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/data1/2460795/zen.2460795.21090.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/data1/2460795/zen.2460795.21090.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.')
23.361% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 1134 integrations and 38 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 1 times previously flagged 10.00% or more.
Flagging an additional 243 integrations and 10 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 8 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 1 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. 61.321% of waterfall flagged after flagging whole times and channels with median z > 1.0. 62.136% of waterfall flagged after flagging z > 4.0 outliers.
65.018% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 98 channels previously flagged 25.00% or more. Flagging 495 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. 76.792% 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
83.043% of flagging channels that are 4.0σ outliers after delay filtering the time average.
83.477% 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/data1/2460795/zen.2460795.21090.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/data1/2460795/2460795_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460795.210783646, 2460795.265589218], [2460795.26961575, 2460795.270957927], [2460795.271852712, 2460795.2724119527], [2460795.272523801, 2460795.2737541297], [2460795.273865978, 2460795.2745370665], [2460795.2747607627, 2460795.2753200033], [2460795.276214788, 2460795.2763266363], [2460795.277109573, 2460795.27789251], [2460795.278004358, 2460795.2785635986], [2460795.2786754468, 2460795.279234687], [2460795.2799057756, 2460795.281247953], [2460795.281471649, 2460795.283149371], [2460795.283261219, 2460795.283373067], [2460795.2835967634, 2460795.2838204596], [2460795.284044156, 2460795.2846033964], [2460795.2850507884, 2460795.2858337252], [2460795.286392966, 2460795.2871759026], [2460795.2878469913, 2460795.2879588394], [2460795.2880706876, 2460795.288629928], [2460795.2891891687, 2460795.289301017], [2460795.289412865, 2460795.2897484093], [2460795.2899721055, 2460795.290755042], [2460795.2912024343, 2460795.2923209155], [2460795.29265646, 2460795.293551245], [2460795.2941104854, 2460795.2952289665], [2460795.2959000547, 2460795.296235599], [2460795.296347447, 2460795.296906688], [2460795.297018536, 2460795.2974659284], [2460795.2978014727, 2460795.298248865], [2460795.2985844095, 2460795.29914365], [2460795.2997028907, 2460795.300150283], [2460795.3005976756, 2460795.301045068], [2460795.3018280044, 2460795.302163549], [2460795.302499093, 2460795.30328203], [2460795.303505726, 2460795.3039531186], [2460795.304288663, 2460795.3048479035], [2460795.305295296, 2460795.306078233], [2460795.3069730178, 2460795.3073085616], [2460795.307867802, 2460795.3080914984], [2460795.3083151947, 2460795.3089862834], [2460795.3092099796, 2460795.309433676], [2460795.3098810683, 2460795.3101047645], [2460795.310775853, 2460795.3109995495], [2460795.311670638, 2460795.3117824863], [2460795.313124663, 2460795.3132365113], [2460795.3134602075, 2460795.313683904], [2460795.3151379293, 2460795.3152497774], [2460795.3153616255, 2460795.3155853217], [2460795.31569717, 2460795.315920866], [2460795.3161445623, 2460795.3163682586], [2460795.3164801067, 2460795.316591955], [2460795.316703803, 2460795.3170393473], [2460795.317710436, 2460795.3181578284], [2460795.318381524, 2460795.3186052204], [2460795.3188289166, 2460795.319052613], [2460795.319164461, 2460795.3197237016], [2460795.3205066384, 2460795.3206184865], [2460795.321065879, 2460795.321289575], [2460795.3214014233, 2460795.3215132714], [2460795.3224080564, 2460795.323190993], [2460795.3233028413, 2460795.3235265375], [2460795.3241976257, 2460795.324421322], [2460795.32453317, 2460795.3248687144], [2460795.3250924107, 2460795.325204259], [2460795.325539803, 2460795.3257634994], [2460795.3268819805, 2460795.327217525], [2460795.327553069, 2460795.3280004617], [2460795.328224158, 2460795.328336006], [2460795.328447854, 2460795.3286715504], [2460795.3288952466, 2460795.3290070947], [2460795.329118943, 2460795.329342639], [2460795.329790031, 2460795.3300137273], [2460795.3303492717, 2460795.33046112], [2460795.331467753, 