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 1567 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data1/2460775/zen.2460775.21085.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 1567 *.sum.smooth.calfits files starting with /mnt/sn1/data1/2460775/zen.2460775.21085.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.')
24.779% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 0 integrations and 15 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 2 times previously flagged 10.00% or more.
Flagging an additional 1 integrations and 8 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 0 integrations and 0 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more. 26.093% of waterfall flagged after flagging whole times and channels with median z > 1.0. 26.872% of waterfall flagged after flagging z > 4.0 outliers.
28.534% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 26 channels previously flagged 25.00% or more. Flagging 183 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. 33.669% 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
37.136% of flagging channels that are 4.0σ outliers after delay filtering the time average.
37.506% 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 1567 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data1/2460775/zen.2460775.21085.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/data1/2460775/2460775_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460775.2162181144, 2460775.2163299625], [2460775.2194617093, 2460775.2208038867], [2460775.224047482, 2460775.2246067226], [2460775.226396292, 2460775.22650814], [2460775.2267318363, 2460775.227179229], [2460775.2276266213, 2460775.2279621656], [2460775.228521406, 2460775.2286332543], [2460775.230199128, 2460775.230310976], [2460775.230422824, 2460775.230870216], [2460775.2310939124, 2460775.2313176086], [2460775.23243609, 2460775.232659786], [2460775.2340019634, 2460775.2342256596], [2460775.2343375077, 2460775.234673052], [2460775.236127077, 2460775.2364626215], [2460775.2384758876, 2460775.238699584], [2460775.238811432, 2460775.2392588244], [2460775.240824698, 2460775.240936546], [2460775.244068293, 2460775.2444038372], [2460775.2447393816, 2460775.2448512297], [2460775.2467526477, 2460775.24720004], [2460775.2491014577, 2460775.249437002], [2460775.252345053, 2460775.2525687492], [2460775.263865408, 2460775.2640891043], [2460775.2648720406, 2460775.265319433], [2460775.2674445473, 2460775.2676682435], [2460775.269234117, 2460775.2693459652], [2460775.2696815096, 2460775.269905206], [2460775.270688142, 2460775.2711355346], [2460775.27672794, 2460775.276951636], [2460775.2771753324, 2460775.277734573], [2460775.277846421, 2460775.277958269], [2460775.2823203453, 2460775.2824321934], [2460775.291380042, 2460775.2917155866], [2460775.2930577635, 2460775.2931696116], [2460775.2932814597, 2460775.293505156], [2460775.2938407003, 2460775.2941762446], [2460775.294623637, 2460775.2948473333], [2460775.2976435362, 2460775.2977553844], [2460775.29797908, 2460775.2983146245], [2460775.3003278906, 2460775.3004397387], [2460775.3036833336, 2460775.30390703], [2460775.3067032327, 2460775.306926929], [2460775.307038777, 2460775.3072624733], [2460775.3091638912, 2460775.3092757394], [2460775.309387587, 2460775.309499435], [2460775.3096112832, 2460775.3097231314], [2460775.3098349795, 2460775.3100586757], [2460775.314085208, 2460775.314197056], [2460775.319565765, 2460775.319677613], [2460775.3199013094, 