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 1851 *.sum.red_avg_zscore.h5 files starting with /mnt/sn1/data2/2460626/zen.2460626.25241.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 1851 *.sum.smooth.calfits files starting with /mnt/sn1/data2/2460626/zen.2460626.25241.sum.smooth.calfits.
In [6]:
assert len(zscore_files) == len(cal_files)
In [7]:
# extract z-scores and correct by a single number per polarization to account for biases created by filtering
zscore = {pol: uvf.metric_array[:, :, np.argwhere(uvf.polarization_array == utils.polstr2num(pol, x_orientation=uvf.x_orientation))[0][0]] for pol in ['ee', 'nn']}
zscore = {pol: zscore[pol] - np.nanmedian(zscore[pol]) for pol in zscore}
In [8]:
freqs = uvf.freq_array
times = uvf.time_array
In [9]:
extent = [freqs[0] / 1e6, freqs[-1] / 1e6, times[-1] - int(times[0]), times[0] - int(times[0])]
In [10]:
def plot_max_z_score(zscore, flags=None, vmin=-5, vmax=5):
if flags is None:
flags = np.any(~np.isfinite(list(zscore.values())), axis=0)
plt.figure(figsize=(14,10), dpi=100)
plt.imshow(np.where(flags, np.nan, np.nanmax([zscore['ee'], zscore['nn']], axis=0)), aspect='auto',
cmap='coolwarm', interpolation='none', vmin=vmin, vmax=vmax, extent=extent)
plt.colorbar(location='top', label='Max z-score of either polarization', extend='both', aspect=40, pad=.02)
plt.xlabel('Frequency (MHz)')
plt.ylabel(f'JD - {int(times[0])}')
plt.tight_layout()
Figure 1: Waterfall of Maximum z-Score of Either Polarization Before Round 2 Flagging¶
Shows the worse of the two results from delay_filtered_average_zscore from either polarization. Dips near flagged channels are expected, due to overfitting of noise. Positive-going excursions are problematic and likely evidence of RFI.
In [11]:
plot_max_z_score(zscore)
All-NaN axis encountered
In [12]:
def plot_histogram():
plt.figure(figsize=(14,4), dpi=100)
bins = np.arange(-50, 100, .1)
hist_ee = plt.hist(np.ravel(zscore['ee']), bins=bins, density=True, label='ee-polarized z-scores', alpha=.5)
hist_nn = plt.hist(np.ravel(zscore['nn']), bins=bins, density=True, label='nn-polarized z-scores', alpha=.5)
plt.plot(bins, (2*np.pi)**-.5 * np.exp(-bins**2 / 2), 'k:', label='Gaussian approximate\nnoise-only distribution')
plt.axvline(WS_Z_THRESH, c='r', ls='--', label='Watershed z-score')
plt.axvline(Z_THRESH, c='r', ls='-', label='Threshold z-score')
plt.yscale('log')
all_densities = np.concatenate([hist_ee[0][hist_ee[0] > 0], hist_nn[0][hist_nn[0] > 0]])
plt.ylim(np.min(all_densities) / 2, np.max(all_densities) * 2)
plt.xlim([-50, 100])
plt.legend()
plt.xlabel('z-score')
plt.ylabel('Density')
plt.tight_layout()
Figure 2: Histogram of z-scores¶
Shows a comparison of the histogram of z-scores in this file (one per polarization) to a Gaussian approximation of what one might expect from thermal noise. Without filtering, the actual distribution is a weighted sum of Rayleigh distributions. Filtering further complicates this. To make the z-scores more reliable, a single per-polarization median is subtracted from each waterfall, which allows us to flag low-level outliers with more confidence. Any points beyond the solid red line are flagged. Any points neighboring a flag beyond the dashed red line are also flagged. Finally, flagging is performed for low-level outliers in whole times or channels.
