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 1556 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data2/2460800/zen.2460800.21086.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 1556 *.sum.smooth.calfits files starting with /mnt/sn1/data2/2460800/zen.2460800.21086.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.')
19.402% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 261 integrations and 27 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 2 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 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. 28.380% of waterfall flagged after flagging whole times and channels with median z > 1.0. 30.842% of waterfall flagged after flagging z > 4.0 outliers.
34.807% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 140 channels previously flagged 25.00% or more. Flagging 654 times previously flagged 10.00% or more.
Mean of empty slice Mean of empty slice
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. 53.402% 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
63.410% of flagging channels that are 4.0σ outliers after delay filtering the time average.
64.317% 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 1556 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data2/2460800/zen.2460800.21086.sum.flag_waterfall_round_2.h5.
In [23]:
# write summary of entirely flagged times/freqs/ants to yaml
all_flagged_times = np.all(flags, axis=1)
all_flagged_freqs = np.all(flags, axis=0)
all_flagged_ants = [(int(ant[0]), ant[1]) for ant in sorted(always_flagged_ants)]
dt = np.median(np.diff(times))
out_yml_str = 'JD_flags: ' + str([[float(times[flag_stretch][0] - dt / 2), float(times[flag_stretch][-1] + dt / 2)]
for flag_stretch in true_stretches(all_flagged_times)])
df = np.median(np.diff(freqs))
out_yml_str += '\n\nfreq_flags: ' + str([[float(freqs[flag_stretch][0] - df / 2), float(freqs[flag_stretch][-1] + df / 2)]
for flag_stretch in true_stretches(all_flagged_freqs)])
out_yml_str += '\n\nex_ants: ' + str(all_flagged_ants).replace("'", "").replace('(', '[').replace(')', ']')
print(f'Writing the following to {out_yaml_file}\n' + '-' * (25 + len(out_yaml_file)))
print(out_yml_str)
with open(out_yaml_file, 'w') as outfile:
outfile.writelines(out_yml_str)
Writing the following to /mnt/sn1/data2/2460800/2460800_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460800.2107488783, 2460800.2205915116], [2460800.2207033597, 2460800.222045537], [2460800.222157385, 2460800.2242824994], [2460800.2249535876, 2460800.226295765], [2460800.227302398, 2460800.2275260943], [2460800.2279734868, 2460800.2312170817], [2460800.2340132846, 2460800.2362502464], [2460800.237704272, 2460800.2383753606], [2460800.2384872087, 2460800.240500475], [2460800.240724171, 2460800.241283411], [2460800.2413952593, 2460800.2419545], [2460800.244974399, 2460800.2455336396], [2460800.245757336, 2460800.2466521207], [2460800.2469876646, 2460800.2481061458], [2460800.248329842, 2460800.2486653863], [2460800.2496720194, 2460800.250119412], [2460800.251349741, 2460800.2514615892], [2460800.25202083, 2460800.252132678], [2460800.2634293367, 2460800.263988577], [2460800.267232172, 2460800.267791413], [2460800.26913359, 2460800.2692454383], [2460800.269357286, 2460800.269580982], [2460800.2708113114, 2460800.2709231596], [2460800.2718179445, 2460800.2719297926], [2460800.272153489, 2460800.272265337], [2460800.2726008813, 2460800.2738312106], [2460800.2752852356, 2460800.275844476], [2460800.277634046, 2460800.2780814385], [2460800.2783051347, 2460800.279311768], [2460800.2799828565, 2460800.280653945], [2460800.280989489, 2460800.281772426], [2460800.283002755, 2460800.2831146033], [2460800.2885951605, 2460800.2887070086], [2460800.291055819, 2460800.2928453884], [2460800.293292781, 2460800.2938520215], [2460800.297654857, 2460800.297766705], [2460800.2979904013, 2460800.2981022494], [2460800.298885186, 2460800.2989970343], [2460800.3011221485, 