Second Round of Full Day RFI Flagging¶
by Josh Dillon, last updated October 13, 2024
This notebook is synthesizes information from individual delay_filtered_average_zscore notebooks to find low-level RFI and flag it. That notebook takes smooth_calibrated data, redundantly averages it, performs a high-pass delay filter, and then incoherently averages across baselines, creating a per-polarization z-score. This notebook then takes that whole night of z-scores and finds a new set of flags to both add to the smooth_cal files, which are updated in place, and to write down as new UVFlag waterfall-type .h5 files.
Here's a set of links to skip to particular figures and tables:
• Figure 1: Waterfall of Maximum z-Score of Either Polarization Before Round 2 Flagging¶
• Figure 2: Histogram of z-scores¶
• Figure 3: Waterfall of Maximum z-Score of Either Polarization After Round 2 Flagging¶
• Figure 4: Spectra of Time-Averaged z-Scores¶
• Figure 5: Summary of Flags Before and After Round 2 Flagging¶
In [1]:
import time
tstart = time.time()
In [2]:
import os
os.environ['HDF5_USE_FILE_LOCKING'] = 'FALSE'
import h5py
import hdf5plugin # REQUIRED to have the compression plugins available
import numpy as np
import glob
import matplotlib.pyplot as plt
import matplotlib
import copy
import warnings
from pyuvdata import UVFlag, UVCal
from hera_cal import utils
from hera_qm import xrfi
from hera_qm.time_series_metrics import true_stretches
from hera_filters import dspec
from IPython.display import display, HTML
%matplotlib inline
display(HTML("<style>.container { width:100% !important; }</style>"))
_ = np.seterr(all='ignore') # get rid of red warnings
%config InlineBackend.figure_format = 'retina'
In [3]:
# get input data file names
SUM_FILE = os.environ.get("SUM_FILE", None)
# SUM_FILE = '/lustre/aoc/projects/hera/h6c-analysis/IDR2/2459861/zen.2459861.25297.sum.uvh5'
SUM_SUFFIX = os.environ.get("SUM_SUFFIX", 'sum.uvh5')
# get input and output suffixes
SMOOTH_CAL_SUFFIX = os.environ.get("SMOOTH_CAL_SUFFIX", 'sum.smooth.calfits')
ZSCORE_SUFFIX = os.environ.get("ZSCORE_SUFFIX", 'sum.red_avg_zscore.h5')
FLAG_WATERFALL2_SUFFIX = os.environ.get("FLAG_WATERFALL2_SUFFIX", 'sum.flag_waterfall_round_2.h5')
OUT_YAML_SUFFIX = os.environ.get("OUT_YAML_SUFFIX", '_aposteriori_flags.yaml')
OUT_YAML_DIR = os.environ.get("OUT_YAML_DIR", None)
# build globs
sum_glob = '.'.join(SUM_FILE.split('.')[:-3]) + '.*.' + SUM_SUFFIX
cal_files_glob = sum_glob.replace(SUM_SUFFIX, SMOOTH_CAL_SUFFIX)
zscore_glob = sum_glob.replace(SUM_SUFFIX, ZSCORE_SUFFIX)
# build out yaml file
if OUT_YAML_DIR is None:
OUT_YAML_DIR = os.path.dirname(SUM_FILE)
out_yaml_file = os.path.join(OUT_YAML_DIR, SUM_FILE.split('.')[-4] + OUT_YAML_SUFFIX)
# get flagging parameters
Z_THRESH = float(os.environ.get("Z_THRESH", 4))
WS_Z_THRESH = float(os.environ.get("WS_Z_THRESH", 2))
AVG_Z_THRESH = float(os.environ.get("AVG_Z_THRESH", 1))
MAX_FREQ_FLAG_FRAC = float(os.environ.get("MAX_FREQ_FLAG_FRAC", .25))
MAX_TIME_FLAG_FRAC = float(os.environ.get("MAX_TIME_FLAG_FRAC", .1))
AVG_SPECTRUM_FILTER_DELAY = float(os.environ.get("AVG_SPECTRUM_FILTER_DELAY", 250)) # in ns
EIGENVAL_CUTOFF = float(os.environ.get("EIGENVAL_CUTOFF", 1e-12))
TIME_AVG_DELAY_FILT_SNR_THRESH = float(os.environ.get("TIME_AVG_DELAY_FILT_SNR_THRESH", 4.0))
TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE = float(os.environ.get("TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE", 1.5))
for setting in ['Z_THRESH', 'WS_Z_THRESH', 'AVG_Z_THRESH', 'MAX_FREQ_FLAG_FRAC', 'MAX_TIME_FLAG_FRAC', 'AVG_SPECTRUM_FILTER_DELAY',
'EIGENVAL_CUTOFF', 'TIME_AVG_DELAY_FILT_SNR_THRESH', 'TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE']:
print(f'{setting} = {eval(setting)}')
Z_THRESH = 4.0 WS_Z_THRESH = 2.0 AVG_Z_THRESH = 1.0 MAX_FREQ_FLAG_FRAC = 0.25 MAX_TIME_FLAG_FRAC = 0.1 AVG_SPECTRUM_FILTER_DELAY = 250.0 EIGENVAL_CUTOFF = 1e-12 TIME_AVG_DELAY_FILT_SNR_THRESH = 4.0 TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE = 1.5
Load z-scores¶
In [4]:
# load z-scores
zscore_files = sorted(glob.glob(zscore_glob))
print(f'Found {len(zscore_files)} *.{ZSCORE_SUFFIX} files starting with {zscore_files[0]}.')
uvf = UVFlag(zscore_files, use_future_array_shapes=True)
Found 1571 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data2/2460792/zen.2460792.21084.sum.red_avg_zscore.h5.