In [13]:
plot_histogram()
Perform flagging¶
In [14]:
def iteratively_flag_on_averaged_zscore(flags, zscore, avg_func=np.nanmean, avg_z_thresh=AVG_Z_THRESH, verbose=True):
'''Flag whole integrations or channels based on average z-score. This is done
iteratively to prevent bad times affecting channel averages or vice versa.'''
flagged_chan_count = 0
flagged_int_count = 0
while True:
zspec = avg_func(np.where(flags, np.nan, zscore), axis=0)
ztseries = avg_func(np.where(flags, np.nan, zscore), axis=1)
if (np.nanmax(zspec) < avg_z_thresh) and (np.nanmax(ztseries) < avg_z_thresh):
break
if np.nanmax(zspec) >= np.nanmax(ztseries):
flagged_chan_count += np.sum((zspec >= np.nanmax(ztseries)) & (zspec >= avg_z_thresh))
flags[:, (zspec >= np.nanmax(ztseries)) & (zspec >= avg_z_thresh)] = True
else:
flagged_int_count += np.sum((ztseries >= np.nanmax(zspec)) & (ztseries >= avg_z_thresh))
flags[(ztseries >= np.nanmax(zspec)) & (ztseries >= avg_z_thresh), :] = True
if verbose:
print(f'\tFlagging an additional {flagged_int_count} integrations and {flagged_chan_count} channels.')
def impose_max_chan_flag_frac(flags, max_flag_frac=MAX_FREQ_FLAG_FRAC, verbose=True):
'''Flag channels already flagged more than max_flag_frac (excluding completely flagged times).'''
unflagged_times = ~np.all(flags, axis=1)
frequently_flagged_chans = np.mean(flags[unflagged_times, :], axis=0) >= max_flag_frac
if verbose:
print(f'\tFlagging {np.sum(frequently_flagged_chans) - np.sum(np.all(flags, axis=0))} channels previously flagged {max_flag_frac:.2%} or more.')
flags[:, frequently_flagged_chans] = True
def impose_max_time_flag_frac(flags, max_flag_frac=MAX_TIME_FLAG_FRAC, verbose=True):
'''Flag times already flagged more than max_flag_frac (excluding completely flagged channels).'''
unflagged_chans = ~np.all(flags, axis=0)
frequently_flagged_times = np.mean(flags[:, unflagged_chans], axis=1) >= max_flag_frac
if verbose:
print(f'\tFlagging {np.sum(frequently_flagged_times) - np.sum(np.all(flags, axis=1))} times previously flagged {max_flag_frac:.2%} or more.')
flags[frequently_flagged_times, :] = True
def time_avg_zscore_dly_filt_SNRs(flags, filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF):
"""Produces SNRs after time-averaging z-scores and delay filtering, accounting for flagging's effect on the filter."""
# figure out high and low band based on FM gap at 100 MHz
flagged_stretches = true_stretches(np.all(flags, axis=0))
FM_gap = [fs for fs in flagged_stretches if fs.start <= np.argmin(np.abs(freqs - 100e6)) < fs.stop][0]
low_band = slice((0 if flagged_stretches[0].start != 0 else flagged_stretches[0].stop), FM_gap.start)
high_band = slice(FM_gap.stop, (len(freqs) if flagged_stretches[-1].stop != len(freqs) else flagged_stretches[-1].start))
filt_SNR = {}
for pol in zscore:
# calculate timeavg_SNR and filter
noise_prediction = 1.0 / np.sum(~flags, axis=0)**.5
timeavg_SNR = np.nanmean(np.where(flags, np.nan, zscore[pol] / noise_prediction), axis=0)
wgts = np.where(np.isfinite(timeavg_SNR), 1, 0)
model = np.zeros_like(timeavg_SNR)
for band in [low_band, high_band]:
model[band], _, _ = dspec.fourier_filter(freqs[band], np.where(np.isfinite(timeavg_SNR[band]), timeavg_SNR[band], 0),
wgts[band], [0], [AVG_SPECTRUM_FILTER_DELAY / 1e9], mode="dpss_solve",
eigenval_cutoff=[EIGENVAL_CUTOFF], suppression_factors=[EIGENVAL_CUTOFF])
filt_SNR[pol] = timeavg_SNR - model
# correct for impact of filter
correction_factors = np.ones_like(wgts) * np.nan
for band in [low_band, high_band]:
X = dspec.dpss_operator(freqs[band], [0], filter_half_widths=[AVG_SPECTRUM_FILTER_DELAY / 1e9], eigenval_cutoff=[EIGENVAL_CUTOFF])[0]
W = wgts[band]
leverage = np.diag(X @ np.linalg.pinv(np.dot(X.T * W, X)) @ (X.T * W))
correction_factors[band] = np.where(leverage > 0, (1 - leverage)**.5, np.nan) # because the underlying data should be gaussian
filt_SNR[pol] /= correction_factors
return filt_SNR
def iteratively_flag_on_delay_filtered_time_avg_zscore(flags, thresh=TIME_AVG_DELAY_FILT_SNR_THRESH, dynamic_range=TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE,
filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF):
"""Flag whole channels based on their outlierness after delay-filterd time-averaged zscores.
