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 1845 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data1/2460705/zen.2460705.25252.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 1845 *.sum.smooth.calfits files starting with /mnt/sn1/data1/2460705/zen.2460705.25252.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.')
45.733% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 0 integrations and 16 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 1 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 10 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 1 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. 46.807% of waterfall flagged after flagging whole times and channels with median z > 1.0. 47.401% of waterfall flagged after flagging z > 4.0 outliers.
49.087% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 32 channels previously flagged 25.00% or more. Flagging 223 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. 54.280% 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
56.468% of flagging channels that are 4.0σ outliers after delay filtering the time average.
56.868% of waterfall flagged after another round of watershed flagging on z > 2.0 neighbors of prior flags.
Show results of flagging¶
Figure 3: Waterfall of Maximum z-Score of Either Polarization After Round 2 Flagging¶
The same as Figure 1, but after the flagging performed in this notebook.
In [16]:
plot_max_z_score(zscore, flags=flags)
All-NaN axis encountered
In [17]:
def zscore_spectra(ylim=[-3, 3], flags=flags):
fig, axes = plt.subplots(2, 1, figsize=(14,6), dpi=100, sharex=True, sharey=True, gridspec_kw={'hspace': 0})
for ax, pol in zip(axes, ['ee', 'nn']):
ax.plot(freqs / 1e6, np.nanmean(zscore[pol], axis=0),'r', label=f'{pol}-Polarization Before Round 2 Flagging', lw=.5)
ax.plot(freqs / 1e6, np.nanmean(np.where(flags, np.nan, zscore[pol]), axis=0), label=f'{pol}-Polarization After Round 2 Flagging')
ax.legend(loc='lower right')
ax.set_ylabel('Time-Averged Z-Score\n(Excluding Flags)')
ax.set_ylim(ylim)
axes[1].set_xlabel('Frequency (MHz)')
plt.tight_layout()
Figure 4: Spectra of Time-Averaged z-Scores¶
The average along the time axis of Figures 1 and 3 (though now separated per-polarization). This plot is useful for showing channels with repeated low-level RFI.
In [18]:
zscore_spectra()
Mean of empty slice Mean of empty slice
In [19]:
def summarize_flagging(flags=flags):
plt.figure(figsize=(14,10), dpi=100)
cmap = matplotlib.colors.ListedColormap(((0, 0, 0),) + matplotlib.cm.get_cmap("Set2").colors[0:2])
plt.imshow(np.where(np.any(~np.isfinite(list(zscore.values())), axis=0), 1, np.where(flags, 2, 0)),
aspect='auto', cmap=cmap, interpolation='none', extent=extent)
plt.clim([-.5, 2.5])
cbar = plt.colorbar(location='top', aspect=40, pad=.02)
cbar.set_ticks([0, 1, 2])
cbar.set_ticklabels(['Unflagged', 'Previously Flagged', 'Flagged Here Using Delayed Filtered z-Scores'])
plt.xlabel('Frequency (MHz)')
plt.ylabel(f'JD - {int(times[0])}')
plt.tight_layout()
Figure 5: Summary of Flags Before and After Round 2 Flagging¶
This plot shows which times and frequencies were flagged before and after this notebook. It is directly comparable to Figure 5 of the first round full_day_rfi notebook.
In [20]:
summarize_flagging()
The get_cmap function was deprecated in Matplotlib 3.7 and will be removed two minor releases later. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap(obj)`` instead.
