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 1850 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data1/2460627/zen.2460627.25257.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 1850 *.sum.smooth.calfits files starting with /mnt/sn1/data1/2460627/zen.2460627.25257.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.')
25.843% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 0 integrations and 7 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 1 integrations and 5 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. 26.522% of waterfall flagged after flagging whole times and channels with median z > 1.0. 27.278% of waterfall flagged after flagging z > 4.0 outliers.
29.580% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 24 channels previously flagged 25.00% or more.
Mean of empty slice Mean of empty slice
Flagging 448 times previously flagged 10.00% or more. Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more.
Flagging 0 times previously flagged 10.00% or more. Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more.
Flagging 0 times previously flagged 10.00% or more. Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
38.989% 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
41.907% of flagging channels that are 4.0σ outliers after delay filtering the time average.
42.393% 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 1850 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data1/2460627/zen.2460627.25257.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/data1/2460627/2460627_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460627.254587232, 2460627.2548109284], [2460627.2593967007, 2460627.259732245], [2460627.260515182, 2460627.260738878], [2460627.261521815, 2460627.2620810554], [2460627.2631995366, 2460627.263423233], [2460627.268791942, 2460627.2694630306], [2460627.2724829293, 2460627.2725947774], [2460627.2727066255, 2460627.2728184736], [2460627.274943588, 2460627.275614676], [2460627.2759502204, 2460627.2760620685], [2460627.2761739166, 2460627.2762857648], [2460627.2769568535, 2460627.277292398], [2460627.284786221, 2460627.284898069], [2460627.285681006, 2460627.2859047023], [2460627.287135031, 2460627.2874705754], [2460627.289819386, 2460627.289931234], [2460627.2903786264, 2460627.2904904746], [2460627.2906023227, 2460627.290826019], [2460627.294405158, 2460627.2946288544], [2460627.29608288, 2460627.296194728], [2460627.296306576, 2460627.2967539686], [2460627.2968658167, 2460627.296977665], [2460627.2978724497, 2460627.2982079936], [2460627.3010041965, 2460627.3012278927], [2460627.301339741, 2460627.301451589], [2460627.303912247, 2460627.304024095], [2460627.316439235, 2460627.3166629313], [2460627.3167747795, 2460627.3168866276], [2460627.31733402, 2460627.317445868], [2460627.31912359, 2460627.3196828305], [2460627.3239330584, 2460627.3240449065], [2460627.3259463245, 2460627.3260581726], [2460627.330755793, 2460627.3309794893], [2460627.3325453624, 2460627.332992755], [2460627.3334401473, 2460627.3336638436], [2460627.3343349323, 2460627.3344467804], [2460627.336124502, 2460627.3364600465], [2460627.336795591, 2460627.336907439], [2460627.337131135, 2460627.3372429833], [2460627.344401262, 2460627.345407895], [2460627.3471974647, 2460627.347309313], [2460627.3512239964, 2460627.3513358445], [2460627.353908351, 2460627.3540201993], [2460627.355250528, 2460627.355362376], [2460627.3574874904, 2460627.3577111866], [2460627.357934883, 2460627.358270427], [2460627.360283693, 2460627.360507389], [2460627.362408807, 2460627.362520655], [2460627.3646457694, 2460627.3647576175], [2460627.365876098, 2460627.3660997944], [2460627.383548099, 2460627.3837717953], [2460627.3857850614, 2460627.3860087576], [2460627.3920485554, 2460627.3921604035], [2460627.397864657, 2460627.397976505], [2460627.398759442, 2460627.39887129], [2460627.402226733, 2460627.402338581], [2460627.4061414166, 2460627.4062532648], [2460627.406365113, 2460627.406588809], [2460627.4249318987, 2460627.425043747], [2460627.429517671, 2460627.429629519], [2460627.4341034433, 2460627.4343271395], [2460627.4354456207, 2460627.435557469], [2460627.435781165, 2460627.4361167094], [2460627.4407024817, 2460627.440926178], [2460627.4433868364, 2460627.4434986846], [2460627.4500977225, 2460627.4503214187], [2460627.450433267, 2460627.450545115], [2460627.4519991404, 2460627.452670229], [2460627.4547953433, 2460627.4550190396], [2460627.455801976, 2460627.456025672], [2460627.4574796977, 2460627.457703394], [2460627.4591574194, 