Skip to content

Commit

Permalink
Inherit straxen records and peaklets (#126)
Browse files Browse the repository at this point in the history 
* inherit from straxen

* inheirt

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
  • Loading branch information
@FaroutYLq @pre-commit-ci
FaroutYLq and pre-commit-ci[bot] authored Apr 23, 2024
1 parent dfdf4b0 commit 7e159ee
Show file tree
Hide file tree
Showing 3 changed files with 3 additions and 310 deletions.
2 changes: 1 addition & 1 deletion jobs/job.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -175,7 +175,7 @@ def main():
print_versions()
_, runid = sys.argv
runid = int(runid)

# Process the saltax desired mode
logging.info("Loading context...")
settings = load_config()
Expand Down
185 changes: 1 addition & 184 deletions saltax/plugins/peaklets.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,6 @@
import strax
from strax.processing.peak_building import _build_hit_waveform
from strax import utils
from immutabledict import immutabledict
from strax.processing.general import _touching_windows
from strax.dtypes import DIGITAL_SUM_WAVEFORM_CHANNEL
from .records import SCHANNEL_STARTS_AT
Expand All @@ -24,7 +23,7 @@
help="'data', 'simu', or 'salt'",
),
)
class SPeaklets(strax.Plugin):
class SPeaklets(straxen.Peaklets):
"""
Split records into:
- peaklets
Expand All @@ -45,141 +44,6 @@ class SPeaklets(strax.Plugin):
extension overlaps with any peaks or other hits.
"""

depends_on = ("records",)
provides = ("peaklets", "lone_hits")
data_kind = dict(peaklets="peaklets", lone_hits="lone_hits")
parallel = "process"
compressor = "zstd"

__version__ = "0.0.4"

peaklet_gap_threshold = straxen.URLConfig(
default=700, infer_type=False, help="No hits for this many ns triggers a new peak"
)

peak_left_extension = straxen.URLConfig(
default=30, infer_type=False, help="Include this many ns left of hits in peaks"
)

peak_right_extension = straxen.URLConfig(
default=200, infer_type=False, help="Include this many ns right of hits in peaks"
)

peak_min_pmts = straxen.URLConfig(
default=2,
infer_type=False,
help="Minimum number of contributing PMTs needed to define a peak",
)

peak_split_gof_threshold = straxen.URLConfig(
# See https://xe1t-wiki.lngs.infn.it/doku.php?id=
# xenon:xenonnt:analysis:strax_clustering_classification
# #natural_breaks_splitting
# for more information
default=(None, ((0.5, 1.0), (6.0, 0.4)), ((2.5, 1.0), (5.625, 0.4))), # Reserved
infer_type=False,
help="Natural breaks goodness of fit/split threshold to split "
"a peak. Specify as tuples of (log10(area), threshold).",
)

peak_split_filter_wing_width = straxen.URLConfig(
default=70,
infer_type=False,
help="Wing width of moving average filter for " "low-split natural breaks",
)

peak_split_min_area = straxen.URLConfig(
default=40.0,
infer_type=False,
help="Minimum area to evaluate natural breaks criterion. " "Smaller peaks are not split.",
)

peak_split_iterations = straxen.URLConfig(
default=20, infer_type=False, help="Maximum number of recursive peak splits to do."
)

diagnose_sorting = straxen.URLConfig(
track=False,
default=False,
infer_type=False,
help="Enable runtime checks for sorting and disjointness",
)

gain_model = straxen.URLConfig(
infer_type=False, help="PMT gain model. Specify as URL or explicit value"
)

gain_model_mc = straxen.URLConfig(
infer_type=False, help="PMT gain model. Specify as URL or explicit value"
)

tight_coincidence_window_left = straxen.URLConfig(
default=50,
infer_type=False,
help="Time range left of peak center to call a hit a tight coincidence (ns)",
)

tight_coincidence_window_right = straxen.URLConfig(
default=50,
infer_type=False,
help="Time range right of peak center to call a hit a tight coincidence (ns)",
)

n_tpc_pmts = straxen.URLConfig(type=int, help="Number of TPC PMTs")

n_top_pmts = straxen.URLConfig(type=int, help="Number of top TPC array PMTs")