Save results¶
In [21]:
add_to_history = 'by full_day_rfi_round_2 notebook with the following environment:\n' + '=' * 65 + '\n' + os.popen('conda env export').read() + '=' * 65
In [22]:
tind = 0
always_flagged_ants = set()
ever_unflagged_ants = set()
for cal_file in cal_files:
with warnings.catch_warnings():
warnings.simplefilter("ignore")
# update cal_file
uvc = UVCal()
uvc.read(cal_file, use_future_array_shapes=True)
uvc.flag_array |= (flags[tind:tind + len(uvc.time_array), :].T)[None, :, :, None]
uvc.history += 'Modified ' + add_to_history
uvc.write_calfits(cal_file, clobber=True)
# keep track of flagged antennas
for antnum in uvc.ant_array:
for antpol in ['Jee', 'Jnn']:
if np.all(uvc.get_flags(antnum, antpol)):
if (antnum, antpol) not in ever_unflagged_ants:
always_flagged_ants.add((antnum, antpol))
else:
ever_unflagged_ants.add((antnum, antpol))
always_flagged_ants.discard((antnum, antpol))
# Create new flag object
uvf_out = UVFlag(uvc, waterfall=True, mode='flag')
uvf_out.flag_array |= flags[tind:tind + len(uvc.time_array), :, None]
uvf_out.history += 'Produced ' + add_to_history
uvf_out.write(cal_file.replace(SMOOTH_CAL_SUFFIX, FLAG_WATERFALL2_SUFFIX), clobber=True)
# increment time index
tind += len(uvc.time_array)
print(f'Saved {len(cal_files)} *.{FLAG_WATERFALL2_SUFFIX} files starting with {cal_files[0].replace(SMOOTH_CAL_SUFFIX, FLAG_WATERFALL2_SUFFIX)}.')
Saved 1845 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data1/2460705/zen.2460705.25252.sum.flag_waterfall_round_2.h5.
In [23]:
# write summary of entirely flagged times/freqs/ants to yaml
all_flagged_times = np.all(flags, axis=1)
all_flagged_freqs = np.all(flags, axis=0)
all_flagged_ants = sorted(always_flagged_ants)
dt = np.median(np.diff(times))
out_yml_str = 'JD_flags: ' + str([[times[flag_stretch][0] - dt / 2, times[flag_stretch][-1] + dt / 2]
for flag_stretch in true_stretches(all_flagged_times)])
df = np.median(np.diff(freqs))
out_yml_str += '\n\nfreq_flags: ' + str([[freqs[flag_stretch][0] - df / 2, freqs[flag_stretch][-1] + df / 2]
for flag_stretch in true_stretches(all_flagged_freqs)])
out_yml_str += '\n\nex_ants: ' + str(all_flagged_ants).replace("'", "").replace('(', '[').replace(')', ']')
print(f'Writing the following to {out_yaml_file}\n' + '-' * (25 + len(out_yaml_file)))
print(out_yml_str)
with open(out_yaml_file, 'w') as outfile:
outfile.writelines(out_yml_str)
Writing the following to /mnt/sn1/data1/2460705/2460705_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460705.252404121, 2460705.3763318234], [2460705.3764436715, 2460705.3775621527], [2460705.378345089, 2460705.3787924815], [2460705.379128026, 2460705.379351722], [2460705.37946357, 2460705.3795754183], [2460705.380134659, 2460705.380358355], [2460705.3805820514, 2460705.381364988], [2460705.3814768363, 2460705.3815886844], [2460705.3818123806, 2460705.382371621], [2460705.3828190137, 2460705.382930862], [2460705.383266406, 2460705.3833782542], [2460705.383713798, 2460705.3841611906], [2460705.384720431, 2460705.3848322793], [2460705.3851678236, 2460705.385503368], [2460705.385727064, 2460705.3859507605], [2460705.3860626086, 2460705.3861744567], [2460705.386510001, 2460705.386621849], [2460705.3867336973, 2460705.3869573935], [2460705.3870692416, 2460705.3871810897], [2460705.387516634, 2460705.387628482], [2460705.3877403303, 2460705.3878521784], [2460705.388523267, 2460705.3887469633], [2460705.3890825077, 2460705.389306204], [2460705.3894180516, 2460705.3895298997], [2460705.389753596, 2460705.3902009884], [2460705.390872077, 2460705.3913194695], [2460705.391543166, 2460705.3921024064], [2460705.3922142545, 2460705.3924379507], [2460705.392885343, 2460705.3931090394], [2460705.3933327356, 2460705.3941156724], [2460705.3948986093, 2460705.395122305], [2460705.3953460013, 2460705.3956815456], [2460705.39601709, 2460705.396128938], [2460705.3963526343, 2460705.3964644824], [2460705.3966881786, 2460705.397023723], [2460705.397135571, 2460705.3975829636], [2460705.398030356, 2460705.3982540523], [2460705.3984777485, 2460705.3985895966], [2460705.3992606853, 2460705.3995962297], [2460705.399819926, 2460705.4001554702], [2460705.4008265585, 2460705.4010502547], [2460705.401162103, 2460705.4017213434], [2460705.4018331915, 2460705.402280584], [2460705.40250428, 2460705.4026161283], [2460705.4027279764, 2460705.4028398246], [2460705.403063521, 2460705.403287217], [2460705.4038464576, 2460705.404182002], [2460705.4049649388, 2460705.405300483], [2460705.4058597237, 2460705.406530812], [2460705.4068663563, 2460705.4072019006], [2460705.4073137487, 2460705.407649293], [2460705.407761141, 2460705.4079848374], [2460705.4102217997, 2460705.410557344], [2460705.4107810403, 2460705.411452129], [2460705.4124587616, 2460705.4131298503], [2460705.413800939, 2460705.414583876], [2460705.4150312683, 2460705.4153668126], [2460705.4154786607, 2460705.415814205], [2460705.415926053, 2460705.4162615975], [2460705.416485294, 2460705.416820838], [2460705.4171563825, 2460705.4176037745], [2460705.417939319, 2460705.418274863], [2460705.4187222556, 2460705.419169648], [2460705.4197288887, 2460705.419952585], [2460705.420064433, 2460705.4203999774], [2460705.4214066104, 2460705.4217421548], [2460705.421854003, 2460705.422189547], [2460705.4236435723, 2460705.424314661], [2460705.424538357, 2460705.425209446], [2460705.425433142, 2460705.4257686865], [2460705.426327927, 2460705.4269990157], [2460705.427446408, 2460705.4276701044], [2460705.4277819525, 2460705.428117497], [2460705.4286767375, 2460705.4287885856], [2460705.4289004332, 2460705.4292359776], [2460705.429459674, 2460705.4299070663], [2460705.4302426106, 2460705.430578155], [2460705.4311373956, 2460705.431696636], [2460705.432591421, 2460705.4328151173], [2460705.4329269654, 2460705.433486206], [2460705.4340454466, 2460705.4365061047], [2460705.436617953, 2460705.436953497], [2460705.4370653452, 2460705.4371771934], [2460705.4372890415, 2460705.4382956745], [2460705.4384075226, 2460705.438743067], [2460705.4391904594, 2460705.439861548], [2460705.440085244, 2460705.4404207882], [2460705.441315573, 2460705.4416511175], [2460705.4424340543, 2460705.4427695987], [2460705.443105143, 2460705.4433288393], [2460705.4443354723, 2460705.4444473204], [2460705.4445591685, 2460705.444894713], [2460705.4626785615, 2460705.4629022577], [2460705.4677117267, 2460705.468047271], [2460705.4701723848, 2460705.470284233], [2460705.4717382584, 2460705.4718501065], [2460705.472409347, 2460705.4727448914], [2460705.4728567395, 2460705.4729685877], [2460705.473192284, 2460705.47341598], [2460705.474534461, 2460705.4749818533], [2460705.4822519803, 2460705.4823638285], [2460705.4825875247, 2460705.482699373], [2460705.498917348, 2460705.499588437], [2460705.503167576, 2460705.5033912724], [2460705.5054045385, 2460705.5056282347], [2460705.5087599815, 2460705.5089836777], [2460705.5149116274, 2460705.5150234755], [2460705.520839577, 2460705.520951425], [2460705.521622514, 2460705.521734362], [2460705.5254253494, 2460705.5257608937], [2460705.529004489, 2460705.5292281853], [2460705.5293400334, 2460705.5294518815], [2460705.535156135, 2460705.535267983], [2460705.536386464, 2460705.536498312], [2460705.537393097, 2460705.537728641], [2460705.5380641855, 2460705.53839973], [2460705.540412996, 2460705.5408603884], [2460705.5449987683, 2460705.5451106164], [2460705.54667649, 