In [5]:
# get calibration solution files
cal_files = sorted(glob.glob(cal_files_glob))
print(f'Found {len(cal_files)} *.{SMOOTH_CAL_SUFFIX} files starting with {cal_files[0]}.')
Found 1571 *.sum.smooth.calfits files starting with /mnt/sn1/data2/2460792/zen.2460792.21084.sum.smooth.calfits.
In [6]:
assert len(zscore_files) == len(cal_files)
In [7]:
# extract z-scores and correct by a single number per polarization to account for biases created by filtering
zscore = {pol: uvf.metric_array[:, :, np.argwhere(uvf.polarization_array == utils.polstr2num(pol, x_orientation=uvf.x_orientation))[0][0]] for pol in ['ee', 'nn']}
zscore = {pol: zscore[pol] - np.nanmedian(zscore[pol]) for pol in zscore}
In [8]:
freqs = uvf.freq_array
times = uvf.time_array
In [9]:
extent = [freqs[0] / 1e6, freqs[-1] / 1e6, times[-1] - int(times[0]), times[0] - int(times[0])]
In [10]:
def plot_max_z_score(zscore, flags=None, vmin=-5, vmax=5):
if flags is None:
flags = np.any(~np.isfinite(list(zscore.values())), axis=0)
plt.figure(figsize=(14,10), dpi=100)
plt.imshow(np.where(flags, np.nan, np.nanmax([zscore['ee'], zscore['nn']], axis=0)), aspect='auto',
cmap='coolwarm', interpolation='none', vmin=vmin, vmax=vmax, extent=extent)
plt.colorbar(location='top', label='Max z-score of either polarization', extend='both', aspect=40, pad=.02)
plt.xlabel('Frequency (MHz)')
plt.ylabel(f'JD - {int(times[0])}')
plt.tight_layout()
Figure 1: Waterfall of Maximum z-Score of Either Polarization Before Round 2 Flagging¶
Shows the worse of the two results from delay_filtered_average_zscore from either polarization. Dips near flagged channels are expected, due to overfitting of noise. Positive-going excursions are problematic and likely evidence of RFI.
In [11]:
plot_max_z_score(zscore)
All-NaN axis encountered
In [12]:
def plot_histogram():
plt.figure(figsize=(14,4), dpi=100)
bins = np.arange(-50, 100, .1)
hist_ee = plt.hist(np.ravel(zscore['ee']), bins=bins, density=True, label='ee-polarized z-scores', alpha=.5)
hist_nn = plt.hist(np.ravel(zscore['nn']), bins=bins, density=True, label='nn-polarized z-scores', alpha=.5)
plt.plot(bins, (2*np.pi)**-.5 * np.exp(-bins**2 / 2), 'k:', label='Gaussian approximate\nnoise-only distribution')
plt.axvline(WS_Z_THRESH, c='r', ls='--', label='Watershed z-score')
plt.axvline(Z_THRESH, c='r', ls='-', label='Threshold z-score')
plt.yscale('log')
all_densities = np.concatenate([hist_ee[0][hist_ee[0] > 0], hist_nn[0][hist_nn[0] > 0]])
plt.ylim(np.min(all_densities) / 2, np.max(all_densities) * 2)
plt.xlim([-50, 100])
plt.legend()
plt.xlabel('z-score')
plt.ylabel('Density')
plt.tight_layout()
Figure 2: Histogram of z-scores¶
Shows a comparison of the histogram of z-scores in this file (one per polarization) to a Gaussian approximation of what one might expect from thermal noise. Without filtering, the actual distribution is a weighted sum of Rayleigh distributions. Filtering further complicates this. To make the z-scores more reliable, a single per-polarization median is subtracted from each waterfall, which allows us to flag low-level outliers with more confidence. Any points beyond the solid red line are flagged. Any points neighboring a flag beyond the dashed red line are also flagged. Finally, flagging is performed for low-level outliers in whole times or channels.
In [13]:
plot_histogram()
Perform flagging¶
In [14]:
def iteratively_flag_on_averaged_zscore(flags, zscore, avg_func=np.nanmean, avg_z_thresh=AVG_Z_THRESH, verbose=True):
'''Flag whole integrations or channels based on average z-score. This is done
iteratively to prevent bad times affecting channel averages or vice versa.'''
