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 1545 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data1/2460799/zen.2460799.21203.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 1545 *.sum.smooth.calfits files starting with /mnt/sn1/data1/2460799/zen.2460799.21203.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.')
30.155% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 1113 integrations and 174 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 12 times previously flagged 10.00% or more.
Flagging an additional 166 integrations and 8 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.
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. 68.442% of waterfall flagged after flagging whole times and channels with median z > 1.0. 69.120% of waterfall flagged after flagging z > 4.0 outliers.
71.581% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 128 channels previously flagged 25.00% or more. Flagging 405 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. 80.590% 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
85.457% of flagging channels that are 4.0σ outliers after delay filtering the time average.
85.583% 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 1545 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data1/2460799/zen.2460799.21203.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/data1/2460799/2460799_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460799.211914401, 2460799.251844175], [2460799.2519560233, 2460799.2669436694], [2460799.2670555175, 2460799.2755559734], [2460799.276003366, 2460799.276115214], [2460799.2764507583, 2460799.2786877207], [2460799.278799569, 2460799.279246961], [2460799.2796943537, 2460799.280029898], [2460799.2803654424, 2460799.280589138], [2460799.2808128344, 2460799.2811483787], [2460799.281260227, 2460799.2829379486], [2460799.2837208854, 2460799.2839445816], [2460799.284280126, 2460799.2847275184], [2460799.2848393666, 2460799.2864052397], [2460799.286852632, 2460799.2881948096], [2460799.288418506, 2460799.28875405], [2460799.2892014426, 2460799.289536987], [2460799.2899843794, 2460799.2902080757], [2460799.291326557, 2460799.291997645], [2460799.2922213413, 2460799.292780582], [2460799.293116126, 2460799.2933398224], [2460799.293899063, 2460799.2942346074], [2460799.294905696, 2460799.2951293923], [2460799.2953530885, 2460799.295800481], [2460799.2962478735, 2460799.296695266], [2460799.296918962, 2460799.2978137466], [2460799.2979255947, 2460799.298372987], [2460799.2984848353, 2460799.2987085315], [2460799.298932228, 2460799.2996033165], [2460799.2997151646, 2460799.300274405], [2460799.3004981014, 2460799.300945494], [2460799.3013928863, 2460799.3016165826], [2460799.301840279, 2460799.302063975], [2460799.3025113675, 2460799.3027350637], [2460799.3029587595, 2460799.3031824557], [2460799.303406152, 2460799.3038535444], [2460799.3040772406, 2460799.304524633], [2460799.3049720256, 2460799.305195722], [2460799.305419418, 2460799.3058668105], [2460799.306314203, 2460799.306537899], [2460799.3069852917, 2460799.307208988], [2460799.3078800766, 2460799.308327469], [2460799.3085511653, 2460799.3089985573], [2460799.3092222535, 2460799.3094459497], [2460799.3101170384, 2460799.310564431], [2460799.3110118234, 2460799.311459216], [2460799.311682912, 2460799.3123540008], [2460799.312577697, 2460799.312801393], [2460799.3130250894, 2460799.3132487857], [2460799.3138080263, 2460799.3139198744], [2460799.3140317225, 2460799.3141435706], [2460799.3143672664, 2460799.3145909626], [2460799.315038355, 2460799.3154857475], [2460799.31593314, 2460799.317051621], [2460799.3172753174, 