snap-combine 0.8

SuperNova Async Pipeline: combining

snap-combine

Code for combining SN significance from several detectors/experiments.

This package contains plugins for SNAP framework.

Documentation

Installation

python -m pip install snap-combine

Plugins

This package contains plugins which can be used in new snap pipeline configurations.

Monitoring: snap.util.monitor

• tqdm_ticker: a provides tqdm-based ticker, which counts the data flow and data rate through current step.

Generating client data: snap.client.fake

• sample_ts: Generates the neutrino interaction timestamps, using the given rate. Can simulate the supernova signal at the given time.

Processing client data: snap.client

• sigcalc.ShapeAnalysis: calculate supernova significance using shape analysis
• setId: change the datablock ID

Combination: snap.combine:

• Buffer: accumulate the data to synchrnize before combining
• methods.Fisher: combine data using Fisher's combination method
• methods.Stouffer: combine data using Stouffer's combination method

Triggering: snap.util.threshold

• Threshold: select and forward only portions of data with significance above given threshold

Misc: snap.util

• dump_to_file: Dump data to given file

Example

Example configurations provided the examples dir use these steps to

• client_sender.yml: generate client data, calculate significance and send it (via ipc zeromq socket) to combiner node
• combine.yml: receive client significance time series, combine them and apply thresholds.

On client side run

snap client_sender.yml -n node1

Optionally in another session:

snap client_sender.yml -n node2

In another session (combination side):

snap combine.yml

You should see the monitoring ticks for all nodes. Also you should see the files with the data output: data_received.dat,data_combined.dat and data_triggered.dat.

After about 120s from the client start, a supernova signal will be emitted, so the significance becomes elevated, and the Trigger counter will advance.