flagged_chan_count = 0
flagged_int_count = 0
while True:
zspec = avg_func(np.where(flags, np.nan, zscore), axis=0)
ztseries = avg_func(np.where(flags, np.nan, zscore), axis=1)
if (np.nanmax(zspec) < avg_z_thresh) and (np.nanmax(ztseries) < avg_z_thresh):
break
if np.nanmax(zspec) >= np.nanmax(ztseries):
flagged_chan_count += np.sum((zspec >= np.nanmax(ztseries)) & (zspec >= avg_z_thresh))
flags[:, (zspec >= np.nanmax(ztseries)) & (zspec >= avg_z_thresh)] = True
else:
flagged_int_count += np.sum((ztseries >= np.nanmax(zspec)) & (ztseries >= avg_z_thresh))
flags[(ztseries >= np.nanmax(zspec)) & (ztseries >= avg_z_thresh), :] = True
if verbose:
print(f'\tFlagging an additional {flagged_int_count} integrations and {flagged_chan_count} channels.')
def impose_max_chan_flag_frac(flags, max_flag_frac=MAX_FREQ_FLAG_FRAC, verbose=True):
'''Flag channels already flagged more than max_flag_frac (excluding completely flagged times).'''
unflagged_times = ~np.all(flags, axis=1)
frequently_flagged_chans = np.mean(flags[unflagged_times, :], axis=0) >= max_flag_frac
if verbose:
print(f'\tFlagging {np.sum(frequently_flagged_chans) - np.sum(np.all(flags, axis=0))} channels previously flagged {max_flag_frac:.2%} or more.')
flags[:, frequently_flagged_chans] = True
def impose_max_time_flag_frac(flags, max_flag_frac=MAX_TIME_FLAG_FRAC, verbose=True):
'''Flag times already flagged more than max_flag_frac (excluding completely flagged channels).'''
unflagged_chans = ~np.all(flags, axis=0)
frequently_flagged_times = np.mean(flags[:, unflagged_chans], axis=1) >= max_flag_frac
if verbose:
print(f'\tFlagging {np.sum(frequently_flagged_times) - np.sum(np.all(flags, axis=1))} times previously flagged {max_flag_frac:.2%} or more.')
flags[frequently_flagged_times, :] = True
def time_avg_zscore_dly_filt_SNRs(flags, filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF):
"""Produces SNRs after time-averaging z-scores and delay filtering, accounting for flagging's effect on the filter."""
# figure out high and low band based on FM gap at 100 MHz
flagged_stretches = true_stretches(np.all(flags, axis=0))
FM_gap = [fs for fs in flagged_stretches if fs.start <= np.argmin(np.abs(freqs - 100e6)) < fs.stop][0]
low_band = slice((0 if flagged_stretches[0].start != 0 else flagged_stretches[0].stop), FM_gap.start)
high_band = slice(FM_gap.stop, (len(freqs) if flagged_stretches[-1].stop != len(freqs) else flagged_stretches[-1].start))
filt_SNR = {}
for pol in zscore:
# calculate timeavg_SNR and filter
noise_prediction = 1.0 / np.sum(~flags, axis=0)**.5
timeavg_SNR = np.nanmean(np.where(flags, np.nan, zscore[pol] / noise_prediction), axis=0)
wgts = np.where(np.isfinite(timeavg_SNR), 1, 0)
model = np.zeros_like(timeavg_SNR)
for band in [low_band, high_band]:
model[band], _, _ = dspec.fourier_filter(freqs[band], np.where(np.isfinite(timeavg_SNR[band]), timeavg_SNR[band], 0),
wgts[band], [0], [AVG_SPECTRUM_FILTER_DELAY / 1e9], mode="dpss_solve",
eigenval_cutoff=[EIGENVAL_CUTOFF], suppression_factors=[EIGENVAL_CUTOFF])
filt_SNR[pol] = timeavg_SNR - model
# correct for impact of filter
correction_factors = np.ones_like(wgts) * np.nan
for band in [low_band, high_band]:
X = dspec.dpss_operator(freqs[band], [0], filter_half_widths=[AVG_SPECTRUM_FILTER_DELAY / 1e9], eigenval_cutoff=[EIGENVAL_CUTOFF])[0]
W = wgts[band]
leverage = np.diag(X @ np.linalg.pinv(np.dot(X.T * W, X)) @ (X.T * W))
correction_factors[band] = np.where(leverage > 0, (1 - leverage)**.5, np.nan) # because the underlying data should be gaussian
filt_SNR[pol] /= correction_factors
return filt_SNR
def iteratively_flag_on_delay_filtered_time_avg_zscore(flags, thresh=TIME_AVG_DELAY_FILT_SNR_THRESH, dynamic_range=TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE,
filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF):
"""Flag whole channels based on their outlierness after delay-filterd time-averaged zscores.
This is done iteratively since the delay filter can be unduly influenced by large outliers."""
filt_SNR = time_avg_zscore_dly_filt_SNRs(flags, filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF)
while True:
largest_SNR = np.nanmax(list(filt_SNR.values()))
if largest_SNR < thresh:
break
#
cut = np.max([thresh, largest_SNR / dynamic_range])
for pol in filt_SNR:
flags[:, filt_SNR[pol] > cut] = True
filt_SNR = time_avg_zscore_dly_filt_SNRs(flags, filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF)
In [15]:
flags = np.any(~np.isfinite(list(zscore.values())), axis=0)
print(f'{np.mean(flags):.3%} of waterfall flagged to start.')