sum_waveform_top_array = straxen.URLConfig(
default=True, type=bool, help="Digitize the sum waveform of the top array separately"
)

saturation_correction_on = straxen.URLConfig(
default=True, infer_type=False, help="On off switch for saturation correction"
)

saturation_reference_length = straxen.URLConfig(
default=100,
infer_type=False,
help="Maximum number of reference sample used " "to correct saturated samples",
)

saturation_min_reference_length = straxen.URLConfig(
default=20,
infer_type=False,
help="Minimum number of reference sample used " "to correct saturated samples",
)

peaklet_max_duration = straxen.URLConfig(
default=int(10e6), infer_type=False, help="Maximum duration [ns] of a peaklet"
)

channel_map = straxen.URLConfig(
track=False,
type=immutabledict,
help="immutabledict mapping subdetector to (min, max) " "channel number.",
)

hit_min_amplitude = straxen.URLConfig(
track=True,
infer_type=False,
default="cmt://hit_thresholds_tpc?version=ONLINE&run_id=plugin.run_id",
help="Minimum hit amplitude in ADC counts above baseline. "
"Specify as a tuple of length n_tpc_pmts, or a number,"
'or a string like "pmt_commissioning_initial" which means calling'
"hitfinder_thresholds.py"
"or a tuple like (correction=str, version=str, nT=boolean),"
"which means we are using cmt.",
)

def infer_dtype(self):
return dict(
peaklets=strax.peak_dtype(
n_channels=self.n_tpc_pmts,
digitize_top=self.sum_waveform_top_array,
),
lone_hits=strax.hit_dtype,
)

def setup(self):
if self.peak_min_pmts > 2:
# Can fix by re-splitting,
Expand Down Expand Up @@ -422,53 +286,6 @@ def compute(self, records, start, end):

return dict(peaklets=peaklets, lone_hits=lone_hits)

def natural_breaks_threshold(self, peaks):
"""
Pasted from https://github.com/XENONnT/straxen/blob/5f232eb2c1ab39e11fb14d4e6ee2db369ed2c2ec/straxen/plugins/peaklets/peaklets.py#L332-L348
"""
rise_time = -peaks["area_decile_from_midpoint"][:, 1]

# This is ~1 for an clean S2, ~0 for a clean S1,
# and transitions gradually in between.
f_s2 = 8 * np.log10(rise_time.clip(1, 1e5) / 100)
f_s2 = 1 / (1 + np.exp(-f_s2))

log_area = np.log10(peaks["area"].clip(1, 1e7))
thresholds = self.peak_split_gof_threshold
return f_s2 * np.interp(log_area, *np.transpose(thresholds[2])) + (1 - f_s2) * np.interp(
log_area, *np.transpose(thresholds[1])
)

@staticmethod
def clip_peaklet_times(peaklets, start, end):
straxen.plugins.peaklets.Peaklets.clip_peaklet_times(peaklets, start, end)

@staticmethod
def create_outside_peaks_region(peaklets, start, end):
"""Creates time intervals which are outside peaks.
:param peaklets: Peaklets for which intervals should be
computed.
:param start: Chunk start
:param end: Chunk end
:return: array of strax.time_fields dtype.
"""
outside_peaks = straxen.plugins.peaklets.Peaklets.create_outside_peaks_region(
peaklets, start, end
)
return outside_peaks

@staticmethod
def add_hit_features(hitlets, hit_max_times, peaklets):
"""Create hits timing features :param hitlets_max: hitlets with only
max height time.
:param peaklets: Peaklets for which intervals should be
computed.
:return: array of peaklet_timing dtype.
"""
straxen.plugins.peaklets.Peaklets.add_hit_features(hitlets, hit_max_times, peaklets)


@numba.jit(nopython=True, nogil=True, cache=False)
def peak_saturation_correction(
Expand Down
126 changes: 1 addition & 125 deletions saltax/plugins/records.py
View file
Original file line number Diff line number Diff line change
@@ -1,4 +1,3 @@
from immutabledict import immutabledict
from straxen.plugins.records.records import count_pulses
import numpy as np
import strax
Expand All @@ -11,7 +10,7 @@


@export
class SPulseProcessing(strax.Plugin):
class SPulseProcessing(straxen.PulseProcessing):
"""
Split raw_records into:
- (tpc) records
Expand All @@ -33,131 +32,8 @@ class SPulseProcessing(strax.Plugin):
"""

