=None)
# grism sources
ax.scatter(f_i['ra'], f_i['dec'], c=c[0][fsel & msel][so], alpha=0.2,
marker='.', vmin=c[1], vmax=c[2], cmap='magma', label=None)
ax.set_aspect(1./np.cos(f_i['dec'][0]/180*np.pi))
ax.set_xlim(ax.get_xlim()[::-1])
ax.set_xlabel('RA'); ax.set_ylabel('Dec.')
fig.tight_layout(pad=0.2)
#overlaps.draw_axis_labels(ax=ax, nlabel=3, format='latex')
# Quality redshifts
qual = (f_i['bic_diff'] > 10) | (f_i['zw1'] < 0.02)
# Show 4 redshift peaks
max_npeak = 3
d1, d2 = 0.5, 0.18
ax2 = ax.inset_axes((1-d1, 1-d2, d1, d2))
xt = []
# Spectroscopic redshifts from parent catalog
nmin = 1
zcol = 'z_spec'
zhist = ex['ra', 'dec', 'z_spec', 'z_spec_src'][ex['z_spec'] > 0]
if GRISM_HISTOGRAM | (len(zhist) == 0):
# Grism objects
fq = fsel & ((tab['bic_diff'] > 10) | (tab['zw1'] < 0.02))
nmin = 6
zcol = 'z_map'
zhist = tab[fq]['ra', 'dec', zcol]
print('{0} Ntot={1} Nmag={2} Nphot={3} n({4})={5}'.format(field, fsel.sum(), (fsel & msel).sum(),
len(ex), zcol, len(zhist)))
yh, xh = np.histogram(zhist[zcol], bins=spbins)
bso = np.argsort(yh)[::-1]
for i, ib in enumerate(bso):
if (yh[ib] < nmin) | (i > max_npeak):
break
ii = (zhist[zcol] >= xh[ib]) & (zhist[zcol] < xh[ib+1])
zsp = zhist[zcol][ii].mean()
if zcol == 'z_spec':
zsrc = '\n'+','.join(list(np.unique(zhist['z_spec_src'][ii])))
else:
zsrc = '\nself(grism)'
plw = ax2.plot([np.log(1+zsp)]*2, [15, 20], color='w', linewidth=4, alpha=0.9, zorder=8)
pl = ax2.plot([np.log(1+zsp)]*2, [15, 20], linewidth=1, alpha=0.9, zorder=10)
color_i = pl[0].get_color()
b_i = np.abs((bins-zsp)/(1+zsp)) < 0.1
ax2.hist(np.log(1+f_i['z_map'][qual]), bins=np.log(1+bins[b_i]), color='0.7', zorder=-5)
xt.append(zsp)
# Plot positions and histogram of objects near the z_spec (QSO) objects
near = qual & (np.abs((f_i['z_map']-zsp)/(1+zsp)) < 0.01)
ax2.hist(np.log(1+f_i['z_map'][near]), bins=np.log(1+bins[b_i]), color=color_i, alpha=0.8)
label = 'z={0:.3f} N={1}{2}'.format(zsp, near.sum(), zsrc)
dzi = 0.01*(1+zsp)
URLi = URL.format(field, zsp-dzi, zsp+dzi)
links.append(HTML('<a href={0}>{1} / {2}</a>'.format(URLi, field, label)))
if (zcol == 'z_spec') & (ii.sum() == 1):
ax.scatter(zhist['ra'][ii][:2], zhist['dec'][ii][:2],
edgecolor=color_i, facecolor='None', alpha=1,
s=220, marker='*', zorder=100, linewidth=2,
label=label)
else:
ax.scatter(zhist['ra'][ii].mean(), zhist['dec'][ii].mean(),
edgecolor=color_i, facecolor='None', alpha=1,
s=300, marker='o', zorder=100, linewidth=2, linestyle='--',
label=label)
ax.scatter(f_i['ra'][near], f_i['dec'][near], marker='x',
color=color_i, alpha=0.8, zorder=1000,
label=None)
ax2.set_xticks(np.log(1+np.array(xt)));
ax2.set_xticklabels(['{0:.2f}'.format(xti) for xti in xt], rotation=45*(len(xt) > 1))
ax2.set_ylim(0,20); ax2.set_yticks([0,5,10,15])
ax2.grid()
ax2.tick_params(labelsize=8)
ax2.xaxis.set_ticks_position('top')
ax2.yaxis.set_ticks_position('right')
ax.legend(loc='lower left', fontsize=8, title=field, ncol=1+(ib > 2)*1)
else:
ax.text(0.03, 0.03, field, ha='left', va='bottom',
transform=ax.transAxes, fontsize=9,
bbox={'facecolor':'w', 'alpha':0.5, 'edgecolor':'None'})
if overlaps is not None:
overlaps.draw_axis_labels(ax=ax, nlabel=3, format='latex')
ax.grid()
# Show URLS
display(*links)
In [22]:
# Fields around SDSS QSOs
# Quasars
# sdss-dr15-QSO
SQLtxt = """
SELECT DISTINCT root
FROM photometry_apcorr, redshift_fit
WHERE phot_root = p_root AND id = p_id
AND (z_spec_src = 'sdss-dr15-QSO' OR z_spec_src = 'LBQS' OR z_spec_src = 'KODIAQ')
AND status > 4 AND root LIKE 'j%%';
"""
print(SQLtxt)
parent = grizli_db.from_sql(SQLtxt, engine)
print('N total: {0} \nN fields: {1}'.format(len(parent), len(np.unique(parent['root']))))
In [23]:
SQLtxt = """
SELECT * FROM (SELECT root,target,field_t_g800l,field_t_g141,proposal_pi,count(root) as n
FROM redshift_fit, charge_fields
WHERE field_root = phot_root AND status=5 {field_query} AND gaia5 < 10
GROUP BY root,target,proposal_pi,field_t_g141,field_t_g800l) as sel WHERE n > 25 ORDER BY n DESC;
""".format(field_query='')
print(SQLtxt)
parent = grizli_db.from_sql(SQLtxt, engine)
print('N total: {0} \nN fields: {1}'.format(len(parent), len(np.unique(parent['root']))))
print('total timeout: ', parent['n'].sum())
parent
Out[23]: