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/data1/2460697/zen.2460697.25250.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/data1/2460697/zen.2460697.25250.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.')
20.351% of waterfall flagged to start.
All-NaN slice encountered
Flagging an additional 2 integrations and 10 channels. Flagging 0 channels previously flagged 25.00% or more. Flagging 0 times previously flagged 10.00% or more.
Flagging an additional 924 integrations and 3 channels. Flagging 1 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. 42.078% of waterfall flagged after flagging whole times and channels with median z > 1.0. 42.786% of waterfall flagged after flagging z > 4.0 outliers.
45.164% of waterfall flagged after watershed flagging on z > 2.0 neighbors of prior flags. Flagging an additional 0 integrations and 0 channels. Flagging 46 channels previously flagged 25.00% or more. Flagging 337 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. 52.625% 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
54.697% of flagging channels that are 4.0σ outliers after delay filtering the time average.
55.121% 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/data1/2460697/zen.2460697.25250.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/data1/2460697/2460697_aposteriori_flags.yaml ------------------------------------------------------------------------------ JD_flags: [[2460697.2555244206, 2460697.2570902943], [2460697.257761383, 2460697.2591035604], [2460697.2592154085, 2460697.2594391047], [2460697.2595509524, 2460697.2596628005], [2460697.260222041, 2460697.2604457373], [2460697.261004978, 2460697.261116826], [2460697.2614523703, 2460697.2615642184], [2460697.2617879147, 2460697.2625708515], [2460697.2627945477, 2460697.262906396], [2460697.263130092, 2460697.2645841176], [2460697.264919662, 2460697.265367054], [2460697.2690580417, 2460697.26916989], [2460697.270176523, 2460697.2707357635], [2460697.2714068517, 2460697.271854244], [2460697.273420118, 2460697.2738675103], [2460697.275768928, 2460697.2772229533], [2460697.280802093, 2460697.280913941], [2460697.2812494854, 2460697.28158503], [2460697.283039055, 2460697.283374599], [2460697.285052321, 2460697.2858352577], [2460697.287065587, 2460697.2880722196], [2460697.2888551564, 2460697.289414397], [2460697.289526245, 2460697.2898617894], [2460697.2910921187, 2460697.2916513593], [2460697.294894954, 2460697.2951186504], [2460697.295566043, 2460697.2959015872], [2460697.296684524, 2460697.2969082203], [2460697.2970200684, 2460697.2972437646], [2460697.2979148533, 2460697.298362246], [2460697.298474094, 2460697.298585942], [2460697.2988096382, 2460697.2989214864], [2460697.300151815, 2460697.3003755114], [2460697.3004873595, 2460697.3007110558], [2460697.305073132, 2460697.305296828], [2460697.307981183, 2460697.308316727], [2460697.3084285753, 2460697.308987816], [2460697.309099664, 2460697.3095470564], [2460697.312566955, 2460697.313349892], [2460697.3148039174, 2460697.3149157655], [2460697.3182712086, 2460697.3183830567], [2460697.319501538, 2460697.319613386], [2460697.3211792596, 2460697.3212911077], [2460697.3227451327, 2460697.322856981], [2460697.32822569, 2460697.328337538], [2460697.328449386, 2460697.3286730824], [2460697.3287849305, 2460697.329120475], [2460697.3297915636, 2460697.3299034117], [2460697.332028526, 2460697.332252222], [2460697.337397235, 2460697.337509083], [2460697.346792476, 2460697.347016172], [2460697.350259767, 2460697.3509308556], [2460697.3510427037, 2460697.352049337], [2460697.3528322736, 2460697.3535033623], [2460697.355628476, 2460697.3559640204], [2460697.358648375, 2460697.3590957676], [2460697.363569692, 2460697.3643526286], [2460697.364688173, 2460697.364911869], [2460697.3650237173, 2460697.3651355654], [2460697.3652474135, 