2460705.547012034], [2460705.550143781, 2460705.55081487], [2460705.550926718, 2460705.5512622623], [2460705.5514859585, 2460705.552045199], [2460705.5522688953, 2460705.5526044397], [2460705.5540584647, 2460705.554505857], [2460705.5550650978, 2460705.555176946], [2460705.5557361864, 2460705.556071731], [2460705.5663617565, 2460705.566809149], [2460705.5698290477, 2460705.570164592], [2460705.5779939597, 2460705.578217656], [2460705.5813494027, 2460705.581461251], [2460705.5819086432, 2460705.5820204914], [2460705.592422365, 2460705.5925342133], [2460705.5966725936, 2460705.597231834], [2460705.603830872, 2460705.604501961], [2460705.606515227, 2460705.606627075], [2460705.6072981637, 2460705.60752186], [2460705.608528493, 2460705.608640341], [2460705.6106536067, 2460705.610877303], [2460705.611100999, 2460705.6113246954], [2460705.6237398353, 2460705.6238516835], [2460705.624410924, 2460705.6248583165], [2460705.6249701646, 2460705.6250820127], [2460705.6449909755, 2460705.6452146717], [2460705.647227938, 2460705.647451634], [2460705.6484582666, 2460705.648681963], [2460705.6494648997, 2460705.649688596], [2460705.650695229, 2460705.650807077], [2460705.6635577613, 2460705.6636696095]] freq_flags: [[49911499.0234375, 50155639.6484375], [51376342.7734375, 51620483.3984375], [61874389.6484375, 61996459.9609375], [62240600.5859375, 62728881.8359375], [69931030.2734375, 70053100.5859375], [87387084.9609375, 108016967.7734375], [109725952.1484375, 110336303.7109375], [112289428.7109375, 113021850.5859375], [113265991.2109375, 113388061.5234375], [113632202.1484375, 113754272.4609375], [116073608.3984375, 116195678.7109375], [116439819.3359375, 116561889.6484375], [116683959.9609375, 116806030.2734375], [124618530.2734375, 125228881.8359375], [127548217.7734375, 127670288.0859375], [129989624.0234375, 130233764.6484375], [136337280.2734375, 136459350.5859375], [136825561.5234375, 138290405.2734375], [138656616.2109375, 138778686.5234375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [142074584.9609375, 142318725.5859375], [142440795.8984375, 143661499.0234375], [143783569.3359375, 144027709.9609375], [145492553.7109375, 145614624.0234375], [145858764.6484375, 145980834.9609375], [147445678.7109375, 147567749.0234375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [153427124.0234375, 153793334.9609375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155380249.0234375], [155868530.2734375, 155990600.5859375], [156967163.0859375, 157089233.3984375], [157211303.7109375, 157333374.0234375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158432006.8359375], [158676147.4609375, 158798217.7734375], [159164428.7109375, 159286499.0234375], [159530639.6484375, 159652709.9609375], [160140991.2109375, 160385131.8359375], [168807983.3984375, 168930053.7109375], [169906616.2109375, 170639038.0859375], [170883178.7109375, 171005249.0234375], [171249389.6484375, 171615600.5859375], [171737670.8984375, 171859741.2109375], [174789428.7109375, 175033569.3359375], [175155639.6484375, 175399780.2734375], [175521850.5859375, 175643920.8984375], [181137084.9609375, 181381225.5859375], [181503295.8984375, 181625366.2109375], [183212280.2734375, 183334350.5859375], [186386108.3984375, 186508178.7109375], [187362670.8984375, 187606811.5234375], [189926147.4609375, 190048217.7734375], [190658569.3359375, 190902709.9609375], [191024780.2734375, 191757202.1484375], [193222045.8984375, 193344116.2109375], [195419311.5234375, 195541381.8359375], [195663452.1484375, 195785522.4609375], [195907592.7734375, 196029663.0859375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [199203491.2109375, 199325561.5234375], [200057983.3984375, 