2460795.331579601], [2460795.3319151453, 2460795.3321388415], [2460795.332474386, 2460795.332698082], [2460795.33280993, 2460795.3329217783], [2460795.3331454745, 2460795.333369171], [2460795.333481019, 2460795.3338165632], [2460795.3342639557, 2460795.334487652], [2460795.3348231963, 2460795.3350468925], [2460795.3357179807, 2460795.3362772213], [2460795.3365009176, 2460795.3366127657], [2460795.337060158, 2460795.3376193987], [2460795.337731247, 2460795.3381786393], [2460795.3385141836, 2460795.33873788], [2460795.3391852723, 2460795.3395208167], [2460795.339632665, 2460795.339744513], [2460795.339968209, 2460795.3405274497], [2460795.340639298, 2460795.340751146], [2460795.3408629936, 2460795.341198538], [2460795.341422234, 2460795.3418696267], [2460795.342205171, 2460795.3426525635], [2460795.3434355003, 2460795.3435473484], [2460795.3436591965, 2460795.3438828927], [2460795.344330285, 2460795.3445539814], [2460795.345113222, 2460795.34522507], [2460795.3456724626, 2460795.346119855], [2460795.347350184, 2460795.347462032], [2460795.347685728, 2460795.3479094245], [2460795.3488042094, 2460795.3490279056], [2460795.349475298, 2460795.3496989943], [2460795.3499226905, 2460795.3500345387], [2460795.350370083, 2460795.3508174755], [2460795.3510411717, 2460795.351264868], [2460795.3520478047, 2460795.352159653], [2460795.3523833486, 2460795.3530544373], [2460795.3532781336, 2460795.35350183], [2460795.3539492222, 2460795.3541729185], [2460795.3543966147, 2460795.354620311], [2460795.3559624883, 2460795.3561861846], [2460795.356633577, 2460795.3568572733], [2460795.3577520577, 2460795.357975754], [2460795.35819945, 2460795.3586468427], [2460795.35998902, 2460795.3602127163], [2460795.3604364125, 2460795.3606601087], [2460795.3612193493, 2460795.3614430455], [2460795.362785223, 2460795.3631207673], [2460795.3635681593, 2460795.3637918555], [2460795.3641274, 2460795.3649103367], [2460795.365134033, 2460795.365357729], [2460795.3658051216, 2460795.366252514], [2460795.3668117546, 2460795.3674828433], [2460795.3677065396, 2460795.368042084], [2460795.3684894764, 2460795.369384261], [2460795.3698316533, 2460795.3700553495], [2460795.3709501345, 2460795.3710619826], [2460795.3711738307, 2460795.371621223], [2460795.3718449194, 2460795.372516008], [2460795.3728515524, 2460795.373187097], [2460795.3738581855, 2460795.374305578], [2460795.374529274, 2460795.3754240586], [2460795.375983299, 2460795.376766236], [2460795.3769899323, 2460795.377549173], [2460795.3782202615, 2460795.378555806], [2460795.378667654, 2460795.379562439], [2460795.379674287, 2460795.3804572234], [2460795.381128312, 2460795.3825823376], [2460795.3826941857, 2460795.38302973], [2460795.384148211, 2460795.3848193], [2460795.384931148, 2460795.3854903886], [2460795.385825933, 2460795.3866088693], [2460795.3868325655, 2460795.387391806], [2460795.3877273505, 2460795.388174743], [2460795.388398439, 2460795.389964313], [2460795.3905235534, 2460795.3918657303], [2460795.3928723633, 2460795.3937671483], [2460795.394326389, 2460795.394661933], [2460795.3951093256, 2460795.39544487], [2460795.3961159587, 2460795.396563351], [2460795.3967870474, 2460795.3970107436], [2460795.3981292243, 2460795.3990240092], [2460795.399918794, 2460795.400813579], [2460795.4010372753, 2460795.401708364], [2460795.4019320603, 2460795.402714997], [2460795.4028268447, 2460795.404169022], [2460795.4047282627, 2460795.4052875033], [2460795.4053993514, 2460795.4055111995], [2460795.4056230476, 2460795.4057348957], [2460795.40607044, 2460795.406965225], [2460795.4075244656, 2460795.40786001], [2460795.4081955543, 2460795.4087547944], [2460795.4089784906, 2460795.4107680605], [2460795.4117746935, 2460795.4126694784], [2460795.4127813266, 2460795.413228719], [2460795.413340567, 2460795.4136761115], [2460795.4138998077, 2460795.414794592], [2460795.4151301365, 2460795.4152419847], [2460795.415577529, 2460795.415689377], [2460795.416472314, 2460795.4171434026], [2460795.4172552507, 2460795.417702643], [2460795.4187092762, 2460795.4201633013], [2460795.4202751494, 2460795.420722542], [2460795.42083439, 2460795.4211699343], [2460795.42262396, 2460795.422959504], [2460795.4231832004, 2460795.423630593], [2460795.4241898335, 2460795.424860922], [2460795.4251964665, 2460795.4255320104], [2460795.425979403, 2460795.426203099], [2460795.4265386434, 