2460799.317946406], [2460799.318058254, 2460799.318953039], [2460799.3190648872, 2460799.3195122797], [2460799.319735976, 2460799.320183368], [2460799.320407064, 2460799.3208544566], [2460799.321078153, 2460799.321301849], [2460799.3217492416, 2460799.322196634], [2460799.3224203303, 2460799.3226440265], [2460799.3227558746, 2460799.323091419], [2460799.323203267, 2460799.323315115], [2460799.323986204, 2460799.3245454445], [2460799.324880989, 2460799.325104685], [2460799.3258876214, 2460799.3262231657], [2460799.326446862, 2460799.3268942544], [2460799.327341647, 2460799.3277890394], [2460799.328236432, 2460799.328460128], [2460799.3289075205, 2460799.329354913], [2460799.3296904573, 2460799.3299141536], [2460799.3300260017, 2460799.3306970904], [2460799.331815571, 2460799.3320392673], [2460799.332822204, 2460799.3330459003], [2460799.3333814447, 2460799.3339406853], [2460799.3342762296, 2460799.334499926], [2460799.3351710145, 2460799.335618407], [2460799.335842103, 2460799.3362894957], [2460799.3367368877, 2460799.336960584], [2460799.3385264575, 2460799.338862002], [2460799.33897385, 2460799.3396449387], [2460799.340092331, 2460799.3403160274], [2460799.34076342, 2460799.3412108123], [2460799.341658205, 2460799.3423292935], [2460799.3427766855, 2460799.3430003817], [2460799.343224078, 2460799.3436714704], [2460799.344118863, 2460799.344342559], [2460799.345013648, 2460799.345237344], [2460799.345349192, 2460799.3459084327], [2460799.346132129, 2460799.3465795214], [2460799.347026914, 2460799.34725061], [2460799.348369091, 2460799.348592787], [2460799.350606053, 2460799.351277142], [2460799.352283775, 2460799.3527311673], [2460799.354856281, 2460799.354968129], [2460799.3551918254, 2460799.3554155217], [2460799.356757699, 2460799.3570932434], [2460799.357652484, 2460799.3582117246], [2460799.358547269, 2460799.3605605345], [2460799.361119775, 2460799.3613434713], [2460799.361790864, 2460799.3622382563], [2460799.3624619525, 2460799.3686135984], [2460799.369172839, 2460799.3695083833], [2460799.370067624, 2460799.37029132], [2460799.370626864, 2460799.3709624084], [2460799.3711861046, 2460799.373423067], [2460799.3738704594, 2460799.3739823075], [2460799.374541548, 2460799.3749889405], [2460799.3756600292, 2460799.3758837255], [2460799.3763311175, 2460799.3772259024], [2460799.3774495986, 2460799.377673295], [2460799.377896991, 2460799.37856808], [2460799.378791776, 2460799.3790154723], [2460799.3794628647, 2460799.379686561], [2460799.3801339534, 2460799.380581346], [2460799.3810287383, 2460799.381699827], [2460799.381923523, 2460799.382035371], [2460799.3824827634, 2460799.3828183077], [2460799.383042004, 2460799.3834893964], [2460799.383936789, 2460799.384048637], [2460799.3846078776, 2460799.38505527], [2460799.3855026625, 2460799.388187017], [2460799.388410713, 2460799.3890818018], [2460799.3906476754, 2460799.3908713716], [2460799.391095068, 2460799.391318764], [2460799.391989853, 2460799.392213549], [2460799.3926609415, 2460799.3928846377], [2460799.393779422, 2460799.3942268146], [2460799.394674207, 2460799.395345296], [2460799.395568992, 2460799.3957926882], [2460799.3960163845, 2460799.396463777], [2460799.396687473, 2460799.3978059543], [2460799.398253347, 2460799.398477043], [2460799.3995955237, 2460799.39981922], [2460799.4011613973, 2460799.40160879], [2460799.401832486, 2460799.4021680304], [2460799.4025035747, 2460799.402727271], [2460799.402950967, 2460799.4030628153], [2460799.4044049922, 2460799.4046286885], [2460799.4056353215, 2460799.4057471696], [2460799.4058590177, 2460799.405970866], [2460799.4065301064, 2460799.406977499], [2460799.4077604357, 