# flag whole integrations or channels using outliers in median
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
iteratively_flag_on_averaged_zscore(flags, zscore[pol], avg_func=np.nanmedian, avg_z_thresh=AVG_Z_THRESH, verbose=True)
impose_max_chan_flag_frac(flags, max_flag_frac=MAX_FREQ_FLAG_FRAC, verbose=True)
impose_max_time_flag_frac(flags, max_flag_frac=MAX_TIME_FLAG_FRAC, verbose=True)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after flagging whole times and channels with median z > {AVG_Z_THRESH}.')
# flag largest outliers
for pol in ['ee', 'nn']:
flags |= (zscore[pol] > Z_THRESH)
print(f'{np.mean(flags):.3%} of waterfall flagged after flagging z > {Z_THRESH} outliers.')
# watershed flagging
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
flags |= xrfi._ws_flag_waterfall(zscore[pol], flags, WS_Z_THRESH)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after watershed flagging on z > {WS_Z_THRESH} neighbors of prior flags.')
# flag whole integrations or channels using outliers in mean
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
iteratively_flag_on_averaged_zscore(flags, zscore[pol], avg_func=np.nanmean, avg_z_thresh=AVG_Z_THRESH, verbose=True)
impose_max_chan_flag_frac(flags, max_flag_frac=MAX_FREQ_FLAG_FRAC, verbose=True)
impose_max_time_flag_frac(flags, max_flag_frac=MAX_TIME_FLAG_FRAC, verbose=True)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after flagging whole times and channels with average z > {AVG_Z_THRESH}.')
# flag channels based on delay filter
iteratively_flag_on_delay_filtered_time_avg_zscore(flags, thresh=TIME_AVG_DELAY_FILT_SNR_THRESH, dynamic_range=TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE,
filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF)
print(f'{np.mean(flags):.3%} of flagging channels that are {TIME_AVG_DELAY_FILT_SNR_THRESH}σ outliers after delay filtering the time average.')
# watershed flagging again
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
flags |= xrfi._ws_flag_waterfall(zscore[pol], flags, WS_Z_THRESH)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after another round of watershed flagging on z > {WS_Z_THRESH} neighbors of prior flags.')
22.941% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 47 integrations and 31 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 1 times previously flagged 10.00% or more.
Flagging an additional 36 integrations and 80 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 2 times previously flagged 10.00% or more.
Flagging an additional 19 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 1 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 1 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. 32.158% of waterfall flagged after flagging whole times and channels with median z > 1.0. 35.059% of waterfall flagged after flagging z > 4.0 outliers.
38.735% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 166 channels previously flagged 25.00% or more. Flagging 484 times previously flagged 10.00% or more.
Mean of empty slice Mean of empty slice
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. 53.302% of waterfall flagged after flagging whole times and channels with average z > 1.0.
Mean of empty slice Casting complex values to real discards the imaginary part Casting complex values to real discards the imaginary part
64.830% of flagging channels that are 4.0σ outliers after delay filtering the time average.
65.709% of waterfall flagged after another round of watershed flagging on z > 2.0 neighbors of prior flags.
Show results of flagging¶
Figure 3: Waterfall of Maximum z-Score of Either Polarization After Round 2 Flagging¶
The same as Figure 1, but after the flagging performed in this notebook.
In [16]:
plot_max_z_score(zscore, flags=flags)
All-NaN axis encountered
In [17]:
def zscore_spectra(ylim=[-3, 3], flags=flags):
fig, axes = plt.subplots(2, 1, figsize=(14,6), dpi=100, sharex=True, sharey=True, gridspec_kw={'hspace': 0})
for ax, pol in zip(axes, ['ee', 'nn']):
ax.plot(freqs / 1e6, np.nanmean(zscore[pol], axis=0),'r', label=f'{pol}-Polarization Before Round 2 Flagging', lw=.5)
ax.plot(freqs / 1e6, np.nanmean(np.where(flags, np.nan, zscore[pol]), axis=0), label=f'{pol}-Polarization After Round 2 Flagging')
ax.legend(loc='lower right')
ax.set_ylabel('Time-Averged Z-Score\n(Excluding Flags)')
ax.set_ylim(ylim)
axes[1].set_xlabel('Frequency (MHz)')
plt.tight_layout()
Figure 4: Spectra of Time-Averaged z-Scores¶
The average along the time axis of Figures 1 and 3 (though now separated per-polarization). This plot is useful for showing channels with repeated low-level RFI.
In [18]:
zscore_spectra()
Mean of empty slice Mean of empty slice
In [19]:
def summarize_flagging(flags=flags):
plt.figure(figsize=(14,10), dpi=100)
cmap = matplotlib.colors.ListedColormap(((0, 0, 0),) + matplotlib.cm.get_cmap("Set2").colors[0:2])
plt.imshow(np.where(np.any(~np.isfinite(list(zscore.values())), axis=0), 1, np.where(flags, 2, 0)),
aspect='auto', cmap=cmap, interpolation='none', extent=extent)
plt.clim([-.5, 2.5])
cbar = plt.colorbar(location='top', aspect=40, pad=.02)
cbar.set_ticks([0, 1, 2])
cbar.set_ticklabels(['Unflagged', 'Previously Flagged', 'Flagged Here Using Delayed Filtered z-Scores'])
plt.xlabel('Frequency (MHz)')
plt.ylabel(f'JD - {int(times[0])}')
plt.tight_layout()
Figure 5: Summary of Flags Before and After Round 2 Flagging¶
This plot shows which times and frequencies were flagged before and after this notebook. It is directly comparable to Figure 5 of the first round full_day_rfi notebook.