2460627.4592692675], [2460627.4678815715, 2460627.4681052677], [2460627.46855266, 2460627.4687763564], [2460627.470454078, 2460627.4706777744], [2460627.471125167, 2460627.471237015], [2460627.47269104, 2460627.4729147363], [2460627.4948369656, 2460627.4949488137], [2460627.495619902, 2460627.49573175], [2460627.4989753454, 2460627.4991990416], [2460627.501771548, 2460627.501995244], [2460627.5026663328, 2460627.502890029], [2460627.5041203583, 2460627.5042322064], [2460627.505574384, 2460627.50579808], [2460627.505909928, 2460627.5060217762], [2460627.5068047126, 2460627.507028409], [2460627.5079231937, 2460627.508035042], [2460627.509041675, 2460627.509265371], [2460627.5106075485, 2460627.5108312448], [2460627.5117260297, 2460627.511837878], [2460627.519890941, 2460627.52056203], [2460627.5211212705, 2460627.5212331186], [2460627.5233582323, 2460627.5235819286], [2460627.523805625, 2460627.524029321], [2460627.525818891, 2460627.525930739], [2460627.526825524, 2460627.526937372], [2460627.535437828, 2460627.5357733723], [2460627.537339246, 2460627.5381221827], [2460627.538457727, 2460627.5387932714], [2460627.5389051195, 2460627.5391288158], [2460627.539576208, 2460627.5402472964], [2460627.541701322, 2460627.5421487144], [2460627.550761019, 2460627.551320259], [2460627.5517676515, 2460627.552215044], [2460627.5571363606, 2460627.5572482087], [2460627.5582548417, 2460627.558590386], [2460627.5625050697, 2460627.562840614], [2460627.569104108, 2460627.569215956], [2460627.5714529185, 2460627.5715647666], [2460627.5724595515, 2460627.572906944], [2460627.5759268426, 2460627.576374235], [2460627.576486083, 2460627.5765979313], [2460627.5768216276, 2460627.577045324], [2460627.57726902, 2460627.5777164125], [2460627.578163805, 2460627.5786111974], [2460627.5800652225, 2460627.580512615], [2460627.5811837036, 2460627.581631096], [2460627.582637729, 2460627.583644362], [2460627.584986539, 2460627.585098387], [2460627.5853220834, 2460627.5854339316], [2460627.5865524127, 2460627.586776109], [2460627.5875590458, 2460627.587670894], [2460627.5902434005, 2460627.5903552487], [2460627.591250033, 2460627.5913618812], [2460627.592256666, 2460627.592927755], [2460627.5957239578, 2460627.595835806], [2460627.595947654, 2460627.5967305903], [2460627.5968424384, 2460627.5969542866], [2460627.597289831, 2460627.598408312], [2460627.59852016, 2460627.5986320083], [2460627.599191249, 2460627.599526793], [2460627.5998623376, 2460627.6006452744], [2460627.6009808187, 2460627.601092667], [2460627.601204515, 2460627.6016519074], [2460627.6017637555, 2460627.602434844], [2460627.602546692, 2460627.6029940844], [2460627.6031059325, 2460627.603777021], [2460627.6040007174, 2460627.623238592], [2460627.62335044, 2460627.624580769], [2460627.624692617, 2460627.6300613265], [2460627.630173174, 2460627.6311798072], [2460627.6312916554, 2460627.631962744], [2460627.632074592, 2460627.632745681], [2460627.6331930733, 2460627.6333049214], [2460627.633752314, 2460627.633864162], [2460627.63397601, 2460627.6340878583], [2460627.6344234026, 2460627.634647099], [2460627.634758947, 2460627.634982643], [2460627.6358774276, 2460627.636101124], [2460627.63632482, 2460627.6367722126], [2460627.637331453, 2460627.6375551494], [2460627.6377788456, 2460627.6378906937], [2460627.6384499343, 2460627.6385617824], [2460627.6386736305, 2460627.6394565674], [2460627.6397921117, 2460627.640910593], [2460627.6412461367, 2460627.6421409217], [2460627.64225277, 2460627.642364618], [2460627.642476466, 2460627.6431475547], [2460627.643594947, 2460627.644266036], [2460627.6449371246, 2460627.6458319095], [2460627.646279302, 2460627.646502998], [2460627.6472859345, 2460627.6473977827], [2460627.647845175, 2460627.6479570232], [2460627.6482925676, 2460627.666300113]] freq_flags: [[49911499.0234375, 50155639.6484375], [62240600.5859375, 63339233.3984375], [66146850.5859375, 66757202.1484375], [69931030.2734375, 70053100.5859375], [87387084.9609375, 108016967.7734375], [109603881.8359375, 109725952.1484375], [109970092.7734375, 110092163.0859375], [112167358.3984375, 113754272.4609375], [115585327.1484375, 115707397.4609375], [116073608.3984375, 116195678.7109375], [116439819.3359375, 116561889.6484375], [121322631.8359375, 121810913.0859375], [124618530.2734375, 125473022.4609375], [127548217.7734375, 127670288.0859375], [128280639.6484375, 128402709.9609375], [128524780.2734375, 128646850.5859375], [129989624.0234375, 130111694.3359375], [131210327.1484375, 131332397.4609375], [131454467.7734375, 131576538.0859375], [132308959.9609375, 