2460800.301569541], [2460800.302688022, 2460800.3031354146], [2460800.303806503, 2460800.3043657434], [2460800.3087278195, 2460800.309175212], [2460800.3118595667, 2460800.312306959], [2460800.3166690352, 2460800.31756382], [2460800.3189059976, 2460800.319241542], [2460800.31935339, 2460800.3196889344], [2460800.3202481745, 2460800.320695567], [2460800.322485137, 2460800.322596985], [2460800.3231562255, 2460800.3233799217], [2460800.3250576435, 2460800.325393188], [2460800.325505036, 2460800.325952428], [2460800.329867112, 2460800.330090808], [2460800.334341036, 2460800.3347884286], [2460800.335347669, 2460800.3355713654], [2460800.336130606, 2460800.3365779985], [2460800.343624429, 2460800.343848125], [2460800.3441836694, 2460800.3444073657], [2460800.3472035686, 2460800.3473154167], [2460800.350111619, 2460800.3503353153], [2460800.3514537965, 2460800.3516774927], [2460800.351901189, 2460800.352124885], [2460800.3525722777, 2460800.352795974], [2460800.35301967, 2460800.3531315182], [2460800.3534670626, 2460800.3535789107], [2460800.3536907583, 2460800.3538026065], [2460800.357269898, 2460800.3576054424], [2460800.3585002273, 2460800.3587239236], [2460800.3601779486, 2460800.360401645], [2460800.361408278, 2460800.361520126], [2460800.361743822, 2460800.3620793666], [2460800.363533392, 2460800.3637570883], [2460800.3648755695, 2460800.3650992652], [2460800.3652111134, 2460800.3653229615], [2460800.365770354, 2460800.3661058983], [2460800.367559924, 2460800.3680073163], [2460800.3682310125, 2460800.3684547087], [2460800.371139063, 2460800.371250911], [2460800.3714746074, 2460800.3716983036], [2460800.3727049367, 2460800.372928633], [2460800.3735997216, 2460800.373823418], [2460800.3744945065, 2460800.3747182027], [2460800.374830051, 2460800.374941899], [2460800.375053747, 2460800.3753892914], [2460800.3776262533, 2460800.3780736458], [2460800.378521038, 2460800.3787447345], [2460800.379415823, 2460800.3796395194], [2460800.380086912, 2460800.380310608], [2460800.3808698487, 2460800.3818764817], [2460800.38254757, 2460800.382771266], [2460800.3832186586, 2460800.383554203], [2460800.3838897473, 2460800.3841134435], [2460800.3860148615, 2460800.3862385578], [2460800.3894821526, 2460800.389817697], [2460800.389929545, 2460800.3902650895], [2460800.3911598744, 2460800.3914954187], [2460800.391719115, 2460800.391830963], [2460800.391942811, 2460800.3922783555], [2460800.394179773, 2460800.3944034693], [2460800.398094457, 2460800.3984300015], [2460800.3988773935, 2460800.3991010897], [2460800.401338052, 2460800.4015617482], [2460800.4017854445, 2460800.4020091407], [2460800.4027920775, 2460800.4030157737], [2460800.403463166, 2460800.4036868624], [2460800.404022407, 2460800.404246103], [2460800.4088318753, 2460800.4090555715], [2460800.4115162296, 2460800.411739926], [2460800.4122991664, 2460800.4125228627], [2460800.416102002, 2460800.41621385], [2460800.418898205, 2460800.419010053], [2460800.4195692935, 2460800.419904838], [2460800.422365496, 2460800.422589192], [2460800.4246024583, 2460800.4247143064], [2460800.426392028, 2460800.4269512687], [2460800.4303067117, 2460800.43041856], [2460800.437017598, 2460800.4374649907], [2460800.43925456, 2460800.4397019525], [2460800.440037497, 2460800.440373041], [2460800.4408204337, 2460800.441267826], [2460800.442050763, 2460800.4428336998], [2460800.443169244, 2460800.4435047884], [2460800.4439521804, 2460800.444399573], [2460800.444623269, 2460800.4447351173], [2460800.4450706616, 2460800.445518054], [2460800.4461891428, 2460800.4467483833], [2460800.4470839277, 2460800.44753132], [2460800.4478668645, 2460800.4480905607], [2460800.4486498013, 2460800.449097194], [2460800.44932089, 2460800.449768282], [2460800.4499919782, 2460800.4504393707], [2460800.450663067, 2460800.4511104594], [2460800.4513341556, 2460800.451781548], [2460800.4520052443, 2460800.452452637], [2460800.452676333, 2460800.4532355736], [2460800.453571118, 