2460795.426986036], [2460795.427209732, 2460795.4284400614], [2460795.4285519095, 2460795.428887454], [2460795.4294466944, 2460795.429894087], [2460795.4311244157, 2460795.43145996], [2460795.431571808, 2460795.4317955044], [2460795.4345917073, 2460795.4349272517], [2460795.4350391, 2460795.4357101885], [2460795.4358220366, 2460795.436045733], [2460795.436381277, 2460795.436604973], [2460795.4390656315, 2460795.439513024], [2460795.4425329226, 2460795.442868467], [2460795.4434277075, 2460795.444098796], [2460795.444658037, 2460795.444993581], [2460795.446783151, 2460795.4471186954], [2460795.447566088, 2460795.4479016317], [2460795.449243809, 2460795.4494675053], [2460795.450138594, 2460795.4503622903], [2460795.4506978346, 2460795.451033379], [2460795.451257075, 2460795.4517044676], [2460795.452040012, 2460795.4523755563], [2460795.4527111007, 2460795.453046645], [2460795.453494037, 2460795.4539414295], [2460795.454388822, 2460795.4548362144], [2460795.455283607, 2460795.4556191512], [2460795.4559546956, 2460795.456178392], [2460795.456402088, 2460795.4570731767], [2460795.457520569, 2460795.4579679617], [2460795.458415354, 2460795.4587508985], [2460795.4590864424, 2460795.4593101386], [2460795.459645683, 2460795.4599812273], [2460795.4604286198, 2460795.460652316], [2460795.4613234047, 2460795.461547101], [2460795.4622181896, 2460795.462441886], [2460795.4628892783, 2460795.4632248227], [2460795.463560367, 2460795.4640077595], [2460795.464455152, 2460795.464790696], [2460795.464902544, 2460795.4653499364], [2460795.466021025, 2460795.4664684176], [2460795.466692114, 2460795.4671395062], [2460795.4675868987, 2460795.468034291], [2460795.4682579874, 2460795.46870538], [2460795.469040924, 2460795.469712013], [2460795.469823861, 2460795.4702712535], [2460795.4707186455, 2460795.471277886], [2460795.4716134304, 2460795.472060823], [2460795.4725082153, 2460795.472955608], [2460795.473291152, 2460795.4736266965], [2460795.474074089, 2460795.4744096333], [2460795.4747451777, 2460795.474968874], [2460795.4754162664, 2460795.475863659], [2460795.4760873546, 2460795.476646595], [2460795.4768702914, 2460795.477429532], [2460795.47754138, 2460795.4781006207], [2460795.478548013, 2460795.4791072537], [2460795.479442798, 2460795.4797783424], [2460795.480337583, 2460795.4807849755], [2460795.4810086717, 2460795.481232368], [2460795.48167976, 2460795.4821271524], [2460795.482574545, 2460795.4830219373], [2460795.4831337854, 2460795.483693026], [2460795.4840285704, 2460795.484699659], [2460795.4850352034, 2460795.4853707477], [2460795.4859299883, 2460795.486377381], [2460795.486601077, 2460795.4870484695], [2460795.4872721653, 2460795.4877195577], [2460795.487943254, 2460795.489173583], [2460795.4892854313, 2460795.489844672], [2460795.4901802163, 2460795.490963153], [2460795.4911868493, 2460795.492752723], [2460795.493200115, 2460795.495213381], [2460795.4955489254, 2460795.4958844697], [2460795.496108166, 2460795.498568824], [2460795.4990162165, 2460795.499687305], [2460795.5001346976, 2460795.50058209], [2460795.5010294826, 2460795.502595356], [2460795.5028190524, 2460795.562210397]] freq_flags: [[46859741.2109375, 50888061.5234375], [51010131.8359375, 51376342.7734375], [51620483.3984375, 51742553.7109375], [52230834.9609375, 52352905.2734375], [54183959.9609375, 55282592.7734375], [55648803.7109375, 56015014.6484375], [56137084.9609375, 56503295.8984375], [56747436.5234375, 56991577.1484375], [57235717.7734375, 58212280.2734375], [58456420.8984375, 59555053.7109375], [60287475.5859375, 60409545.8984375], [60531616.2109375, 60775756.8359375], [61019897.4609375, 61264038.0859375], [61508178.7109375, 64682006.8359375], [64804077.1484375, 64926147.4609375], [65170288.0859375, 66513061.5234375], [67123413.0859375, 67367553.7109375], [69808959.9609375, 70175170.8984375], [71884155.2734375, 72128295.8984375], [72494506.8359375, 72860717.7734375], [72982788.0859375, 73226928.7109375], [73593139.6484375, 75668334.9609375], [76034545.8984375, 76156616.2109375], [76278686.5234375, 82015991.2109375], [82870483.3984375, 82992553.7109375], [83358764.6484375, 83480834.9609375], [83847045.8984375, 83969116.2109375], [84335327.1484375, 84579467.7734375], [84823608.3984375, 85067749.0234375], [85311889.6484375, 85556030.2734375], [85800170.8984375, 86044311.5234375], [86166381.8359375, 