__version__ = "0.0.2"

parallel = "process"
rechunk_on_save = immutabledict(records=False, veto_regions=True, pulse_counts=True)
compressor = "zstd"

depends_on = ("raw_records", "raw_records_simu")

provides = ("records", "veto_regions", "pulse_counts")
data_kind = {k: k for k in provides}
save_when = immutabledict(
records=strax.SaveWhen.TARGET,
veto_regions=strax.SaveWhen.TARGET,
pulse_counts=strax.SaveWhen.ALWAYS,
)

hev_gain_model = straxen.URLConfig(
default=None, infer_type=False, help="PMT gain model used in the software high-energy veto."
)

baseline_samples = straxen.URLConfig(
default=40,
infer_type=False,
help="Number of samples to use at the start of the pulse to determine " "the baseline",
)

# Tail veto options
tail_veto_threshold = straxen.URLConfig(
default=0,
infer_type=False,
help=(
"Minimum peakarea in PE to trigger tail veto."
"Set to None, 0 or False to disable veto."
),
)

tail_veto_duration = straxen.URLConfig(
default=int(3e6), infer_type=False, help="Time in ns to veto after large peaks"
)

tail_veto_resolution = straxen.URLConfig(
default=int(1e3),
infer_type=False,
help="Time resolution in ns for pass-veto waveform summation",
)

tail_veto_pass_fraction = straxen.URLConfig(
default=0.05, infer_type=False, help="Pass veto if maximum amplitude above max * fraction"
)

tail_veto_pass_extend = straxen.URLConfig(
default=3,
infer_type=False,
help="Extend pass veto by this many samples (tail_veto_resolution!)",
)

max_veto_value = straxen.URLConfig(
default=None,
infer_type=False,
help="Optionally pass a HE peak that exceeds this absolute area. "
"(if performing a hard veto, can keep a few statistics.)",
)

# PMT pulse processing options
pmt_pulse_filter = straxen.URLConfig(
default=None, infer_type=False, help="Linear filter to apply to pulses, will be normalized."
)

save_outside_hits = straxen.URLConfig(
default=(3, 20),
infer_type=False,
help="Save (left, right) samples besides hits; cut the rest",
)

n_tpc_pmts = straxen.URLConfig(type=int, help="Number of TPC PMTs")

check_raw_record_overlaps = straxen.URLConfig(
default=True,
track=False,
infer_type=False,
help="Crash if any of the pulses in raw_records overlap with others " "in the same channel",
)

allow_sloppy_chunking = straxen.URLConfig(
default=False,
track=False,
infer_type=False,
help=(
"Use a default baseline for incorrectly chunked fragments. "
"This is a kludge for improperly converted XENON1T data."
),
)

hit_min_amplitude = straxen.URLConfig(
track=True,
infer_type=False,
default="cmt://hit_thresholds_tpc?version=ONLINE&run_id=plugin.run_id",
help="Minimum hit amplitude in ADC counts above baseline. "
"Specify as a tuple of length n_tpc_pmts, or a number,"
'or a string like "pmt_commissioning_initial" which means calling'
"hitfinder_thresholds.py"
"or a tuple like (correction=str, version=str, nT=boolean),"
"which means we are using cmt.",
)

def infer_dtype(self):
# Get record_length from the plugin making raw_records
self.record_length = strax.record_length_from_dtype(
self.deps["raw_records"].dtype_for("raw_records")
)

dtype = dict()
for p in self.provides:
if "records" in p:
dtype[p] = strax.record_dtype(self.record_length)
dtype["veto_regions"] = strax.hit_dtype
dtype["pulse_counts"] = straxen.pulse_count_dtype(self.n_tpc_pmts)

return dtype

def setup(self):
self.hev_enabled = self.hev_gain_model is not None and self.tail_veto_threshold
if self.hev_enabled:
self.to_pe = self.hev_gain_model
self.hit_thresholds = self.hit_min_amplitude

def compute(self, raw_records, raw_records_simu, start, end):
if self.check_raw_record_overlaps:
straxen.check_overlaps(raw_records, n_channels=3000)
Expand Down

0 comments on commit 7e159ee

Please sign in to comment.