2460799.407872284], [2460799.408990765, 2460799.409102613], [2460799.4110040306, 2460799.411227727], [2460799.4139120816, 2460799.414135778], [2460799.4149187147, 2460799.415254259], [2460799.415813499, 2460799.4160371954], [2460799.417826765, 2460799.4179386133], [2460799.4234191705, 2460799.423642867], [2460799.428228639, 2460799.4284523353], [2460799.4349395256, 2460799.435163222], [2460799.4384068167, 2460799.438630513], [2460799.438742361, 2460799.4389660573], [2460799.4403082347, 2460799.440531931], [2460799.440643779, 2460799.4409793233], [2460799.441426716, 2460799.441650412], [2460799.4423215007, 2460799.442545197], [2460799.4429925894, 2460799.443439982], [2460799.443887374, 2460799.4442229182], [2460799.4443347664, 2460799.444782159], [2460799.4452295513, 2460799.4454532475], [2460799.44590064, 2460799.446124336], [2460799.446795425, 2460799.446907273], [2460799.447019121, 2460799.4472428174], [2460799.447913906, 2460799.4481376023], [2460799.448808691, 2460799.449032387], [2460799.449479779, 2460799.4498153236], [2460799.450374564, 2460799.4505982604], [2460799.4508219566, 2460799.451269349], [2460799.4514930453, 2460799.451940438], [2460799.452275982, 2460799.4527233746], [2460799.453170767, 2460799.4536181595], [2460799.4538418557, 2460799.454289248], [2460799.4545129444, 2460799.454848489], [2460799.4551840327, 2460799.455519577], [2460799.4561906657, 2460799.4567499063], [2460799.4568617544, 2460799.457420995], [2460799.457644691, 2460799.4580920837], [2460799.458427628, 2460799.4588750205], [2460799.4590987167, 2460799.4596579573], [2460799.4598816535, 2460799.460329046], [2460799.460440894, 2460799.46066459], [2460799.460888286, 2460799.4612238305], [2460799.4613356786, 2460799.461559375], [2460799.461671223, 2460799.4623423116], [2460799.462789704, 2460799.463460793], [2460799.4639081853, 2460799.4643555777], [2460799.464579274, 2460799.4650266664], [2460799.4652503626, 2460799.465697755], [2460799.4659214513, 2460799.4663688433], [2460799.4665925396, 2460799.4677110207], [2460799.468046565, 2460799.468829502], [2460799.4691650462, 2460799.4693887425], [2460799.469724287, 2460799.470842768], [2460799.471066464, 2460799.4739745148], [2460799.474198211, 2460799.4746456034], [2460799.474981148, 2460799.4757640846], [2460799.475987781, 2460799.478672135], [2460799.4788958314, 2460799.479343224], [2460799.47956692, 2460799.4801261607], [2460799.480238009, 2460799.4809090975], [2460799.4811327937, 2460799.4818038824], [2460799.482251275, 2460799.4826986673], [2460799.4829223636, 2460799.4840408443], [2460799.4842645405, 2460799.484711933], [2460799.4851593254, 2460799.485606718], [2460799.4860541103, 2460799.486501503], [2460799.486725199, 2460799.4871725915], [2460799.487619984, 2460799.4880673764], [2460799.488514769, 2460799.489074009], [2460799.4892977052, 2460799.4907517307], [2460799.491087275, 2460799.4924294525], [2460799.4926531487, 2460799.4946664143], [2460799.4947782625, 2460799.55931462]] freq_flags: [[46859741.2109375, 79330444.3359375], [79452514.6484375, 79818725.5859375], [79940795.8984375, 83724975.5859375], [83847045.8984375, 84335327.1484375], [84457397.4609375, 84823608.3984375], [84945678.7109375, 85678100.5859375], [85800170.8984375, 86532592.7734375], [86654663.0859375, 114120483.3984375], [116073608.3984375, 116195678.7109375], [116439819.3359375, 116561889.6484375], [116683959.9609375, 116806030.2734375], [120223999.0234375, 120346069.3359375], [124740600.5859375, 125228881.8359375], [126693725.5859375, 129379272.4609375], [129867553.7109375, 130355834.9609375], [130722045.8984375, 134017944.3359375], [135726928.7109375, 