200180053.7109375], [201766967.7734375, 201889038.0859375], [203231811.5234375, 203353881.8359375], [204696655.2734375, 204818725.5859375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [207138061.5234375, 207260131.8359375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [211898803.7109375, 212020874.0234375], [212142944.3359375, 212265014.6484375], [212387084.9609375, 212509155.2734375], [215194702.1484375, 215316772.4609375], [220565795.8984375, 220809936.5234375], [221176147.4609375, 221298217.7734375], [222763061.5234375, 223739624.0234375], [225692749.0234375, 225814819.3359375], [227279663.0859375, 228012084.9609375], [229110717.7734375, 229354858.3984375], [229721069.3359375, 229843139.6484375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231307983.3984375], [232162475.5859375, 234359741.2109375]] ex_ants: [[7, Jee], [7, Jnn], [8, Jee], [8, Jnn], [9, Jee], [9, Jnn], [10, Jee], [10, Jnn], [15, Jnn], [16, Jee], [18, Jnn], [19, Jee], [19, Jnn], [20, Jee], [20, Jnn], [21, Jee], [21, Jnn], [22, Jee], [22, Jnn], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [31, Jee], [31, Jnn], [32, Jee], [32, Jnn], [33, Jee], [33, Jnn], [34, Jee], [34, Jnn], [35, Jee], [35, Jnn], [37, Jee], [37, Jnn], [40, Jee], [40, Jnn], [41, Jee], [41, Jnn], [42, Jee], [42, Jnn], [45, Jee], [46, Jee], [47, Jee], [47, Jnn], [48, Jee], [48, Jnn], [49, Jee], [49, Jnn], [51, Jee], [51, Jnn], [54, Jee], [54, Jnn], [55, Jee], [55, Jnn], [56, Jee], [56, Jnn], [57, Jee], [57, Jnn], [61, Jee], [61, Jnn], [62, Jee], [62, Jnn], [63, Jee], [63, Jnn], [64, Jee], [64, Jnn], [68, Jee], [68, Jnn], [69, Jee], [69, Jnn], [70, Jee], [70, Jnn], [71, Jee], [71, Jnn], [72, Jee], [72, Jnn], [73, Jee], [77, Jee], [77, Jnn], [78, Jee], [78, Jnn], [80, Jnn], [81, Jee], [83, Jnn], [86, Jee], [87, Jee], [87, Jnn], [88, Jee], [88, Jnn], [90, Jee], [90, Jnn], [92, Jee], [93, Jee], [95, Jee], [97, Jnn], [98, Jee], [98, Jnn], [99, Jee], [100, Jnn], [104, Jee], [104, Jnn], [107, Jee], [107, Jnn], [108, Jnn], [109, Jnn], [114, Jee], [114, Jnn], [116, Jee], [120, Jee], [120, Jnn], [121, Jee], [121, Jnn], [126, Jee], [127, Jee], [127, Jnn], [130, Jnn], [131, Jnn], [133, Jee], [135, Jee], [137, Jee], [137, Jnn], [142, Jnn], [147, Jee], [147, Jnn], [148, Jee], [148, Jnn], [149, Jee], [149, Jnn], [155, Jee], [161, Jee], [161, Jnn], [170, Jee], [171, Jnn], [179, Jee], [179, Jnn], [180, Jnn], [182, Jee], [182, Jnn], [184, Jee], [187, Jee], [188, Jnn], [198, Jnn], [199, Jee], [199, Jnn], [200, Jee], [200, Jnn], [201, Jnn], [202, Jnn], [208, Jee], [212, Jnn], [213, Jee], [213, Jnn], [215, Jnn], [216, Jnn], [218, Jee], [218, Jnn], [221, Jee], [232, Jee], [238, Jnn], [239, Jee], [240, Jee], [240, Jnn], [241, Jee], [241, Jnn], [242, Jee], [242, Jnn], [243, Jee], [243, Jnn], [245, Jnn], [246, Jee], [250, Jee], [251, Jee], [253, Jnn], [255, Jnn], [262, Jee], [262, Jnn], [320, Jee], [320, Jnn], [321, Jee], [321, Jnn], [323, Jee], [323, Jnn], [324, Jee], [324, Jnn], [325, Jee], [325, Jnn], [326, Jee], [326, Jnn], [327, Jee], [327, Jnn], [328, Jee], [328, Jnn], [329, Jee], [329, Jnn], [331, Jee], [331, Jnn], [332, Jee], [332, Jnn], [333, Jee], [333, Jnn], [336, Jee], [336, Jnn], [340, Jee], [340, Jnn]]
Metadata¶
In [24]:
for repo in ['hera_cal', 'hera_qm', 'hera_filters', 'hera_notebook_templates', 'pyuvdata']:
exec(f'from {repo} import __version__')
print(f'{repo}: {__version__}')
hera_cal: 3.6.2.dev110+g0529798 hera_qm: 2.2.0 hera_filters: 0.1.6.dev1+g297dcce
hera_notebook_templates: 0.1.dev936+gdc93cad pyuvdata: 3.0.1.dev70+g283dda3
In [25]:
print(f'Finished execution in {(time.time() - tstart) / 60:.2f} minutes.')
Finished execution in 149.18 minutes.