In [20]:
summarize_flagging()
The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
Save results¶
In [21]:
add_to_history = 'by full_day_rfi_round_2 notebook with the following environment:\n' + '=' * 65 + '\n' + os.popen('conda env export').read() + '=' * 65
In [22]:
tind = 0
always_flagged_ants = set()
ever_unflagged_ants = set()
for cal_file in cal_files:
with warnings.catch_warnings():
warnings.simplefilter("ignore")
# update cal_file
uvc = UVCal()
uvc.read(cal_file, use_future_array_shapes=True)
uvc.flag_array |= (flags[tind:tind + len(uvc.time_array), :].T)[None, :, :, None]
uvc.history += 'Modified ' + add_to_history
uvc.write_calfits(cal_file, clobber=True)
# keep track of flagged antennas
for antnum in uvc.ant_array:
for antpol in ['Jee', 'Jnn']:
if np.all(uvc.get_flags(antnum, antpol)):
if (antnum, antpol) not in ever_unflagged_ants:
always_flagged_ants.add((antnum, antpol))
else:
ever_unflagged_ants.add((antnum, antpol))
always_flagged_ants.discard((antnum, antpol))
# Create new flag object
uvf_out = UVFlag(uvc, waterfall=True, mode='flag')
uvf_out.flag_array |= flags[tind:tind + len(uvc.time_array), :, None]
uvf_out.history += 'Produced ' + add_to_history
uvf_out.write(cal_file.replace(SMOOTH_CAL_SUFFIX, FLAG_WATERFALL2_SUFFIX), clobber=True)
# increment time index
tind += len(uvc.time_array)
print(f'Saved {len(cal_files)} *.{FLAG_WATERFALL2_SUFFIX} files starting with {cal_files[0].replace(SMOOTH_CAL_SUFFIX, FLAG_WATERFALL2_SUFFIX)}.')
Saved 1571 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data2/2460792/zen.2460792.21084.sum.flag_waterfall_round_2.h5.
In [23]:
# write summary of entirely flagged times/freqs/ants to yaml
all_flagged_times = np.all(flags, axis=1)
all_flagged_freqs = np.all(flags, axis=0)
all_flagged_ants = [(int(ant[0]), ant[1]) for ant in sorted(always_flagged_ants)]
dt = np.median(np.diff(times))
out_yml_str = 'JD_flags: ' + str([[float(times[flag_stretch][0] - dt / 2), float(times[flag_stretch][-1] + dt / 2)]
for flag_stretch in true_stretches(all_flagged_times)])
df = np.median(np.diff(freqs))
out_yml_str += '\n\nfreq_flags: ' + str([[float(freqs[flag_stretch][0] - df / 2), float(freqs[flag_stretch][-1] + df / 2)]
for flag_stretch in true_stretches(all_flagged_freqs)])
out_yml_str += '\n\nex_ants: ' + str(all_flagged_ants).replace("'", "").replace('(', '[').replace(')', ']')
print(f'Writing the following to {out_yaml_file}\n' + '-' * (25 + len(out_yaml_file)))
print(out_yml_str)
with open(out_yaml_file, 'w') as outfile:
outfile.writelines(out_yml_str)
Writing the following to /mnt/sn1/data2/2460792/2460792_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460792.210726412, 2460792.2200098047], [2460792.220345349, 2460792.223029704], [2460792.223141552, 2460792.223365248], [2460792.2238126406, 2460792.224036337], [2460792.224371881, 2460792.2247074256], [2460792.2252666657, 2460792.225378514], [2460792.2259377544, 2460792.2260496025], [2460792.2262732987, 2460792.228398413], [2460792.2287339573, 2460792.2288458054], [2460792.229293198, 2460792.2297405903], [2460792.2298524384, 2460792.230187983], [2460792.231865704, 2460792.2324249446], [2460792.2332078815, 2460792.2343263626], [2460792.2344382107, 2460792.234661907], [2460792.236339628, 2460792.237122565], [2460792.2394713755, 2460792.2396950717], [2460792.24450454, 2460792.245063781], [2460792.2467415025, 2460792.247636287], [2460792.2480836795, 2460792.248419224], [2460792.2502087937, 2460792.2509917305], [2460792.2542353254, 2460792.2545708697], [2460792.2564722877, 2460792.2570315283], [2460792.2722428706, 2460792.272690263], [2460792.275710162, 2460792.275933858], [2460792.276716795, 2460792.277723428], [2460792.2791774534, 2460792.2792893015], [2460792.280855175, 2460792.2813025676], [2460792.282980289, 2460792.284210618], [2460792.284881707, 2460792.2856646436], [2460792.286783125, 2460792.2871186687], [2460792.2877897574, 2460792.2879016055], [2460792.2905859603, 2460792.2910333527], [2460792.292487378, 2460792.29293477], [2460792.293494011, 2460792.293829555], [2460792.2960665175, 2460792.296737606], [2460792.3074750244, 2460792.308034265], [2460792.309040898, 2460792.3093764423], [2460792.3118371004, 2460792.312508189], [2460792.3138503665, 2460792.3140740627], [2460792.3159754802, 