2460800.4542422066], [2460800.454465903, 2460800.4549132953], [2460800.455136991, 2460800.45580808], [2460800.456031776, 2460800.4564791685], [2460800.4567028647, 2460800.457150257], [2460800.4573739534, 2460800.4575976497], [2460800.458045042, 2460800.4583805865], [2460800.458716131, 2460800.4591635233], [2460800.4596109157, 2460800.460058308], [2460800.4607293964, 2460800.461176789], [2460800.4628545106, 2460800.463190055], [2460800.46408484, 2460800.4647559286], [2460800.464979625, 2460800.4654270173], [2460800.4657625617, 2460800.466657346], [2460800.4671047386, 2460800.469453549], [2460800.472026055, 2460800.4723615996], [2460800.4730326883, 2460800.4733682326], [2460800.474934106, 2460800.4752696506], [2460800.4762762836, 2460800.4763881317], [2460800.478960638, 2460800.479072486], [2460800.4795198785, 2460800.480079119], [2460800.4845530433, 2460800.4848885876], [2460800.491487626, 2460800.4918231703], [2460800.493500892, 2460800.4939482845], [2460800.5027842848, 2460800.5097188675], [2460800.510389956, 2460800.5105018043], [2460800.5113965888, 2460800.5167652983], [2460800.5168771464, 2460800.517660083], [2460800.517771931, 2460800.5181074752], [2460800.5183311715, 2460800.5191141083], [2460800.5192259564, 2460800.5211273744], [2460800.5212392225, 2460800.5213510706], [2460800.5214629187, 2460800.521798463], [2460800.5222458555, 2460800.522469552], [2460800.5228050957, 2460800.523364336], [2460800.5243709693, 2460800.5245946655], [2460800.52493021, 2460800.525042058], [2460800.525153906, 2460800.525265754], [2460800.5254894504, 2460800.5256012985], [2460800.5257131467, 2460800.525936843], [2460800.526048691, 2460800.5262723872], [2460800.5263842354, 2460800.526943476], [2460800.527055324, 2460800.5273908684], [2460800.5277264128, 2460800.528173805], [2460800.528397501, 2460800.528509349], [2460800.5326477294, 2460800.5348846912], [2460800.5403652485, 2460800.5405889447], [2460800.5464050462, 2460800.5467405906], [2460800.546964287, 2460800.553451477], [2460800.5537870214, 2460800.554122566], [2460800.554346262, 2460800.5564713757], [2460800.55680692, 2460800.557478009], [2460800.557701705, 2460800.5581490975], [2460800.5582609456, 2460800.55859649], [2460800.558820186, 2460800.5590438824], [2460800.5592675786, 2460800.5600505155], [2460800.5601623636, 2460800.562175629]] freq_flags: [[46859741.2109375, 53085327.1484375], [53207397.4609375, 53451538.0859375], [53573608.3984375, 53939819.3359375], [54061889.6484375, 54550170.8984375], [54672241.2109375, 54794311.5234375], [55160522.4609375, 56015014.6484375], [56259155.2734375, 56381225.5859375], [56625366.2109375, 56991577.1484375], [57113647.4609375, 57479858.3984375], [57601928.7109375, 64315795.8984375], [64559936.5234375, 64926147.4609375], [65292358.3984375, 65414428.7109375], [69808959.9609375, 70175170.8984375], [70785522.4609375, 70907592.7734375], [71395874.0234375, 71517944.3359375], [72616577.1484375, 72860717.7734375], [73104858.3984375, 73959350.5859375], [74081420.8984375, 75302124.0234375], [75424194.3359375, 75668334.9609375], [75912475.5859375, 76156616.2109375], [77743530.2734375, 77987670.8984375], [78109741.2109375, 78231811.5234375], [78353881.8359375, 78475952.1484375], [82382202.1484375, 82504272.4609375], [83969116.2109375, 84091186.5234375], [87020874.0234375, 108383178.7109375], [109725952.1484375, 109848022.4609375], [109970092.7734375, 110092163.0859375], [112045288.0859375, 112167358.3984375], [112289428.7109375, 112411499.0234375], [112655639.6484375, 112777709.9609375], [113265991.2109375, 113510131.8359375], [113632202.1484375, 113754272.4609375], [116073608.3984375, 116195678.7109375], [116439819.3359375, 116561889.6484375], [116683959.9609375, 116806030.2734375], [124862670.8984375, 125228881.8359375], [126205444.3359375, 130233764.6484375], [130599975.5859375, 131576538.0859375], [131698608.3984375, 131942749.0234375], [136947631.8359375, 138534545.8984375], [138656616.2109375, 138778686.5234375], [139511108.3984375, 