This is done iteratively since the delay filter can be unduly influenced by large outliers."""
filt_SNR = time_avg_zscore_dly_filt_SNRs(flags, filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF)
while True:
largest_SNR = np.nanmax(list(filt_SNR.values()))
if largest_SNR < thresh:
break
#
cut = np.max([thresh, largest_SNR / dynamic_range])
for pol in filt_SNR:
flags[:, filt_SNR[pol] > cut] = True
filt_SNR = time_avg_zscore_dly_filt_SNRs(flags, filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF)
In [15]:
flags = np.any(~np.isfinite(list(zscore.values())), axis=0)
print(f'{np.mean(flags):.3%} of waterfall flagged to start.')
# flag whole integrations or channels using outliers in median
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
iteratively_flag_on_averaged_zscore(flags, zscore[pol], avg_func=np.nanmedian, avg_z_thresh=AVG_Z_THRESH, verbose=True)
impose_max_chan_flag_frac(flags, max_flag_frac=MAX_FREQ_FLAG_FRAC, verbose=True)
impose_max_time_flag_frac(flags, max_flag_frac=MAX_TIME_FLAG_FRAC, verbose=True)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after flagging whole times and channels with median z > {AVG_Z_THRESH}.')
# flag largest outliers
for pol in ['ee', 'nn']:
flags |= (zscore[pol] > Z_THRESH)
print(f'{np.mean(flags):.3%} of waterfall flagged after flagging z > {Z_THRESH} outliers.')
# watershed flagging
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
flags |= xrfi._ws_flag_waterfall(zscore[pol], flags, WS_Z_THRESH)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after watershed flagging on z > {WS_Z_THRESH} neighbors of prior flags.')
# flag whole integrations or channels using outliers in mean
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
iteratively_flag_on_averaged_zscore(flags, zscore[pol], avg_func=np.nanmean, avg_z_thresh=AVG_Z_THRESH, verbose=True)
impose_max_chan_flag_frac(flags, max_flag_frac=MAX_FREQ_FLAG_FRAC, verbose=True)
impose_max_time_flag_frac(flags, max_flag_frac=MAX_TIME_FLAG_FRAC, verbose=True)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after flagging whole times and channels with average z > {AVG_Z_THRESH}.')
# flag channels based on delay filter
iteratively_flag_on_delay_filtered_time_avg_zscore(flags, thresh=TIME_AVG_DELAY_FILT_SNR_THRESH, dynamic_range=TIME_AVG_DELAY_FILT_SNR_DYNAMIC_RANGE,
filter_delay=AVG_SPECTRUM_FILTER_DELAY, eigenval_cutoff=EIGENVAL_CUTOFF)
print(f'{np.mean(flags):.3%} of flagging channels that are {TIME_AVG_DELAY_FILT_SNR_THRESH}σ outliers after delay filtering the time average.')
# watershed flagging again
while True:
nflags = np.sum(flags)
for pol in ['ee', 'nn']:
flags |= xrfi._ws_flag_waterfall(zscore[pol], flags, WS_Z_THRESH)
if np.sum(flags) == nflags:
break
print(f'{np.mean(flags):.3%} of waterfall flagged after another round of watershed flagging on z > {WS_Z_THRESH} neighbors of prior flags.')
19.675% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 0 integrations and 28 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 11 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. 21.910% of waterfall flagged after flagging whole times and channels with median z > 1.0. 23.078% of waterfall flagged after flagging z > 4.0 outliers.
24.553% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 30 channels previously flagged 25.00% or more. Flagging 86 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. 27.422% 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
35.438% of flagging channels that are 4.0σ outliers after delay filtering the time average.
