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 1841 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data1/2460623/zen.2460623.25256.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 1841 *.sum.smooth.calfits files starting with /mnt/sn1/data1/2460623/zen.2460623.25256.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.')
21.940% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 3 integrations and 30 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 57 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 2 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 1 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. 27.162% of waterfall flagged after flagging whole times and channels with median z > 1.0. 29.035% of waterfall flagged after flagging z > 4.0 outliers.
31.166% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 61 channels previously flagged 25.00% or more.
Mean of empty slice Mean of empty slice
Flagging 383 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.
40.419% 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
47.877% of flagging channels that are 4.0σ outliers after delay filtering the time average.
48.540% 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 two minor releases later. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap(obj)`` 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 1841 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data1/2460623/zen.2460623.25256.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 = sorted(always_flagged_ants)
dt = np.median(np.diff(times))
out_yml_str = 'JD_flags: ' + str([[times[flag_stretch][0] - dt / 2, 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([[freqs[flag_stretch][0] - df / 2, 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/data1/2460623/2460623_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460623.252673676, 2460623.2531210682], [2460623.2534566126, 2460623.2535684607], [2460623.2541277013, 2460623.2542395494], [2460623.254686942, 2460623.2584897773], [2460623.2586016254, 2460623.260055651], [2460623.260167499, 2460623.260279347], [2460623.2603911953, 2460623.2605030434], [2460623.2607267397, 2460623.260950436], [2460623.261062284, 2460623.2617333727], [2460623.261845221, 2460623.2624044614], [2460623.2625163095, 2460623.2626281576], [2460623.2628518534, 2460623.263299246], [2460623.263411094, 2460623.264417727], [2460623.264529575, 2460623.265200664], [2460623.265312512, 2460623.26542436], [2460623.2659836006, 2460623.26721393], [2460623.267325778, 2460623.267437626], [2460623.2676613224, 2460623.272358943], [2460623.272694487, 2460623.2728063352], [2460623.273477424, 2460623.273589272], [2460623.274036664, 2460623.2742603603], [2460623.2744840565, 2460623.275938082], [2460623.27604993, 2460623.2762736264], [2460623.2763854745, 2460623.2766091707], [2460623.276944715, 2460623.277727652], [2460623.2778395, 2460623.278063196], [2460623.2781750443, 2460623.2782868925], [2460623.278734285, 2460623.2790698293], [2460623.2791816774, 2460623.2795172213], [2460623.2799646137, 2460623.280076462], [2460623.280412006, 2460623.2805238543], [2460623.2807475505, 2460623.2808593987], [2460623.281194943, 2460623.2815304874], [2460623.2818660317, 2460623.282313424], [2460623.283991146, 2460623.2843266902], [2460623.2844385384, 2460623.2845503865], [2460623.284885931, 2460623.285109627], [2460623.286339956, 2460623.2868991964], [2460623.287570285, 2460623.2879058295], [2460623.288353222, 2460623.2886887663], [2460623.289359855, 2460623.2898072475], [2460623.2922679055, 2460623.292715298], [2460623.2930508424, 2460623.2933863867], [2460623.2939456273, 2460623.294728564], [2460623.2965181335, 2460623.2967418297], [2460623.30322902, 2460623.303340868], [2460623.3099399065, 2460623.3100517546], [2460623.3110583876, 2460623.311282084], [2460623.3127361094, 2460623.3129598056], [2460623.313630894, 2460623.313742742], [2460623.3170981854, 2460623.3172100335], [2460623.3183285147, 