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 1570 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data1/2460783/zen.2460783.21087.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 1570 *.sum.smooth.calfits files starting with /mnt/sn1/data1/2460783/zen.2460783.21087.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.')
24.160% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 1 integrations and 6 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 3 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 5 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. 24.880% of waterfall flagged after flagging whole times and channels with median z > 1.0. 26.558% of waterfall flagged after flagging z > 4.0 outliers.
29.146% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 81 channels previously flagged 25.00% or more. Flagging 425 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. 41.388% 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
44.551% of flagging channels that are 4.0σ outliers after delay filtering the time average.
44.950% 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 1570 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data1/2460783/zen.2460783.21087.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/data1/2460783/2460783_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460783.2110965974, 2460783.2113202936], [2460783.212215078, 2460783.2124387743], [2460783.213221711, 2460783.2135572555], [2460783.2180311796, 2460783.218366724], [2460783.219149661, 2460783.219373357], [2460783.220156294, 2460783.220268142], [2460783.22216956, 2460783.2225051043], [2460783.2232880406, 2460783.2233998887], [2460783.223511737, 2460783.223847281], [2460783.226419788, 2460783.2268671803], [2460783.2270908765, 2460783.2272027247], [2460783.2277619652, 2460783.2282093577], [2460783.228433054, 2460783.228544902], [2460783.229104142, 2460783.2293278384], [2460783.232012193, 2460783.2322358894], [2460783.2378282947, 2460783.238163839], [2460783.2409600415, 2460783.2411837378], [2460783.241295586, 2460783.241519282], [2460783.245881358, 2460783.2459932063], [2460783.2461050544, 2460783.2463287506], [2460783.2533751815, 2460783.2534870296], [2460783.2535988777, 2460783.253822574], [2460783.2563950806, 2460783.256618777], [2460783.257066169, 2460783.257289865], [2460783.2587438906, 2460783.259079435], [2460783.2627704223, 2460783.2631059666], [2460783.263441511, 2460783.263665207], [2460783.2650073846, 2460783.265566625], [2460783.2667969544, 2460783.267244347], [2460783.2684746757, 2460783.268922068], [2460783.2692576125, 2460783.2693694606], [2460783.269705005, 2460783.269816853], [2460783.272501208, 2460783.2728367522], [2460783.274067081, 2460783.2742907773], [2460783.2744026254, 2460783.275185562], [2460783.2754092584, 2460783.2755211066], [2460783.2756329547, 2460783.275744803], [2460783.277981765, 2460783.27887655], [2460783.280330575, 2460783.2805542713], [2460783.2807779675, 2460783.2858111323], [2460783.286370373, 2460783.286594069], [2460783.2869296134, 2460783.2871533097], [2460783.287488854, 2460783.2914035376], [2460783.291739082, 2460783.291962778], [2460783.2922983225, 2460783.292857563], [2460783.2931931075, 2460783.293528652], [2460783.2936405, 2460783.293752348], [2460783.2939760443, 2460783.2943115886], [2460783.294535285, 2460783.294870829], [2460783.2974433354, 2460783.298002576], [2460783.298114424, 2460783.3015817157], [2460783.3016935633, 2460783.303147589], [2460783.303371285, 2460783.303483133], [2460783.3035949813, 2460783.304042374], [2460783.30426607, 2460783.3118717414], [2460783.314667944, 2460783.3152271844], [2460783.316793058, 2460783.3170167543], [2460783.31847078, 2460783.3188063237], [2460783.3223854634, 2460783.3226091596], [2460783.323056552, 2460783.3232802483], [2460783.3233920964, 2460783.3246224253], [2460783.3247342734, 2460783.328313413], [2460783.328425261, 2460783.3285371093], [2460783.3287608055, 2460783.3288726537], [2460783.3303266787, 2460783.330774071], [2460783.3337939703, 2460783.3342413628], [2460783.3349124514, 2460783.335247996], [2460783.336030932, 2460783.3364783246], [2460783.3372612614, 2460783.3374849577], [2460783.3438603, 2460783.343972148], [2460783.344195844, 2460783.3447550847], [2460783.3467683503, 