132431030.2734375], [132553100.5859375, 132675170.8984375], [133407592.7734375, 133529663.0859375], [133895874.0234375, 134017944.3359375], [134994506.8359375, 135360717.7734375], [135604858.3984375, 135726928.7109375], [136215209.9609375, 136459350.5859375], [136581420.8984375, 138290405.2734375], [138656616.2109375, 138778686.5234375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [142074584.9609375, 142318725.5859375], [142684936.5234375, 143661499.0234375], [143783569.3359375, 144027709.9609375], [144882202.1484375, 145004272.4609375], [145858764.6484375, 145980834.9609375], [146835327.1484375, 147567749.0234375], [148300170.8984375, 148544311.5234375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [153915405.2734375, 154037475.5859375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [155258178.7109375, 155380249.0234375], [155868530.2734375, 155990600.5859375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158309936.5234375], [159164428.7109375, 159286499.0234375], [160140991.2109375, 160385131.8359375], [169174194.3359375, 169296264.6484375], [169906616.2109375, 170150756.8359375], [170272827.1484375, 170394897.4609375], [170516967.7734375, 170639038.0859375], [170883178.7109375, 171005249.0234375], [171249389.6484375, 171371459.9609375], [175155639.6484375, 175399780.2734375], [181137084.9609375, 181259155.2734375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [189193725.5859375, 189315795.8984375], [189926147.4609375, 190048217.7734375], [190780639.6484375, 190902709.9609375], [191024780.2734375, 191513061.5234375], [191635131.8359375, 191879272.4609375], [192123413.0859375, 192245483.3984375], [192489624.0234375, 192611694.3359375], [195297241.2109375, 195541381.8359375], [195663452.1484375, 196029663.0859375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [199203491.2109375, 199325561.5234375], [201644897.4609375, 201889038.0859375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [207138061.5234375, 207382202.1484375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [211776733.3984375, 212631225.5859375], [215194702.1484375, 215316772.4609375], [220565795.8984375, 220809936.5234375], [220932006.8359375, 221054077.1484375], [221176147.4609375, 221298217.7734375], [222885131.8359375, 223739624.0234375], [227401733.3984375, 227523803.7109375], [227645874.0234375, 227890014.6484375], [229110717.7734375, 229354858.3984375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231185913.0859375], [231307983.3984375, 231430053.7109375], [233993530.2734375, 234359741.2109375]] ex_ants: [[7, Jee], [8, Jee], [9, Jee], [10, Jee], [15, Jnn], [16, Jee], [17, Jnn], [18, Jee], [18, Jnn], [21, Jee], [22, Jee], [22, Jnn], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [29, Jnn], [31, Jnn], [32, Jnn], [33, Jnn], [34, Jee], [34, Jnn], [35, Jee], [35, Jnn], [36, Jee], [37, Jnn], [40, Jnn], [42, Jnn], [45, Jee], [46, Jee], [47, Jee], [47, Jnn], [48, Jee], [48, Jnn], [49, Jee], [49, Jnn], [51, Jee], [54, Jnn], [55, Jee], [61, Jee], [61, Jnn], [62, Jee], [62, Jnn], [63, Jee], [63, Jnn], [64, Jee], [64, Jnn], [69, Jee], [72, Jnn], [73, Jee], [77, Jee], [77, Jnn], [78, Jee], [78, Jnn], [81, Jee], [85, Jnn], [86, Jee], [86, Jnn], [87, Jee], [88, Jee], [88, Jnn], [89, Jee], [90, Jee], [90, Jnn], [92, Jee], [93, Jee], [95, Jee], [97, Jnn], [98, Jnn], [100, Jnn], [103, Jnn], [104, Jnn], [107, Jee], [107, Jnn], [108, Jnn], [109, Jnn], [117, Jee], [120, Jee], [120, Jnn], [121, Jee], [121, Jnn], [125, Jee], [127, Jee], [130, Jee], [130, Jnn], [131, Jnn], [134, Jee], [136, Jee], [136, Jnn], [137, Jee], [142, Jnn], [147, Jnn], [161, Jnn], [164, Jee], [170, Jee], [171, Jnn], [176, Jee], [176, Jnn], [177, Jee], [177, Jnn], [178, Jee], [178, Jnn], [179, Jee], [179, Jnn], [180, Jee], [180, Jnn], [182, Jee], [182, Jnn], [183, Jee], [184, Jee], [188, Jnn], [191, Jee], [193, Jee], [197, Jnn], [198, Jnn], [199, Jee], [199, Jnn], [200, Jee], [200, Jnn], [201, Jnn], [202, Jnn], [204, Jee], [206, Jee], [208, Jee], [209, Jnn], [212, Jnn], [213, Jee], [213, Jnn], [215, Jee], [215, Jnn], [216, Jee], [216, Jnn], [218, Jnn], [231, Jee], [232, Jee], [234, Jnn], [235, Jee], [235, Jnn], [240, Jee], [240, Jnn], [241, Jee], [241, Jnn], [242, Jee], [242, Jnn], [243, Jee], [243, Jnn], [245, Jnn], [246, Jee], [246, Jnn], [250, Jee], [251, Jee], [253, Jnn], [255, Jnn], [256, Jee], [256, Jnn], [261, Jee], [261, Jnn], [262, Jee], [262, Jnn], [268, Jee], [268, Jnn], [270, Jee], [270, Jnn], [272, Jee], [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 46.16 minutes.