35.962% 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 1851 *.sum.flag_waterfall_round_2.h5 files starting with /mnt/sn1/data2/2460626/zen.2460626.25241.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/data2/2460626/2460626_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460626.2771301214, 2460626.2772419695], [2460626.280485565, 2460626.280709261], [2460626.2824988305, 2460626.282946223], [2460626.2837291597, 2460626.283841008], [2460626.2844002484, 2460626.2845120966], [2460626.2846239447, 2460626.284847641], [2460626.290104502, 2460626.290328198], [2460626.301736705, 2460626.3019604012], [2460626.314039997, 2460626.314151845], [2460626.3431205046, 2460626.3432323528], [2460626.3482655175, 2460626.3483773656], [2460626.3498313907, 2460626.349943239], [2460626.35106172, 2460626.3513972643], [2460626.3516209605, 2460626.351956505], [2460626.3535223785, 2460626.354081619], [2460626.3622465306, 2460626.3623583787], [2460626.3652664297, 2460626.365378278], [2460626.367838936, 2460626.368062632], [2460626.368845569, 2460626.369069265], [2460626.3715299238, 2460626.371865468], [2460626.3765630885, 2460626.3767867847], [2460626.378464506, 2460626.378576354], [2460626.383609519, 2460626.383721367], [2460626.398037925, 2460626.3984853174], [2460626.4195127613, 2460626.4198483056], [2460626.4445667374, 2460626.4451259775], [2460626.4460207624, 2460626.4461326106], [2460626.4471392436, 2460626.447474788], [2460626.4559752443, 2460626.4560870924], [2460626.4646993964, 2460626.4648112445], [2460626.4667126625, 2460626.467383751], [2460626.472416916, 2460626.472528764], [2460626.4734235485, 2460626.4735353966], [2460626.4774500807, 2460626.477561929], [2460626.486509777, 2460626.4869571696], [2460626.4923258787, 2460626.492549575], [2460626.495457626, 2460626.495569474], [2460626.502168512, 2460626.50228036], [2460626.519840513, 2460626.519952361], [2460626.5203997535, 2460626.5205116016], [2460626.525097374, 2460626.52532107], [2460626.5461248183, 2460626.5462366664], [2460626.549032869, 2460626.549256565], [2460626.5504868943, 2460626.551157983], [2460626.5518290717, 2460626.5521646156], [2460626.55250016, 2460626.552612008], [2460626.5550726666, 2460626.555296363], [2460626.5570859327, 2460626.557197781], [2460626.56077692, 2460626.5610006163], [2460626.5630138824, 2460626.5631257305], [2460626.5646916036, 2460626.5648034518], [2460626.5727446675, 2460626.5728565156], [2460626.5802384904, 2460626.5803503385], [2460626.5811332753, 2460626.5813569715], [2460626.583593934, 2460626.583705782], [2460626.5919825416, 2460626.592318086], [2460626.593436567, 2460626.5937721115], [2460626.594331352, 2460626.5944432], [2460626.595226137, 2460626.595337985], [2460626.600482998, 2460626.600594846], [2460626.602831808, 2460626.602943656], [2460626.6061872514, 2460626.6064109476], [2460626.610884872, 2460626.611108568], [2460626.6212867456, 2460626.621510442], [2460626.628780569, 2460626.628892417], [2460626.6302345945, 2460626.6303464426], [2460626.633142645, 2460626.6334781894], [2460626.6338137337, 2460626.633925582], [2460626.6434326707, 2460626.643544519], [2460626.6439919113, 2460626.6441037594], [2460626.6442156075, 2460626.644551152], [2460626.6452222406, 2460626.6453340887], [2460626.6470118104, 2460626.6471236586], [2460626.6472355067, 2460626.647459203], [2460626.6482421393, 2460626.6663615326]] freq_flags: [[49911499.0234375, 50155639.6484375], [62362670.8984375, 62728881.8359375], [69931030.2734375, 70053100.5859375], [87387084.9609375, 108139038.0859375], [109970092.7734375, 110092163.0859375], [112045288.0859375, 112411499.0234375], [112533569.3359375, 113021850.5859375], [113265991.2109375, 114120483.3984375], [114364624.0234375, 114608764.6484375], [115585327.1484375, 115707397.4609375], [115829467.7734375, 115951538.0859375], [116073608.3984375, 116928100.5859375], [123641967.7734375, 123764038.0859375], [124740600.5859375, 125228881.8359375], [126205444.3359375, 126327514.6484375], [127548217.7734375, 127670288.0859375], [129989624.0234375, 130111694.3359375], [132553100.5859375, 132675170.8984375], [133895874.0234375, 134017944.3359375], [134140014.6484375, 