2460697.3662540466], [2460697.3663658947, 2460697.497004483], [2460697.497228179, 2460697.4973400272], [2460697.498122964, 2460697.498234812], [2460697.498905901, 2460697.4994651414], [2460697.5012547113, 2460697.5013665594], [2460697.505057547, 2460697.505169395], [2460697.505281243, 2460697.505393091], [2460697.5058404836, 2460697.506287876], [2460697.5093077747, 2460697.509419623], [2460697.517584535, 2460697.517920079], [2460697.533578814, 2460697.5339143584], [2460697.5405133967, 2460697.540625245], [2460697.541408181, 2460697.5415200293], [2460697.549573093, 2460697.5499086375], [2460697.5518100555, 2460697.5520337517], [2460697.5552773466, 2460697.5553891947], [2460697.563442258, 2460697.5636659544], [2460697.574738917, 2460697.574850765], [2460697.578541753, 2460697.578877297], [2460697.590285804, 2460697.5905095004], [2460697.5919635254, 2460697.5920753735], [2460697.5931938547, 2460697.593305703], [2460697.5961019057, 2460697.596325602], [2460697.6086288933, 2460697.6088525895], [2460697.6115369443, 2460697.6116487924], [2460697.613326514, 2460697.6135502104], [2460697.6148923873, 2460697.6151160835], [2460697.619701856, 2460697.619925552], [2460697.6242876286, 2460697.6243994767], [2460697.6266364385, 2460697.626860135], [2460697.6294326414, 2460697.6296563377], [2460697.6306629707, 2460697.630998515], [2460697.6359198317, 2460697.63603168], [2460697.651131174, 2460697.6513548703], [2460697.664441099, 2460697.6664543645]] freq_flags: [[49911499.0234375, 50155639.6484375], [62118530.2734375, 62850952.1484375], [66146850.5859375, 66879272.4609375], [69931030.2734375, 70053100.5859375], [87387084.9609375, 108016967.7734375], [109970092.7734375, 110092163.0859375], [112167358.3984375, 112289428.7109375], [112655639.6484375, 112777709.9609375], [112899780.2734375, 113143920.8984375], [113265991.2109375, 113876342.7734375], [115829467.7734375, 115951538.0859375], [116073608.3984375, 116928100.5859375], [124740600.5859375, 125228881.8359375], [127548217.7734375, 127670288.0859375], [129989624.0234375, 130111694.3359375], [136337280.2734375, 136459350.5859375], [136825561.5234375, 138290405.2734375], [141464233.3984375, 141586303.7109375], [141708374.0234375, 141830444.3359375], [142074584.9609375, 142318725.5859375], [142440795.8984375, 142562866.2109375], [142684936.5234375, 143173217.7734375], [143295288.0859375, 143417358.3984375], [143783569.3359375, 144027709.9609375], [144882202.1484375, 145004272.4609375], [145492553.7109375, 145614624.0234375], [145736694.3359375, 146102905.2734375], [147445678.7109375, 147567749.0234375], [148422241.2109375, 148544311.5234375], [149154663.0859375, 149276733.3984375], [149642944.3359375, 149765014.6484375], [149887084.9609375, 150009155.2734375], [154159545.8984375, 154403686.5234375], [155014038.0859375, 155136108.3984375], [155258178.7109375, 155380249.0234375], [157577514.6484375, 157699584.9609375], [158187866.2109375, 158309936.5234375], [159164428.7109375, 159286499.0234375], [168807983.3984375, 168930053.7109375], [169906616.2109375, 170150756.8359375], [170272827.1484375, 170394897.4609375], [170883178.7109375, 171005249.0234375], [171249389.6484375, 171371459.9609375], [171737670.8984375, 171859741.2109375], [175155639.6484375, 175277709.9609375], [180770874.0234375, 181625366.2109375], [183212280.2734375, 183334350.5859375], [187362670.8984375, 187606811.5234375], [189926147.4609375, 190048217.7734375], [190292358.3984375, 190414428.7109375], [190902709.9609375, 191757202.1484375], [193344116.2109375, 193466186.5234375], [195663452.1484375, 195785522.4609375], [197128295.8984375, 197372436.5234375], [198104858.3984375, 198348999.0234375], [199203491.2109375, 199325561.5234375], [200790405.2734375, 200912475.5859375], [201644897.4609375, 201889038.0859375], [203964233.3984375, 204086303.7109375], [204940795.8984375, 