2460623.318440363], [2460623.319894388, 2460623.320006236], [2460623.320901021, 2460623.321012869], [2460623.3212365652, 2460623.3216839577], [2460623.3233616794, 2460623.3234735276], [2460623.32392092, 2460623.3242564644], [2460623.3243683125, 2460623.3249275526], [2460623.32626973, 2460623.326493426], [2460623.3280593, 2460623.3285066923], [2460623.331638439, 2460623.331750287], [2460623.3332043127, 2460623.333428009], [2460623.3339872495, 2460623.334322794], [2460623.334658338, 2460623.3348820345], [2460623.3349938826, 2460623.3351057307], [2460623.3363360595, 2460623.336671604], [2460623.336783452, 2460623.3371189963], [2460623.3401388954, 2460623.3402507436], [2460623.342040313, 2460623.342152161], [2460623.343270642, 2460623.3433824903], [2460623.3434943384, 2460623.343829883], [2460623.3469616296, 2460623.3470734777], [2460623.347409022, 2460623.3478564145], [2460623.3479682626, 2460623.3480801107], [2460623.348303807, 2460623.348415655], [2460623.353784364, 2460623.3542317566], [2460623.3549028453, 2460623.355350238], [2460623.358817529, 2460623.358929377], [2460623.3594886176, 2460623.3596004657], [2460623.3604952507, 2460623.360607099], [2460623.3622848205, 2460623.3625085168], [2460623.362732213, 2460623.362844061], [2460623.362955909, 2460623.3630677573], [2460623.3634033017, 2460623.36351515], [2460623.364186238, 2460623.364298086], [2460623.3645217824, 2460623.3648573267], [2460623.3657521117, 2460623.36586396], [2460623.3675416815, 2460623.3677653777], [2460623.368995707, 2460623.369107555], [2460623.3696667952, 2460623.3697786434], [2460623.373245935, 2460623.3738051755], [2460623.375706593, 2460623.375818441], [2460623.3760421374, 2460623.3761539855], [2460623.37648953, 2460623.376601378], [2460623.3769369223, 2460623.3770487704], [2460623.379733125, 2460623.3798449733], [2460623.3877861886, 2460623.3878980367], [2460623.3937141383, 2460623.3938259864], [2460623.405681886, 2460623.405793734], [2460623.4109387468, 2460623.411050595], [2460623.4148534304, 2460623.4150771266], [2460623.4151889747, 2460623.415300823], [2460623.416307456, 2460623.416531152], [2460623.419439203, 2460623.419551051], [2460623.421564317, 2460623.421676165], [2460623.426933026, 2460623.427044874], [2460623.4318543426, 2460623.4319661907], [2460623.4336439124, 2460623.4337557605], [2460623.4345386974, 2460623.4346505455], [2460623.440242951, 2460623.440354799], [2460623.4409140395, 2460623.4410258876], [2460623.4416969763, 2460623.4418088244], [2460623.443933938, 2460623.444716875], [2460623.4472893816, 2460623.4474012298], [2460623.454335812, 2460623.45444766], [2460623.457243863, 2460623.457355711], [2460623.460487458, 2460623.4608230023], [2460623.464402142, 2460623.4649613826], [2460623.4684286737, 2460623.468540522], [2460623.4690997624, 2460623.4692116105], [2460623.469547155, 2460623.469659003], [2460623.469770851, 2460623.4701063954], [2460623.470553788, 2460623.470665636], [2460623.474804016, 2460623.474915864], [2460623.478271307, 2460623.478383155], [2460623.4809556617, 2460623.481179358], [2460623.490574599, 2460623.490910143], [2460623.491916776, 2460623.4920286243], [2460623.4967262447, 2460623.496949941], [2460623.503101587, 2460623.503213435], [2460623.5038845236, 2460623.504891156], [2460623.5063451817, 2460623.506568878], [2460623.507687359, 2460623.5077992072], [2460623.5079110553, 2460623.5082465997], [2460623.5099243214, 2460623.5100361696], [2460623.512049435, 2460623.5121612833], [2460623.516970752, 2460623.517194448], [2460623.5176418405, 2460623.5177536886], [2460623.5207735877, 2460623.520885436], [2460623.5220039166, 2460623.5221157647], [2460623.524240879, 2460623.524352727], [2460623.5280437144, 2460623.5282674106], [2460623.5292740436, 2460623.5293858917], [2460623.5459394115, 2460623.5460512596], [2460623.549630399, 2460623.549854095], [2460623.5509725763, 2460623.5511962725], [2460623.5699867546, 2460623.5700986027], [2460623.5711052357, 2460623.571217084], [2460623.5717763244, 2460623.5718881725], [2460623.574572527, 