135848999.0234375], [136337280.2734375, 136459350.5859375], [136947631.8359375, 138290405.2734375], [138534545.8984375, 138900756.8359375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [142074584.9609375, 142318725.5859375], [142562866.2109375, 144149780.2734375], [145492553.7109375, 145614624.0234375], [145858764.6484375, 146102905.2734375], [146224975.5859375, 146347045.8984375], [147445678.7109375, 147567749.0234375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158309936.5234375], [158676147.4609375, 158798217.7734375], [159164428.7109375, 159286499.0234375], [160140991.2109375, 160263061.5234375], [161361694.3359375, 161483764.6484375], [162948608.3984375, 165634155.2734375], [166000366.2109375, 168441772.4609375], [169906616.2109375, 170150756.8359375], [170272827.1484375, 170394897.4609375], [170883178.7109375, 171005249.0234375], [171737670.8984375, 171859741.2109375], [174911499.0234375, 175033569.3359375], [175155639.6484375, 175399780.2734375], [181015014.6484375, 181381225.5859375], [183212280.2734375, 183334350.5859375], [186386108.3984375, 186508178.7109375], [187240600.5859375, 187850952.1484375], [189193725.5859375, 191635131.8359375], [194198608.3984375, 194442749.0234375], [195663452.1484375, 195785522.4609375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [199203491.2109375, 199325561.5234375], [199935913.0859375, 202255249.0234375], [202377319.3359375, 205307006.8359375], [207138061.5234375, 207382202.1484375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [212142944.3359375, 212265014.6484375], [213119506.8359375, 213241577.1484375], [215194702.1484375, 215316772.4609375], [216537475.5859375, 216659545.8984375], [220565795.8984375, 220809936.5234375], [221176147.4609375, 221298217.7734375], [222763061.5234375, 223739624.0234375], [226791381.8359375, 227035522.4609375], [227401733.3984375, 227523803.7109375], [227645874.0234375, 227767944.3359375], [229110717.7734375, 229354858.3984375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231307983.3984375], [232284545.8984375, 234237670.8984375]] ex_ants: [[4, Jee], [7, Jee], [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], [29, Jee], [29, Jnn], [30, Jee], [30, Jnn], [31, Jee], [32, Jnn], [33, Jee], [33, Jnn], [34, Jee], [35, Jnn], [37, Jee], [37, Jnn], [40, Jnn], [42, Jee], [45, Jee], [46, Jee], [47, 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], [75, Jee], [76, Jee], [76, Jnn], [77, Jnn], [78, Jee], [81, Jnn], [82, Jee], [82, Jnn], [86, Jee], [86, Jnn], [87, Jee], [90, Jnn], [99, Jnn], [102, Jnn], [104, Jnn], [105, Jee], [107, Jee], [108, Jnn], [109, Jnn], [111, Jee], [115, Jee], [117, Jee], [120, Jee], [120, Jnn], [121, Jee], [125, Jee], [125, Jnn], [127, Jee], [130, Jee], [130, Jnn], [135, Jee], [136, Jnn], [137, Jnn], [143, Jnn], [153, Jnn], [155, Jnn], [158, Jnn], [159, Jnn], [161, Jnn], [166, Jee], [166, Jnn], [167, Jnn], [170, Jee], [172, Jnn], [173, Jnn], [174, Jee], [175, Jee], [175, Jnn], [176, Jnn], [180, Jee], [180, Jnn], [182, Jee], [184, Jee], [184, Jnn], [185, Jee], [185, Jnn], [186, Jee], [186, Jnn], [188, Jnn], [195, Jee], [195, Jnn], [197, Jnn], [199, Jnn], [200, Jee], [200, Jnn], [202, Jnn], [204, Jnn], [206, Jnn], [209, Jnn], [212, Jnn], [214, Jee], [214, Jnn], [215, Jnn], [216, Jee], [218, Jnn], [226, Jnn], [227, Jee], [227, Jnn], [231, Jee], [231, Jnn], [233, Jnn], [236, Jee], [236, Jnn], [238, Jnn], [239, Jee], [240, Jee], [240, Jnn], [244, Jee], [250, Jee], [251, Jee], [251, Jnn], [252, 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], [269, Jnn], [271, Jee], [271, Jnn], [273, Jee], [273, 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 33.61 minutes.