2460783.3479986796], [2460783.348334224, 2460783.3488934645], [2460783.350235642, 2460783.3505711863], [2460783.3509067306, 2460783.3510185787], [2460783.351130427, 2460783.3516896674], [2460783.3601901233, 2460783.3605256677], [2460783.3634337187, 2460783.364216655], [2460783.3644403513, 2460783.3647758956], [2460783.3666773136, 2460783.367012858], [2460783.3701446047, 2460783.370480149], [2460783.371263086, 2460783.37159863], [2460783.372157871, 2460783.3727171114], [2460783.3738355925, 2460783.3749540737], [2460783.3750659213, 2460783.3751777695], [2460783.377638428, 2460783.3778621242], [2460783.3868099726, 2460783.3879284537], [2460783.390165416, 2460783.3905009604], [2460783.3920668336, 2460783.39229053], [2460783.392402378, 2460783.3928497704], [2460783.3968763025, 2460783.397211847], [2460783.399672505, 2460783.399896201], [2460783.405153062, 2460783.4056004547], [2460783.408620354, 2460783.409067746], [2460783.412087645, 2460783.4128705817], [2460783.424726481, 2460783.4249501773], [2460783.427858228, 2460783.428081924], [2460783.433115089, 2460783.433338785], [2460783.43602314, 2460783.436134988], [2460783.4378127092, 2460783.4380364055], [2460783.4381482536, 2460783.4382601017], [2460783.439490431, 2460783.439714127], [2460783.441615545, 2460783.4418392414], [2460783.442734026, 2460783.44306957], [2460783.454478077, 2460783.4545899252], [2460783.457386128, 2460783.4574979763], [2460783.4674524576, 2460783.467788002], [2460783.4684590907, 2460783.468570939], [2460783.4692420275, 2460783.4693538756], [2460783.4701368124, 2460783.4703605087], [2460783.472261926, 2460783.4724856224], [2460783.475841066, 2460783.476064762], [2460783.4842296736, 2460783.48445337], [2460783.4884799016, 2460783.4885917497], [2460783.4946315475, 2460783.4948552437], [2460783.4985462315, 2460783.4986580797], [2460783.5016779783, 2460783.5017898264], [2460783.510737675, 2460783.511073219], [2460783.513086485, 2460783.5134220296], [2460783.5388115495, 2460783.539035246], [2460783.541384056, 2460783.5418314487], [2460783.5449631955, 2460783.54529874], [2460783.5470883097, 2460783.547535702], [2460783.5479830946, 2460783.5480949427], [2460783.548318639, 2460783.548430487], [2460783.5503319046, 2460783.5510029932], [2460783.552904411, 2460783.5530162593], [2460783.5535755, 2460783.553687348], [2460783.553799196, 2460783.5539110443], [2460783.555029525, 2460783.556259854], [2460783.5563717023, 2460783.562187804]] freq_flags: [[46859741.2109375, 47103881.8359375], [49911499.0234375, 50033569.3359375], [59555053.7109375, 67367553.7109375], [69808959.9609375, 70663452.1484375], [73715209.9609375, 75546264.6484375], [77133178.7109375, 78720092.7734375], [80795288.0859375, 80917358.3984375], [81893920.8984375, 82015991.2109375], [85311889.6484375, 85678100.5859375], [85800170.8984375, 86044311.5234375], [87020874.0234375, 108261108.3984375], [109970092.7734375, 110092163.0859375], [112655639.6484375, 113143920.8984375], [113265991.2109375, 113388061.5234375], [113632202.1484375, 113754272.4609375], [116439819.3359375, 117172241.2109375], [124740600.5859375, 125228881.8359375], [126205444.3359375, 126327514.6484375], [127548217.7734375, 127670288.0859375], [129989624.0234375, 130111694.3359375], [136337280.2734375, 136459350.5859375], [136581420.8984375, 136703491.2109375], [136947631.8359375, 138046264.6484375], [138412475.5859375, 138534545.8984375], [138656616.2109375, 138778686.5234375], [138900756.8359375, 139022827.1484375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [142074584.9609375, 142318725.5859375], [142440795.8984375, 142562866.2109375], [142684936.5234375, 143539428.7109375], [143783569.3359375, 144027709.9609375], [145492553.7109375, 145614624.0234375], [146224975.5859375, 146347045.8984375], [147445678.7109375, 147567749.0234375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [154159545.8984375, 154403686.5234375], [157577514.6484375, 157699584.9609375], [158432006.8359375, 158798217.7734375], [159164428.7109375, 159286499.0234375], [160140991.2109375, 160385131.8359375], [169906616.2109375, 170150756.8359375], [170272827.1484375, 170394897.4609375], [170516967.7734375, 170639038.0859375], [170883178.7109375, 171005249.0234375], [174911499.0234375, 