2460623.574684375], [2460623.576809489, 2460623.5769213373], [2460623.579158299, 2460623.5796056916], [2460623.5837440714, 2460623.5838559195], [2460623.5839677677, 2460623.584079616], [2460623.584303312, 2460623.5859810337], [2460623.586092882, 2460623.58620473], [2460623.5865402743, 2460623.5867639706], [2460623.5886653885, 2460623.5887772366], [2460623.5901194136, 2460623.5902312617], [2460623.5907905023, 2460623.5909023504], [2460623.5936985533, 2460623.5938104014], [2460623.5941459457, 2460623.594257794], [2460623.594928882, 2460623.59504073], [2460623.5989554143, 2460623.5992909586], [2460623.6008568318, 2460623.601192376], [2460623.601304224, 2460623.6017516167], [2460623.6108113136, 2460623.6114824023], [2460623.611706098, 2460623.612377187], [2460623.612712731, 2460623.6150615416], [2460623.615285238, 2460623.6158444784], [2460623.619759162, 2460623.619982858], [2460623.6204302507, 2460623.620542099], [2460623.620653947, 2460623.620877643], [2460623.620989491, 2460623.6223316686], [2460623.6224435167, 2460623.622555365], [2460623.6227790606, 2460623.6228909087], [2460623.623114605, 2460623.623226453], [2460623.6234501493, 2460623.6235619974], [2460623.625016023, 2460623.6255752635], [2460623.6389970365, 2460623.6391088846], [2460623.639444429, 2460623.639668125], [2460623.6412339984, 2460623.6414576946], [2460623.6422406314, 2460623.642688024], [2460623.6443657456, 2460623.644589442], [2460623.649287062, 2460623.6653931895]] freq_flags: [[47714233.3984375, 47958374.0234375], [49911499.0234375, 50155639.6484375], [62362670.8984375, 62728881.8359375], [69931030.2734375, 70053100.5859375], [72250366.2109375, 72372436.5234375], [73471069.3359375, 73593139.6484375], [87387084.9609375, 108383178.7109375], [108749389.6484375, 108871459.9609375], [109603881.8359375, 109725952.1484375], [109970092.7734375, 110092163.0859375], [110580444.3359375, 110702514.6484375], [111068725.5859375, 111312866.2109375], [111557006.8359375, 111923217.7734375], [112045288.0859375, 114242553.7109375], [114364624.0234375, 114730834.9609375], [114852905.2734375, 115463256.8359375], [115585327.1484375, 116683959.9609375], [116806030.2734375, 117294311.5234375], [117416381.8359375, 117782592.7734375], [118026733.3984375, 118392944.3359375], [118759155.2734375, 119003295.8984375], [119247436.5234375, 119491577.1484375], [119979858.3984375, 120101928.7109375], [120468139.6484375, 120834350.5859375], [121932983.3984375, 122055053.7109375], [122909545.8984375, 123275756.8359375], [124130249.0234375, 124496459.9609375], [124740600.5859375, 125350952.1484375], [126205444.3359375, 126449584.9609375], [126815795.8984375, 126937866.2109375], [127059936.5234375, 127182006.8359375], [127304077.1484375, 127426147.4609375], [127548217.7734375, 127670288.0859375], [127792358.3984375, 128158569.3359375], [128524780.2734375, 128646850.5859375], [129867553.7109375, 130233764.6484375], [131454467.7734375, 131576538.0859375], [131698608.3984375, 131942749.0234375], [132431030.2734375, 132797241.2109375], [132919311.5234375, 135360717.7734375], [135482788.0859375, 138778686.5234375], [138900756.8359375, 139144897.4609375], [139389038.0859375, 139755249.0234375], [139877319.3359375, 140487670.8984375], [140609741.2109375, 141098022.4609375], [141220092.7734375, 142440795.8984375], [142562866.2109375, 143661499.0234375], [143783569.3359375, 144271850.5859375], [144638061.5234375, 144760131.8359375], [145492553.7109375, 145614624.0234375], [145858764.6484375, 145980834.9609375], [147445678.7109375, 147567749.0234375], [148422241.2109375, 148544311.5234375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [153305053.7109375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [155258178.7109375, 155380249.0234375], [157211303.7109375, 158065795.8984375], [158187866.2109375, 158309936.5234375], [158432006.8359375, 158554077.1484375], [159164428.7109375, 159286499.0234375], [160140991.2109375, 160385131.8359375], [161361694.3359375, 161483764.6484375], [163681030.2734375, 