2460775.3201250057], [2460775.320796094, 2460775.320907942], [2460775.3222501194, 2460775.3225856638], [2460775.32280936, 2460775.3231449043], [2460775.3233686006, 2460775.3234804487], [2460775.3241515374, 2460775.32459893], [2460775.326164803, 2460775.326276651], [2460775.3268358917, 2460775.32694774], [2460775.3285136134, 2460775.3286254616], [2460775.328849158, 2460775.329072854], [2460775.3296320946, 2460775.3297439427], [2460775.329855791, 2460775.330191335], [2460775.3318690564, 2460775.3319809046], [2460775.332428297, 2460775.3327638414], [2460775.3329875376, 2460775.33343493], [2460775.3336586263, 2460775.3341060188], [2460775.335336348, 2460775.336231133], [2460775.3366785254, 2460775.33757331], [2460775.3379088542, 2460775.3381325505], [2460775.3382443986, 2460775.338468095], [2460775.3399221203, 2460775.340481361], [2460775.3412642977, 2460775.341487994], [2460775.3432775633, 2460775.3433894115], [2460775.3446197407, 2460775.344843437], [2460775.3498766017, 2460775.34998845], [2460775.362515438, 2460775.362739134], [2460775.362850982, 2460775.3630746785], [2460775.365870881, 2460775.36710121], [2460775.3673249064, 2460775.3675486026], [2460775.3676604507, 2460775.3683315394], [2460775.3685552357, 2460775.36889078], [2460775.3691144763, 2460775.3692263244], [2460775.3693381725, 2460775.369673717], [2460775.3701211093, 2460775.3703448055], [2460775.3705685018, 2460775.371015894], [2460775.37123959, 2460775.371351438], [2460775.3716869825, 2460775.3717988306], [2460775.372022527, 2460775.372134375], [2460775.372246223, 2460775.3724699193], [2460775.373252856, 2460775.373364704], [2460775.3737002485, 2460775.373923945], [2460775.3743713372, 2460775.3745950335], [2460775.3747068816, 2460775.375042426], [2460775.3753779703, 2460775.3757135146], [2460775.376049059, 2460775.376160907], [2460775.378621565, 2460775.3788452614], [2460775.382200705, 2460775.3829836412], [2460775.38365473, 2460775.3843258186], [2460775.384661363, 2460775.384885059], [2460775.3849969073, 2460775.3853324517], [2460775.3854443, 2460775.385556148], [2460775.385667996, 2460775.385779844], [2460775.388240502, 2460775.3886878947], [2460775.3907011608, 2460775.390924857], [2460775.392043338, 2460775.3922670344], [2460775.3939447557, 2460775.3942803], [2460775.399537161, 2460775.399760857], [2460775.40255706, 2460775.4031163007], [2460775.403228149, 2460775.403339997], [2460775.4036755413, 2460775.4040110856], [2460775.405800655, 2460775.4062480475], [2460775.4083731617, 2460775.40848501], [2460775.408596858, 2460775.408820554], [2460775.4089324023, 2460775.4091560985], [2460775.409827187, 2460775.4100508834], [2460775.4107219717, 2460775.411057516], [2460775.411169364, 2460775.4112812122], [2460775.4116167566, 2460775.4117286047], [2460775.411840453, 2460775.413070782], [2460775.4132944783, 2460775.413965567], [2460775.4146366557, 2460775.414748504], [2460775.4149722, 2460775.4151958963], [2460775.416202529, 2460775.4166499213], [2460775.4169854657, 2460775.417097314], [2460775.4176565544, 2460775.4177684025], [2460775.4185513393, 2460775.4187750355], [2460775.421347542, 2460775.4214593903], [2460775.4215712384, 2460775.4216830865], [2460775.4232489597, 2460775.423584504], [2460775.431749416, 2460775.4319731123], [2460775.435216707, 2460775.4354404034], [2460775.4363351883, 2460775.4366707327], [2460775.4384603025, 2460775.4385721507], [2460775.438683999, 2460775.439355087], [2460775.44215129, 2460775.442374986], [2460775.4451711886, 2460775.445730429], [2460775.4471844546, 