134262084.9609375], [134994506.8359375, 135116577.1484375], [135238647.4609375, 135360717.7734375], [135482788.0859375, 135726928.7109375], [135848999.0234375, 135971069.3359375], [136215209.9609375, 138290405.2734375], [138656616.2109375, 138778686.5234375], [138900756.8359375, 139144897.4609375], [139266967.7734375, 139755249.0234375], [139877319.3359375, 140365600.5859375], [140609741.2109375, 141098022.4609375], [141220092.7734375, 142440795.8984375], [142562866.2109375, 143661499.0234375], [143783569.3359375, 144271850.5859375], [144638061.5234375, 144882202.1484375], [145248413.0859375, 145614624.0234375], [145736694.3359375, 146102905.2734375], [147445678.7109375, 147567749.0234375], [148178100.5859375, 148422241.2109375], [149154663.0859375, 149276733.3984375], [149887084.9609375, 150009155.2734375], [153427124.0234375, 153549194.3359375], [154159545.8984375, 154403686.5234375], [154891967.7734375, 155136108.3984375], [155258178.7109375, 155380249.0234375], [155990600.5859375, 156112670.8984375], [156845092.7734375, 156967163.0859375], [157577514.6484375, 157699584.9609375], [157943725.5859375, 158554077.1484375], [159164428.7109375, 159286499.0234375], [160140991.2109375, 160385131.8359375], [161361694.3359375, 161483764.6484375], [169906616.2109375, 170150756.8359375], [170394897.4609375, 170639038.0859375], [170883178.7109375, 171005249.0234375], [171859741.2109375, 171981811.5234375], [175155639.6484375, 175277709.9609375], [180770874.0234375, 181625366.2109375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [188949584.9609375, 189071655.2734375], [189193725.5859375, 189315795.8984375], [189926147.4609375, 190048217.7734375], [190780639.6484375, 191879272.4609375], [193222045.8984375, 193344116.2109375], [195663452.1484375, 195785522.4609375], [196395874.0234375, 196517944.3359375], [196884155.2734375, 197006225.5859375], [197128295.8984375, 197372436.5234375], [197860717.7734375, 198715209.9609375], [199203491.2109375, 199325561.5234375], [200057983.3984375, 200180053.7109375], [201644897.4609375, 201889038.0859375], [203231811.5234375, 203353881.8359375], [204452514.6484375, 205429077.1484375], [206283569.3359375, 206527709.9609375], [206771850.5859375, 207870483.3984375], [207992553.7109375, 209823608.3984375], [209945678.7109375, 210922241.2109375], [211410522.4609375, 211532592.7734375], [211776733.3984375, 212631225.5859375], [213485717.7734375, 213607788.0859375], [215194702.1484375, 215316772.4609375], [216659545.8984375, 216781616.2109375], [219833374.0234375, 219955444.3359375], [220321655.2734375, 220565795.8984375], [220687866.2109375, 221176147.4609375], [222763061.5234375, 222885131.8359375], [223007202.1484375, 223739624.0234375], [225692749.0234375, 225814819.3359375], [227035522.4609375, 227890014.6484375], [229110717.7734375, 229354858.3984375], [229965209.9609375, 230087280.2734375], [231063842.7734375, 231185913.0859375]] 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], [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], [57, Jee], [61, Jee], [61, Jnn], [62, Jee], [62, Jnn], [63, Jee], [63, Jnn], [64, Jee], [64, Jnn], [69, Jee], [71, Jee], [71, Jnn], [72, Jee], [72, Jnn], [73, Jee], [77, Jee], [77, Jnn], [78, Jee], [78, Jnn], [81, Jee], [81, Jnn], [82, Jee], [82, Jnn], [83, Jee], [84, Jnn], [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, Jee], [98, Jnn], [100, Jnn], [103, Jnn], [104, Jnn], [107, Jee], [107, Jnn], [108, Jnn], [109, Jnn], [120, Jee], [120, Jnn], [121, Jee], [125, Jee], [125, Jnn], [127, Jee], [127, Jnn], [130, Jee], [130, Jnn], [131, Jee], [133, Jee], [134, Jee], [136, Jee], [136, Jnn], [137, Jee], [142, Jnn], [161, Jee], [161, Jnn], [162, Jee], [162, 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], [194, Jee], [198, Jnn], [199, Jnn], [200, Jee], [200, Jnn], [201, Jnn], [202, Jnn], [204, Jee], [205, Jnn], [206, Jee], [207, Jee], [207, Jnn], [208, Jee], [209, Jnn], [212, Jnn], [213, Jee], [213, Jnn], [215, Jnn], [218, Jnn], [231, Jee], [232, Jee], [234, Jnn], [235, Jee], [235, 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], [256, Jee], [256, 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 35.51 minutes.