139633178.7109375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [142074584.9609375, 144027709.9609375], [145492553.7109375, 145614624.0234375], [145736694.3359375, 145980834.9609375], [146224975.5859375, 146347045.8984375], [146469116.2109375, 147323608.3984375], [147445678.7109375, 147567749.0234375], [147689819.3359375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [152938842.7734375, 153060913.0859375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158309936.5234375], [158676147.4609375, 158798217.7734375], [160140991.2109375, 160385131.8359375], [161361694.3359375, 161483764.6484375], [162704467.7734375, 165878295.8984375], [166000366.2109375, 169296264.6484375], [169906616.2109375, 170639038.0859375], [170883178.7109375, 172103881.8359375], [175033569.3359375, 175399780.2734375], [181015014.6484375, 181503295.8984375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [189926147.4609375, 190048217.7734375], [190292358.3984375, 190414428.7109375], [191024780.2734375, 191513061.5234375], [195663452.1484375, 195785522.4609375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [199203491.2109375, 199325561.5234375], [199935913.0859375, 200057983.3984375], [201522827.1484375, 206527709.9609375], [207138061.5234375, 207382202.1484375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [212142944.3359375, 212265014.6484375], [214584350.5859375, 214706420.8984375], [215194702.1484375, 215316772.4609375], [215682983.3984375, 215805053.7109375], [216049194.3359375, 216293334.9609375], [216415405.2734375, 216537475.5859375], [219955444.3359375, 220443725.5859375], [220687866.2109375, 220809936.5234375], [221176147.4609375, 221298217.7734375], [221542358.3984375, 221786499.0234375], [221908569.3359375, 222152709.9609375], [222640991.2109375, 224105834.9609375], [226913452.1484375, 227035522.4609375], [227401733.3984375, 227767944.3359375], [229110717.7734375, 229354858.3984375], [229721069.3359375, 229843139.6484375], [229965209.9609375, 230087280.2734375], [230209350.5859375, 230575561.5234375], [231063842.7734375, 231185913.0859375], [232040405.2734375, 232162475.5859375], [233139038.0859375, 233505249.0234375], [233627319.3359375, 233749389.6484375], [234115600.5859375, 234359741.2109375]] ex_ants: [[4, Jee], [7, Jee], [7, Jnn], [8, Jnn], [10, Jee], [15, Jee], [15, Jnn], [16, Jee], [17, 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, Jnn], [34, Jee], [35, Jnn], [37, Jnn], [40, Jnn], [42, Jnn], [44, Jee], [46, Jee], [47, Jnn], [51, Jee], [56, Jee], [56, Jnn], [60, Jnn], [62, Jee], [65, 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, Jee], [82, Jnn], [83, Jee], [83, Jnn], [85, Jnn], [86, Jee], [86, Jnn], [87, Jee], [88, Jee], [88, Jnn], [90, Jee], [90, Jnn], [97, Jnn], [99, Jnn], [102, Jnn], [103, Jee], [103, Jnn], [104, Jee], [104, Jnn], [105, Jee], [107, Jee], [107, Jnn], [108, Jee], [109, Jnn], [111, Jee], [115, Jee], [115, Jnn], [117, Jee], [120, Jee], [120, Jnn], [121, Jee], [125, Jee], [125, Jnn], [130, Jee], [130, Jnn], [131, Jnn], [135, Jee], [136, Jnn], [137, Jee], [137, Jnn], [140, Jee], [143, Jee], [143, Jnn], [148, Jee], [153, Jnn], [155, Jnn], [158, Jee], [158, Jnn], [159, Jee], [161, Jnn], [166, Jee], [166, Jnn], [167, Jnn], [169, Jee], [170, Jee], [172, Jnn], [173, Jnn], [174, Jnn], [176, Jnn], [180, Jee], [180, Jnn], [182, Jee], [184, Jee], [184, Jnn], [185, Jee], [185, Jnn], [186, Jee], [186, Jnn], [188, Jnn], [189, Jnn], [194, Jnn], [197, Jnn], [199, Jnn], [200, Jee], [200, Jnn], [202, Jnn], [204, Jnn], [206, Jnn], [209, Jnn], [212, Jnn], [214, Jee], [214, Jnn], [215, Jee], [215, Jnn], [216, Jee], [218, Jnn], [227, Jee], [227, Jnn], [231, Jee], [231, Jnn], [233, Jnn], [236, Jee], [236, Jnn], [238, Jnn], [239, Jee], [240, Jnn], [244, Jee], [246, Jee], [250, Jee], [251, Jee], [251, Jnn], [252, Jnn], [253, Jee], [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], [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 35.22 minutes.