2460775.4472963028], [2460775.450763594, 2460775.4513228345], [2460775.4523294675, 2460775.452665012], [2460775.4542308855, 2460775.454678278], [2460775.4566915436, 2460775.45691524], [2460775.4576981766, 2460775.458257417], [2460775.460046987, 2460775.4604943795], [2460775.4627313414, 2460775.463178734], [2460775.463290582, 2460775.463514278], [2460775.4637379744, 2460775.4639616706], [2460775.471343646, 2460775.471455494], [2460775.4717910383, 2460775.4719028864], [2460775.4721265826, 2460775.472350279], [2460775.472797671, 2460775.473021367], [2460775.4745872407, 2460775.4753701775], [2460775.4761531143, 2460775.4763768106], [2460775.476600507, 2460775.476712355], [2460775.4786137724, 2460775.4788374687], [2460775.478949317, 2460775.479061165], [2460775.481857368, 2460775.482192912], [2460775.4833113933, 2460775.4835350895], [2460775.4856602033, 2460775.4858838995], [2460775.486554988, 2460775.4870023807], [2460775.488344558, 2460775.488456406], [2460775.494608052, 2460775.4947199], [2460775.501095242, 2460775.5017663306], [2460775.5091483057, 2460775.510154939], [2460775.511385268, 2460775.5116089643], [2460775.51362223, 2460775.5139577743], [2460775.5158591922, 2460775.5159710404], [2460775.516753977, 2460775.5169776734], [2460775.5204449645, 2460775.5206686608], [2460775.5286098765, 2460775.5288335728], [2460775.5292809648, 2460775.529616509], [2460775.5309586865, 2460775.5311823827], [2460775.542031649, 2460775.5422553453], [2460775.547064814, 2460775.5479595987], [2460775.555229726, 2460775.555453422], [2460775.5566837513, 2460775.557019295]] freq_flags: [[49911499.0234375, 50155639.6484375], [62362670.8984375, 62728881.8359375], [69931030.2734375, 70053100.5859375], [74935913.0859375, 75057983.3984375], [87265014.6484375, 108139038.0859375], [109970092.7734375, 110092163.0859375], [112655639.6484375, 112777709.9609375], [112899780.2734375, 113143920.8984375], [113265991.2109375, 113388061.5234375], [113632202.1484375, 113754272.4609375], [116073608.3984375, 116195678.7109375], [116439819.3359375, 116561889.6484375], [116683959.9609375, 116806030.2734375], [118392944.3359375, 118515014.6484375], [124740600.5859375, 125228881.8359375], [127548217.7734375, 127670288.0859375], [129989624.0234375, 130111694.3359375], [136215209.9609375, 136459350.5859375], [136825561.5234375, 138290405.2734375], [138412475.5859375, 139022827.1484375], [139511108.3984375, 139633178.7109375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [141952514.6484375, 142318725.5859375], [142562866.2109375, 143539428.7109375], [143783569.3359375, 144027709.9609375], [144638061.5234375, 144760131.8359375], [144882202.1484375, 145004272.4609375], [145370483.3984375, 145614624.0234375], [146224975.5859375, 146347045.8984375], [147445678.7109375, 147567749.0234375], [148178100.5859375, 148544311.5234375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [152938842.7734375, 153060913.0859375], [153305053.7109375, 153549194.3359375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [155258178.7109375, 155380249.0234375], [155502319.3359375, 155624389.6484375], [155868530.2734375, 155990600.5859375], [156845092.7734375, 156967163.0859375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158309936.5234375], [158432006.8359375, 158554077.1484375], [158676147.4609375, 158798217.7734375], [159164428.7109375, 159286499.0234375], [159774780.2734375, 160751342.7734375], [161361694.3359375, 161483764.6484375], [169174194.3359375, 