205062866.2109375], [205184936.5234375, 205307006.8359375], [207138061.5234375, 207382202.1484375], [207504272.4609375, 207626342.7734375], [208480834.9609375, 208724975.5859375], [209945678.7109375, 210067749.0234375], [211166381.8359375, 211288452.1484375], [212142944.3359375, 212265014.6484375], [215194702.1484375, 215316772.4609375], [220565795.8984375, 220809936.5234375], [222763061.5234375, 223739624.0234375], [227401733.3984375, 227523803.7109375], [227645874.0234375, 227767944.3359375], [228744506.8359375, 228866577.1484375], [229110717.7734375, 229354858.3984375], [229476928.7109375, 229843139.6484375], [229965209.9609375, 230087280.2734375], [230941772.4609375, 234359741.2109375]] ex_ants: [[7, Jnn], [8, Jnn], [9, Jee], [9, Jnn], [10, Jee], [10, Jnn], [15, Jnn], [16, Jee], [17, Jnn], [18, Jnn], [19, Jnn], [20, Jnn], [21, Jee], [21, Jnn], [22, Jee], [22, Jnn], [27, Jee], [27, Jnn], [28, Jee], [28, Jnn], [31, Jnn], [32, Jnn], [33, Jnn], [34, Jee], [34, Jnn], [35, Jee], [35, Jnn], [37, Jnn], [40, Jee], [40, Jnn], [41, Jee], [41, Jnn], [42, Jee], [42, Jnn], [45, Jee], [45, Jnn], [46, Jee], [46, Jnn], [47, Jee], [47, Jnn], [48, Jee], [48, Jnn], [49, Jee], [49, Jnn], [51, Jee], [54, Jee], [54, Jnn], [55, Jee], [55, Jnn], [56, Jee], [56, Jnn], [57, Jee], [57, Jnn], [61, Jee], [61, Jnn], [62, Jee], [62, Jnn], [63, Jee], [63, Jnn], [64, Jee], [64, Jnn], [68, Jee], [68, Jnn], [69, Jee], [69, Jnn], [70, Jee], [70, Jnn], [71, Jee], [71, Jnn], [72, Jee], [72, Jnn], [73, Jee], [77, Jee], [77, Jnn], [78, Jee], [78, Jnn], [80, Jnn], [82, Jee], [82, Jnn], [83, Jnn], [84, Jee], [86, Jee], [88, Jee], [88, Jnn], [89, Jee], [90, Jee], [90, Jnn], [92, Jee], [93, Jee], [93, Jnn], [94, Jnn], [95, Jee], [97, Jnn], [98, Jnn], [104, Jee], [104, Jnn], [107, Jee], [107, Jnn], [108, Jnn], [109, Jnn], [110, Jnn], [111, Jnn], [112, Jnn], [113, Jnn], [114, Jee], [114, Jnn], [116, Jee], [116, Jnn], [117, Jee], [117, Jnn], [120, Jee], [120, Jnn], [121, Jee], [121, Jnn], [124, Jee], [125, Jee], [125, Jnn], [127, Jee], [127, Jnn], [129, Jnn], [130, Jnn], [134, Jee], [135, Jee], [136, Jnn], [137, Jee], [142, Jnn], [143, Jee], [143, Jnn], [144, Jee], [144, Jnn], [145, Jee], [145, Jnn], [146, Jee], [146, Jnn], [149, Jnn], [150, Jnn], [152, Jnn], [154, Jnn], [155, Jee], [161, Jee], [161, Jnn], [163, Jee], [163, Jnn], [164, Jee], [164, Jnn], [165, Jee], [165, Jnn], [166, Jee], [166, Jnn], [170, Jee], [170, Jnn], [171, Jnn], [173, Jnn], [179, Jee], [179, Jnn], [180, Jnn], [182, Jee], [182, Jnn], [184, Jee], [184, Jnn], [185, Jee], [185, Jnn], [186, Jee], [186, Jnn], [187, Jee], [187, Jnn], [188, Jnn], [198, Jnn], [199, Jee], [199, Jnn], [200, Jee], [200, Jnn], [201, Jnn], [202, Jnn], [204, Jee], [204, Jnn], [206, Jee], [208, Jee], [209, Jnn], [212, Jnn], [213, Jee], [213, Jnn], [215, Jnn], [218, Jee], [218, Jnn], [232, Jee], [235, Jnn], [240, Jee], [240, Jnn], [241, Jee], [241, Jnn], [242, Jee], [242, Jnn], [243, Jee], [243, Jnn], [245, Jnn], [246, Jee], [250, Jee], [251, Jee], [253, Jnn], [255, Jnn], [262, Jee], [262, Jnn], [268, Jnn], [320, Jee], [320, Jnn], [321, Jee], [321, Jnn], [323, Jee], [323, Jnn], [324, Jee], [324, Jnn], [325, Jee], [325, Jnn], [326, Jee], [326, Jnn], [327, Jee], [327, Jnn], [328, Jee], [328, Jnn], [329, Jee], [329, Jnn], [331, Jee], [331, Jnn], [332, Jee], [332, Jnn], [333, Jee], [333, Jnn], [336, Jee], [336, Jnn], [340, Jee], [340, Jnn]]
Metadata¶
In [24]:
for repo in ['hera_cal', 'hera_qm', 'hera_filters', 'hera_notebook_templates', 'pyuvdata']:
exec(f'from {repo} import __version__')
print(f'{repo}: {__version__}')
hera_cal: 3.6.2.dev110+g0529798 hera_qm: 2.2.0 hera_filters: 0.1.6.dev1+g297dcce
hera_notebook_templates: 0.1.dev936+gdc93cad pyuvdata: 3.0.1.dev70+g283dda3
In [25]:
print(f'Finished execution in {(time.time() - tstart) / 60:.2f} minutes.')
Finished execution in 39.89 minutes.