2460792.3163110246], [2460792.3172058095, 2460792.317653202], [2460792.3179887463, 2460792.3182124426], [2460792.3192190756, 2460792.3193309237], [2460792.319442772, 2460792.3198901643], [2460792.3241403922, 2460792.3244759366], [2460792.3279432277, 2460792.32839062], [2460792.335548899, 2460792.3358844435], [2460792.340470216, 2460792.3412531526], [2460792.3413650007, 2460792.341812393], [2460792.3429308743, 2460792.3430427224], [2460792.3447204437, 2460792.344832292], [2460792.3453915324, 2460792.3457270768], [2460792.346957406, 2460792.347069254], [2460792.347852191, 2460792.348075887], [2460792.35143133, 2460792.3518787227], [2460792.352214267, 2460792.3525498114], [2460792.354563077, 2460792.355681558], [2460792.3559052544, 2460792.3561289506], [2460792.358589609, 2460792.3588133054], [2460792.3590370016, 2460792.359372546], [2460792.359596242, 2460792.3599317865], [2460792.3612739635, 2460792.361609508], [2460792.3632872296, 2460792.363622774], [2460792.3686559387, 2460792.368991483], [2460792.3757023695, 2460792.3758142176], [2460792.376149762, 2460792.3765971544], [2460792.382860648, 2460792.3834198886], [2460792.3843146735, 2460792.3844265216], [2460792.3853213065, 2460792.385656851], [2460792.3864397877, 2460792.3869990283], [2460792.3885649014, 2460792.389124142], [2460792.3905781675, 2460792.3908018637], [2460792.390913712, 2460792.39102556], [2460792.391920345, 2460792.3924795855], [2460792.3925914336, 2460792.393150674], [2460792.393821763, 2460792.3947165473], [2460792.3986312314, 2460792.399078624], [2460792.40165113, 2460792.401762978], [2460792.4101515864, 2460792.4102634345], [2460792.4116056114, 2460792.411941156], [2460792.412947789, 2460792.413059637], [2460792.4133951813, 2460792.414178118], [2460792.415184751, 2460792.4154084474], [2460792.4157439917, 2460792.41585584], [2460792.4167506243, 2460792.4169743205], [2460792.417198017, 2460792.4208890046], [2460792.4210008527, 2460792.426593258], [2460792.4268169543, 2460792.4272643467], [2460792.4275998906, 2460792.428159131], [2460792.4283828274, 2460792.4286065237], [2460792.429053916, 2460792.429836853], [2460792.4301723973, 2460792.4311790303], [2460792.4325212077, 2460792.432633056], [2460792.4330804483, 2460792.433304144], [2460792.4348700177, 2460792.435205562], [2460792.4365477394, 2460792.4366595875], [2460792.4373306762, 2460792.4377780687], [2460792.438001765, 2460792.4383373093], [2460792.4386728536, 2460792.4392320937], [2460792.439343942, 2460792.439679486], [2460792.4410216636, 2460792.4411335117], [2460792.4416927523, 2460792.4418046004], [2460792.443258626, 2460792.443370474], [2460792.444712651, 2460792.4453837397], [2460792.445607436, 2460792.445719284], [2460792.4462785246, 2460792.446502221], [2460792.448403639, 2460792.448515487], [2460792.4488510313, 2460792.4490747275], [2460792.4500813605, 2460792.4501932086], [2460792.4532131073, 2460792.4533249554], [2460792.4553382215, 2460792.4554500696], [2460792.4565685503, 2460792.4567922466], [2460792.457015943, 2460792.457127791], [2460792.4579107277, 2460792.458134424], [2460792.45835812, 2460792.4584699683], [2460792.4594766013, 2460792.4597002976], [2460792.46014769, 2460792.4603713863], [2460792.460930627, 2460792.461042475], [2460792.461378019, 2460792.461601715], [2460792.461713563, 2460792.4619372594], [2460792.462384652, 2460792.462608348], [2460792.462720196, 2460792.4629438925], [2460792.4631675887, 2460792.463279437], [2460792.4639505255, 2460792.4640623736], [2460792.466411184, 2460792.466523032], [2460792.4674178166, 2460792.4675296647], [2460792.4838594887, 2460792.4840831845], [2460792.486655691, 2460792.4869912355], [2460792.490905919, 2460792.4911296153], [2460792.4921362484, 2460792.4923599446], [2460792.5048869327, 2460792.505110629], [2460792.5133873885, 2460792.5134992367], [2460792.521887845, 2460792.522111541], [2460792.5226707817, 2460792.523006326], [2460792.5258025285, 2460792.5260262247], [2460792.527144706, 2460792.527368402], [2460792.529717212, 2460792.5299409083], [2460792.530611997, 2460792.5308356932], [2460792.531506782, 2460792.531730478], [2460792.5330726556, 2460792.533296352], [2460792.533631896, 2460792.5337437443], [2460792.534414833, 2460792.534526681], [2460792.5356451618, 2460792.5362044023], [2460792.536651795, 