175033569.3359375], [175155639.6484375, 175643920.8984375], [180892944.3359375, 181625366.2109375], [183212280.2734375, 183334350.5859375], [183456420.8984375, 183578491.2109375], [187362670.8984375, 187606811.5234375], [189193725.5859375, 189315795.8984375], [189926147.4609375, 190048217.7734375], [190902709.9609375, 191757202.1484375], [193222045.8984375, 193832397.4609375], [193954467.7734375, 194198608.3984375], [195663452.1484375, 195785522.4609375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [198471069.3359375, 198593139.6484375], [199203491.2109375, 199325561.5234375], [201644897.4609375, 201889038.0859375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [206893920.8984375, 207626342.7734375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [212142944.3359375, 212265014.6484375], [213119506.8359375, 213241577.1484375], [215072631.8359375, 215438842.7734375], [220565795.8984375, 220809936.5234375], [221176147.4609375, 221298217.7734375], [222763061.5234375, 223739624.0234375], [223861694.3359375, 223983764.6484375], [227401733.3984375, 227767944.3359375], [229110717.7734375, 229354858.3984375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231185913.0859375], [231796264.6484375, 232040405.2734375], [232772827.1484375, 232894897.4609375], [233139038.0859375, 233505249.0234375]] ex_ants: [[3, Jee], [3, Jnn], [4, Jee], [7, Jee], [7, Jnn], [8, Jee], [8, Jnn], [10, Jee], [15, Jnn], [18, Jee], [18, Jnn], [20, Jnn], [21, Jee], [22, Jnn], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [30, Jee], [30, Jnn], [31, Jee], [32, Jnn], [33, Jnn], [34, Jee], [35, Jnn], [37, Jnn], [40, Jnn], [42, Jee], [42, Jnn], [46, Jee], [47, Jnn], [49, Jnn], [51, Jee], [54, Jee], [56, Jee], [56, Jnn], [57, Jee], [60, Jnn], [62, Jee], [64, Jnn], [65, Jee], [66, Jee], [66, Jnn], [67, Jnn], [68, Jee], [68, Jnn], [70, Jee], [70, Jnn], [71, Jee], [71, Jnn], [72, Jee], [72, Jnn], [76, Jee], [76, Jnn], [78, Jee], [80, Jnn], [81, Jnn], [82, Jnn], [86, Jee], [86, Jnn], [87, Jee], [90, Jnn], [97, Jnn], [98, Jnn], [99, Jnn], [102, Jnn], [104, Jnn], [105, Jee], [107, Jee], [109, Jnn], [113, Jnn], [115, Jee], [115, Jnn], [117, Jee], [120, Jee], [121, Jee], [130, Jee], [130, Jnn], [131, Jnn], [134, Jnn], [135, Jee], [136, Jnn], [144, Jee], [144, Jnn], [148, Jee], [149, Jee], [149, Jnn], [151, Jee], [151, Jnn], [153, Jnn], [154, Jnn], [155, Jnn], [158, Jee], [158, Jnn], [161, Jnn], [166, Jee], [166, Jnn], [167, Jnn], [170, Jee], [171, Jee], [171, Jnn], [172, Jee], [172, Jnn], [176, Jnn], [180, Jee], [180, Jnn], [184, Jee], [184, Jnn], [185, Jee], [185, Jnn], [186, Jee], [186, Jnn], [195, Jnn], [197, Jnn], [199, Jnn], [200, Jee], [200, Jnn], [202, Jnn], [206, Jnn], [209, Jnn], [210, Jee], [210, Jnn], [212, Jnn], [214, Jee], [214, Jnn], [215, Jee], [215, Jnn], [216, Jee], [218, Jnn], [227, Jee], [227, Jnn], [231, Jee], [231, Jnn], [232, Jee], [236, Jee], [236, Jnn], [238, Jnn], [239, Jee], [240, Jee], [240, Jnn], [244, Jee], [246, Jee], [250, Jee], [250, Jnn], [251, Jee], [251, Jnn], [252, Jee], [252, Jnn], [253, Jee], [253, Jnn], [254, Jee], [254, Jnn], [255, Jee], [255, Jnn], [257, Jee], [257, Jnn], [262, Jee], [262, Jnn], [266, Jee], [266, Jnn], [267, Jee], [267, Jnn], [268, Jee], [268, Jnn], [269, Jee], [269, Jnn], [271, Jee], [271, Jnn], [273, Jee], [273, Jnn], [281, Jee], [281, Jnn], [282, Jee], [282, Jnn], [283, Jee], [283, Jnn], [284, Jee], [284, Jnn], [286, Jee], [286, Jnn], [287, Jee], [295, Jee], [295, Jnn], [320, Jee], [320, Jnn], [321, Jee], [321, Jnn], [322, Jee], [322, Jnn], [323, Jee], [323, Jnn], [324, Jee], [324, Jnn], [325, Jee], [325, Jnn], [326, Jee], [326, Jnn], [327, Jee], [327, Jnn], [328, Jee], [328, Jnn], [329, Jee], [329, Jnn], [331, Jee], [331, Jnn], [332, Jee], [332, Jnn], [333, Jee], [333, Jnn], [336, Jee], [336, Jnn], [340, Jee], [340, Jnn]]
Metadata¶
In [24]:
for repo in ['hera_cal', 'hera_qm', 'hera_filters', 'hera_notebook_templates', 'pyuvdata']:
exec(f'from {repo} import __version__')
print(f'{repo}: {__version__}')
hera_cal: 3.7.1.dev45+g4a0c6f1 hera_qm: 2.2.1.dev2+ga535e9e hera_filters: 0.1.6.dev9+gf165ec1
hera_notebook_templates: 0.1.dev989+gee0995d pyuvdata: 3.1.3
In [25]:
print(f'Finished execution in {(time.time() - tstart) / 60:.2f} minutes.')
Finished execution in 31.40 minutes.