163803100.5859375], [169174194.3359375, 169296264.6484375], [169784545.8984375, 171615600.5859375], [171737670.8984375, 172103881.8359375], [174789428.7109375, 175033569.3359375], [175155639.6484375, 175765991.2109375], [179672241.2109375, 179794311.5234375], [181137084.9609375, 181381225.5859375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [189193725.5859375, 189315795.8984375], [189926147.4609375, 190048217.7734375], [191024780.2734375, 191513061.5234375], [192123413.0859375, 192245483.3984375], [192489624.0234375, 192611694.3359375], [192855834.9609375, 192977905.2734375], [193222045.8984375, 193344116.2109375], [195663452.1484375, 195785522.4609375], [197128295.8984375, 197372436.5234375], [197860717.7734375, 198715209.9609375], [199203491.2109375, 199325561.5234375], [200057983.3984375, 200180053.7109375], [201644897.4609375, 201889038.0859375], [203231811.5234375, 203353881.8359375], [204452514.6484375, 205429077.1484375], [206893920.8984375, 207015991.2109375], [207138061.5234375, 207382202.1484375], [207504272.4609375, 207870483.3984375], [208114624.0234375, 209213256.8359375], [209335327.1484375, 209701538.0859375], [209945678.7109375, 210067749.0234375], [210189819.3359375, 210311889.6484375], [211776733.3984375, 212631225.5859375], [215194702.1484375, 215316772.4609375], [219223022.4609375, 219345092.7734375], [219833374.0234375, 219955444.3359375], [220199584.9609375, 221298217.7734375], [221542358.3984375, 221908569.3359375], [222152709.9609375, 222518920.8984375], [222640991.2109375, 223861694.3359375], [223983764.6484375, 224349975.5859375], [224594116.2109375, 224838256.8359375], [227401733.3984375, 227767944.3359375], [229110717.7734375, 229476928.7109375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231307983.3984375], [233993530.2734375, 234359741.2109375]] ex_ants: [[7, Jee], [8, Jee], [9, Jee], [10, Jee], [15, Jnn], [16, Jee], [17, Jnn], [18, Jee], [18, Jnn], [21, Jee], [22, Jee], [22, Jnn], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [29, Jnn], [31, Jnn], [32, Jnn], [33, Jnn], [34, Jee], [34, Jnn], [35, Jee], [35, Jnn], [36, Jee], [37, Jnn], [40, Jnn], [42, Jnn], [45, Jee], [46, Jee], [47, Jee], [47, Jnn], [48, Jee], [48, Jnn], [49, Jee], [49, Jnn], [51, Jee], [54, Jnn], [55, Jee], [61, Jee], [61, Jnn], [62, Jee], [62, Jnn], [63, Jee], [63, Jnn], [64, Jee], [64, Jnn], [69, Jee], [72, Jnn], [73, Jee], [77, Jee], [77, Jnn], [78, Jee], [78, Jnn], [86, Jee], [86, Jnn], [87, Jee], [88, Jee], [88, Jnn], [89, Jee], [90, Jee], [90, Jnn], [92, Jee], [93, Jee], [95, Jee], [97, Jnn], [98, Jnn], [100, Jnn], [103, Jnn], [104, Jnn], [107, Jee], [107, Jnn], [109, Jnn], [117, Jee], [120, Jee], [121, Jee], [121, Jnn], [125, Jnn], [127, Jee], [130, Jee], [130, Jnn], [132, Jee], [134, Jee], [136, Jee], [136, Jnn], [137, Jee], [142, Jnn], [161, Jnn], [164, Jee], [170, Jee], [171, Jnn], [176, Jee], [176, Jnn], [177, Jee], [177, Jnn], [178, Jee], [178, Jnn], [179, Jee], [179, Jnn], [180, Jee], [180, Jnn], [182, Jee], [182, Jnn], [183, Jee], [184, Jee], [188, Jnn], [191, Jee], [193, Jee], [198, Jnn], [199, Jnn], [200, Jee], [200, Jnn], [201, Jnn], [202, Jnn], [206, Jee], [207, Jee], [207, Jnn], [208, Jee], [208, Jnn], [209, Jee], [209, Jnn], [210, Jee], [210, Jnn], [212, Jnn], [213, Jee], [213, Jnn], [215, Jee], [215, Jnn], [216, Jee], [218, Jnn], [231, Jee], [232, Jee], [234, Jnn], [235, Jnn], [241, Jee], [241, Jnn], [242, Jee], [242, Jnn], [243, Jee], [243, Jnn], [245, Jnn], [246, Jee], [250, Jee], [251, Jee], [253, Jnn], [255, Jnn], [256, Jee], [256, Jnn], [262, Jee], [262, Jnn], [268, Jnn], [270, Jee], [270, Jnn], [272, Jee], [320, Jee], [320, Jnn], [321, Jee], [321, 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: 3.6.2.dev110+g0529798 hera_qm: 2.2.0 hera_filters: 0.1.6.dev1+g297dcce
hera_notebook_templates: 0.1.dev936+gdc93cad pyuvdata: 3.0.1.dev70+g283dda3
In [25]:
print(f'Finished execution in {(time.time() - tstart) / 60:.2f} minutes.')
Finished execution in 37.71 minutes.