169296264.6484375], [169906616.2109375, 170761108.3984375], [170883178.7109375, 171005249.0234375], [174545288.0859375, 175765991.2109375], [179306030.2734375, 179550170.8984375], [179672241.2109375, 179794311.5234375], [179916381.8359375, 180038452.1484375], [181137084.9609375, 181381225.5859375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [189193725.5859375, 189315795.8984375], [189926147.4609375, 190048217.7734375], [190780639.6484375, 190902709.9609375], [191024780.2734375, 191513061.5234375], [193222045.8984375, 193344116.2109375], [195297241.2109375, 195419311.5234375], [195663452.1484375, 195785522.4609375], [195907592.7734375, 196029663.0859375], [197128295.8984375, 197372436.5234375], [197860717.7734375, 198593139.6484375], [199203491.2109375, 199325561.5234375], [201644897.4609375, 201889038.0859375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [206893920.8984375, 207015991.2109375], [207138061.5234375, 207382202.1484375], [207504272.4609375, 207626342.7734375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [212142944.3359375, 212265014.6484375], [213119506.8359375, 213241577.1484375], [215194702.1484375, 215316772.4609375], [220565795.8984375, 221786499.0234375], [222030639.6484375, 222152709.9609375], [222274780.2734375, 224105834.9609375], [224349975.5859375, 224594116.2109375], [226791381.8359375, 226913452.1484375], [227157592.7734375, 228256225.5859375], [228500366.2109375, 228744506.8359375], [229110717.7734375, 229354858.3984375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231185913.0859375]] ex_ants: [[4, Jee], [4, Jnn], [5, Jnn], [8, Jee], [8, Jnn], [10, Jnn], [15, Jnn], [17, Jnn], [18, Jee], [18, Jnn], [20, Jnn], [21, Jee], [22, Jnn], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [29, Jee], [29, Jnn], [30, Jee], [31, Jnn], [32, Jnn], [33, Jnn], [34, Jee], [35, Jnn], [37, Jnn], [40, Jnn], [42, Jnn], [44, Jee], [45, Jnn], [46, Jee], [46, Jnn], [49, Jnn], [51, Jee], [59, Jnn], [60, Jnn], [62, Jee], [63, Jee], [63, Jnn], [64, Jee], [64, Jnn], [67, Jnn], [68, Jee], [71, Jnn], [73, Jnn], [75, Jnn], [76, Jee], [76, Jnn], [77, Jnn], [78, Jee], [78, Jnn], [80, Jnn], [81, Jnn], [82, Jee], [82, Jnn], [85, Jnn], [86, Jee], [86, Jnn], [87, Jee], [90, Jnn], [92, Jee], [93, Jnn], [94, Jnn], [97, Jnn], [99, Jnn], [100, Jee], [100, Jnn], [102, Jnn], [104, Jnn], [105, Jee], [107, Jee], [107, Jnn], [108, Jnn], [109, Jnn], [110, Jnn], [111, Jnn], [112, Jee], [113, Jnn], [115, Jee], [115, Jnn], [117, Jee], [119, Jee], [119, Jnn], [120, Jee], [120, Jnn], [121, Jee], [121, Jnn], [125, Jnn], [129, Jnn], [130, Jee], [130, Jnn], [134, Jee], [134, Jnn], [135, Jee], [137, Jee], [148, Jee], [148, Jnn], [149, Jee], [149, Jnn], [150, Jee], [151, Jee], [151, Jnn], [152, Jnn], [153, Jnn], [154, Jnn], [155, Jnn], [161, Jnn], [164, Jnn], [168, Jnn], [169, Jnn], [170, Jee], [170, Jnn], [171, Jee], [171, Jnn], [172, Jee], [172, Jnn], [176, Jnn], [180, Jee], [180, Jnn], [189, Jnn], [190, Jnn], [199, Jnn], [200, Jee], [200, Jnn], [202, Jnn], [206, Jnn], [209, Jnn], [212, Jnn], [213, Jee], [213, Jnn], [214, Jee], [214, Jnn], [215, Jee], [215, Jnn], [216, Jee], [218, Jnn], [221, Jee], [223, Jee], [227, Jee], [227, Jnn], [232, Jee], [233, Jnn], [238, Jnn], [239, Jee], [240, Jee], [240, Jnn], [244, Jee], [246, Jee], [246, Jnn], [250, Jee], [251, Jee], [253, Jnn], [255, Jee], [255, Jnn], [261, Jee], [261, Jnn], [262, Jee], [262, Jnn], [266, Jee], [266, Jnn], [268, Jee], [268, 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 41.58 minutes.