2460792.5370991873], [2460792.5373228835, 2460792.537993972], [2460792.5382176684, 2460792.5383295165], [2460792.538888757, 2460792.5393361496], [2460792.540230934, 2460792.541349415], [2460792.5416849595, 2460792.5417968077], [2460792.541908656, 2460792.542020504], [2460792.543027137, 2460792.5435863775], [2460792.544816707, 2460792.545152251], [2460792.5455996436, 2460792.5457114917], [2460792.5461588837, 2460792.546270732], [2460792.5468299724, 2460792.5470536686], [2460792.5471655168, 2460792.547501061], [2460792.547612909, 2460792.5478366055], [2460792.548395846, 2460792.548507694], [2460792.5486195423, 2460792.5509683527], [2460792.551192049, 2460792.551751289], [2460792.551863137, 2460792.5519749853], [2460792.5520868334, 2460792.5523105296], [2460792.55286977, 2460792.553429011], [2460792.553652707, 2460792.5538764033], [2460792.5548830363, 2460792.5552185806], [2460792.5553304288, 2460792.555442277], [2460792.555554125, 2460792.557902935], [2460792.558014783, 2460792.5585740237], [2460792.558685872, 2460792.5592451124], [2460792.5594688086, 2460792.5601398973], [2460792.5603635935, 2460792.5621531634]] freq_flags: [[46859741.2109375, 70541381.8359375], [70663452.1484375, 71029663.0859375], [71273803.7109375, 71762084.9609375], [71884155.2734375, 72250366.2109375], [72494506.8359375, 72860717.7734375], [72982788.0859375, 73348999.0234375], [73593139.6484375, 75790405.2734375], [75912475.5859375, 76278686.5234375], [76522827.1484375, 76889038.0859375], [77011108.3984375, 77255249.0234375], [77377319.3359375, 79208374.0234375], [79452514.6484375, 79696655.2734375], [79940795.8984375, 80184936.5234375], [80551147.4609375, 80795288.0859375], [81039428.7109375, 81405639.6484375], [81527709.9609375, 82504272.4609375], [82626342.7734375, 83114624.0234375], [83236694.3359375, 83602905.2734375], [83724975.5859375, 84091186.5234375], [84335327.1484375, 84579467.7734375], [84823608.3984375, 85067749.0234375], [85189819.3359375, 86410522.4609375], [86532592.7734375, 108261108.3984375], [109725952.1484375, 109848022.4609375], [109970092.7734375, 110092163.0859375], [110214233.3984375, 110702514.6484375], [111434936.5234375, 112899780.2734375], [113265991.2109375, 113510131.8359375], [113632202.1484375, 113754272.4609375], [116073608.3984375, 116195678.7109375], [116439819.3359375, 116561889.6484375], [116683959.9609375, 116806030.2734375], [119979858.3984375, 120101928.7109375], [123764038.0859375, 123886108.3984375], [124862670.8984375, 129867553.7109375], [129989624.0234375, 130111694.3359375], [130599975.5859375, 131820678.7109375], [136337280.2734375, 136459350.5859375], [136703491.2109375, 136825561.5234375], [136947631.8359375, 138168334.9609375], [138290405.2734375, 138412475.5859375], [138656616.2109375, 138778686.5234375], [139022827.1484375, 139144897.4609375], [139511108.3984375, 139633178.7109375], [140853881.8359375, 141342163.0859375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 144027709.9609375], [144638061.5234375, 144760131.8359375], [145492553.7109375, 145614624.0234375], [146224975.5859375, 146347045.8984375], [146591186.5234375, 147201538.0859375], [147445678.7109375, 147567749.0234375], [147933959.9609375, 148056030.2734375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [152938842.7734375, 153060913.0859375], [153671264.6484375, 153915405.2734375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [156845092.7734375, 156967163.0859375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158309936.5234375], [159164428.7109375, 159286499.0234375], [160140991.2109375, 160507202.1484375], [161361694.3359375, 161483764.6484375], [161727905.2734375, 161849975.5859375], [163070678.7109375, 165756225.5859375], [165878295.8984375, 169418334.9609375], [169906616.2109375, 170761108.3984375], [170883178.7109375, 171005249.0234375], [171249389.6484375, 171615600.5859375], [171737670.8984375, 171859741.2109375], [174911499.0234375, 175033569.3359375], [175155639.6484375, 175399780.2734375], [175643920.8984375, 175765991.2109375], [176010131.8359375, 176132202.1484375], [181137084.9609375, 181381225.5859375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [188583374.0234375, 188705444.3359375], [189926147.4609375, 190048217.7734375], [190780639.6484375, 190902709.9609375], [191024780.2734375, 