108627319.3359375], [109481811.5234375, 110458374.0234375], [111190795.8984375, 111923217.7734375], [112045288.0859375, 112411499.0234375], [112533569.3359375, 113021850.5859375], [113143920.8984375, 113876342.7734375], [114486694.3359375, 114608764.6484375], [115585327.1484375, 117050170.8984375], [117538452.1484375, 120101928.7109375], [121078491.2109375, 121444702.1484375], [124618530.2734375, 125350952.1484375], [127548217.7734375, 128524780.2734375], [128768920.8984375, 129135131.8359375], [129257202.1484375, 132675170.8984375], [135971069.3359375, 136825561.5234375], [136947631.8359375, 138168334.9609375], [138534545.8984375, 138900756.8359375], [141342163.0859375, 141830444.3359375], [141952514.6484375, 142440795.8984375], [142684936.5234375, 143295288.0859375], [143417358.3984375, 143539428.7109375], [143661499.0234375, 144271850.5859375], [145248413.0859375, 145370483.3984375], [145492553.7109375, 145614624.0234375], [145736694.3359375, 146102905.2734375], [146224975.5859375, 146347045.8984375], [147323608.3984375, 147567749.0234375], [149154663.0859375, 149276733.3984375], [149398803.7109375, 149520874.0234375], [149887084.9609375, 150009155.2734375], [154159545.8984375, 154403686.5234375], [157577514.6484375, 157821655.2734375], [159164428.7109375, 159286499.0234375], [161972045.8984375, 165390014.6484375], [165512084.9609375, 167343139.6484375], [169906616.2109375, 170150756.8359375], [170272827.1484375, 170394897.4609375], [170883178.7109375, 171005249.0234375], [174911499.0234375, 175033569.3359375], [175155639.6484375, 175521850.5859375], [180892944.3359375, 181015014.6484375], [181137084.9609375, 181381225.5859375], [182968139.6484375, 183090209.9609375], [183212280.2734375, 183334350.5859375], [187240600.5859375, 187728881.8359375], [189193725.5859375, 189315795.8984375], [189804077.1484375, 190048217.7734375], [191024780.2734375, 191513061.5234375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [198959350.5859375, 202499389.6484375], [202743530.2734375, 204208374.0234375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [207015991.2109375, 207382202.1484375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [212020874.0234375, 212265014.6484375], [213119506.8359375, 213363647.4609375], [215072631.8359375, 215438842.7734375], [215560913.0859375, 215682983.3984375], [218124389.6484375, 218246459.9609375], [220565795.8984375, 221054077.1484375], [221176147.4609375, 221298217.7734375], [222640991.2109375, 223739624.0234375], [225204467.7734375, 225326538.0859375], [225570678.7109375, 225692749.0234375], [227035522.4609375, 227157592.7734375], [227279663.0859375, 227767944.3359375], [228988647.4609375, 229476928.7109375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231307983.3984375], [231918334.9609375, 232406616.2109375], [232528686.5234375, 232650756.8359375], [233383178.7109375, 233505249.0234375], [233627319.3359375, 233871459.9609375], [234115600.5859375, 234237670.8984375]] ex_ants: [[4, Jee], [7, Jee], [7, Jnn], [8, Jee], [8, Jnn], [10, Jee], [15, Jee], [15, Jnn], [18, Jee], [18, Jnn], [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], [51, Jee], [53, Jee], [56, Jee], [56, Jnn], [60, Jnn], [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], [81, Jnn], [82, Jnn], [86, Jee], [86, Jnn], [87, Jee], [90, Jnn], [92, Jee], [97, Jnn], [98, Jnn], [99, Jnn], [100, Jnn], [102, Jnn], [104, Jnn], [105, Jee], [107, Jee], [109, Jnn], [113, Jnn], [115, Jee], [117, Jee], [120, Jee], [120, Jnn], [121, Jee], [127, Jee], [127, Jnn], [130, Jee], [130, Jnn], [135, Jee], [136, Jnn], [137, 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, Jnn], [200, Jee], [200, Jnn], [202, Jnn], [204, Jnn], [206, Jnn], [208, Jee], [209, Jnn], [212, Jnn], [213, Jee], [213, Jnn], [214, Jee], [214, Jnn], [218, Jnn], [227, Jee], [227, Jnn], [231, Jee], [231, Jnn], [236, Jee], [236, Jnn], [238, Jnn], [239, Jee], [240, Jee], [240, Jnn], [244, Jee], [250, Jee], [251, Jee], [251, Jnn], [252, Jnn], [253, Jnn], [254, Jee], [254, Jnn], [255, Jee], [255, Jnn], [256, Jee], [256, Jnn], [257, Jee], [257, Jnn], [262, Jee], [262, Jnn], [266, Jee], [266, Jnn], [267, Jee], [267, Jnn], [268, Jee], [268, Jnn], [269, Jee], [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], [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 37.73 minutes.