191513061.5234375], [193222045.8984375, 193344116.2109375], [195419311.5234375, 195541381.8359375], [195663452.1484375, 195785522.4609375], [195907592.7734375, 196029663.0859375], [197128295.8984375, 197372436.5234375], [197860717.7734375, 202621459.9609375], [203598022.4609375, 203964233.3984375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [207138061.5234375, 207382202.1484375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [210433959.9609375, 210556030.2734375], [212142944.3359375, 212265014.6484375], [213485717.7734375, 213607788.0859375], [215072631.8359375, 215316772.4609375], [215927124.0234375, 216049194.3359375], [216659545.8984375, 216781616.2109375], [219589233.3984375, 219711303.7109375], [219833374.0234375, 219955444.3359375], [220565795.8984375, 220809936.5234375], [221176147.4609375, 221298217.7734375], [222030639.6484375, 222274780.2734375], [222518920.8984375, 222640991.2109375], [222885131.8359375, 223739624.0234375], [227401733.3984375, 227767944.3359375], [229110717.7734375, 229354858.3984375], [229721069.3359375, 229843139.6484375], [229965209.9609375, 230087280.2734375], [230209350.5859375, 230575561.5234375], [231063842.7734375, 231185913.0859375], [232894897.4609375, 233016967.7734375], [233139038.0859375, 233383178.7109375], [233627319.3359375, 233749389.6484375], [233993530.2734375, 234237670.8984375]] ex_ants: [[4, Jee], [7, Jee], [8, Jee], [8, Jnn], [10, Jee], [15, Jee], [15, Jnn], [18, Jee], [18, Jnn], [20, Jee], [20, Jnn], [21, Jee], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [30, Jee], [30, Jnn], [32, Jnn], [33, Jnn], [34, Jee], [35, Jnn], [37, Jnn], [40, Jnn], [42, Jnn], [46, Jee], [47, Jnn], [49, Jnn], [51, Jee], [56, Jee], [56, Jnn], [60, Jnn], [62, Jee], [66, Jee], [66, Jnn], [67, Jnn], [68, Jee], [68, Jnn], [70, Jee], [70, Jnn], [71, Jee], [71, Jnn], [72, Jee], [72, Jnn], [76, Jee], [76, Jnn], [77, Jnn], [78, Jee], [80, Jee], [80, Jnn], [81, Jnn], [82, Jee], [82, Jnn], [83, Jee], [83, Jnn], [86, Jee], [86, Jnn], [87, Jee], [92, Jee], [96, Jee], [96, Jnn], [99, Jnn], [102, Jnn], [104, Jnn], [105, Jee], [107, Jee], [107, Jnn], [108, Jnn], [109, Jnn], [113, Jnn], [114, Jee], [114, Jnn], [115, Jee], [116, Jnn], [117, Jee], [119, Jee], [119, Jnn], [120, Jee], [120, Jnn], [121, Jee], [125, Jnn], [127, Jee], [127, Jnn], [130, Jee], [130, Jnn], [134, Jee], [135, Jee], [136, Jnn], [137, Jee], [137, Jnn], [143, Jee], [143, Jnn], [148, Jee], [153, Jnn], [155, Jnn], [161, Jnn], [166, Jee], [166, Jnn], [167, Jnn], [170, Jee], [172, Jnn], [173, Jnn], [174, Jnn], [175, Jnn], [176, Jnn], [180, Jee], [180, Jnn], [182, Jee], [184, Jee], [184, Jnn], [185, Jee], [185, Jnn], [186, Jee], [186, Jnn], [188, Jnn], [194, Jnn], [197, Jnn], [199, Jee], [199, Jnn], [200, Jee], [200, Jnn], [202, Jnn], [204, Jnn], [206, Jnn], [212, Jnn], [214, Jee], [214, Jnn], [215, Jnn], [218, Jnn], [227, Jee], [227, Jnn], [231, Jee], [231, Jnn], [233, Jnn], [235, Jee], [235, Jnn], [236, Jee], [236, Jnn], [238, Jnn], [239, Jee], [244, Jee], [250, Jee], [251, Jee], [251, Jnn], [252, Jnn], [253, Jee], [253, Jnn], [254, Jee], [254, Jnn], [255, Jee], [255, Jnn], [257, Jee], [257, Jnn], [262, Jee], [262, Jnn], [266, Jee], [266, Jnn], [267, Jee], [267, Jnn], [268, Jee], [268, Jnn], [269, Jee], [269, Jnn], [271, Jee], [271, Jnn], [273, Jee], [273, Jnn], [281, Jnn], [282, Jee], [282, Jnn], [283, Jee], [283, Jnn], [284, Jee], [284, Jnn], [286, Jee], [286, Jnn], [295, Jee], [295, Jnn], [320, Jee], [320, Jnn], [321, Jee], [321, Jnn], [322, Jee], [322, Jnn], [323, Jee], [323, Jnn], [324, Jee], [324, Jnn], [325, Jee], [325, Jnn], [326, Jee], [326, Jnn], [327, Jee], [327, Jnn], [328, Jee], [328, Jnn], [329, Jee], [329, Jnn], [331, Jee], [331, Jnn], [332, Jee], [332, Jnn], [333, Jee], [333, Jnn], [336, Jee], [336, Jnn], [340, Jee], [340, Jnn]]
Metadata¶
In [24]:
for repo in ['hera_cal', 'hera_qm', 'hera_filters', 'hera_notebook_templates', 'pyuvdata']:
exec(f'from {repo} import __version__')
print(f'{repo}: {__version__}')
hera_cal: 3.7.1.dev45+g4a0c6f1 hera_qm: 2.2.1.dev2+ga535e9e hera_filters: 0.1.6.dev9+gf165ec1
hera_notebook_templates: 0.1.dev989+gee0995d pyuvdata: 3.1.3
In [25]:
print(f'Finished execution in {(time.time() - tstart) / 60:.2f} minutes.')
Finished execution in 24.90 minutes.