Speaker
Description
In pump-probe experiments at the Single Pulse Facility (SPF) at MAX IV, the dynamics of ultrafast phenomena are gauged by gathering 2D diffraction patterns at stochastically random varying delay-times, or ping values, between
the pump and probe pulses. To build statistics and aid in data reduction, data frames are sorted based on their ping value and binned together by summing the corresponding 2D patterns. However, these ping values arrive out of sequence.
Previously employed scripts for sorting and binning required to be run after data acquisition had been completed and could take longer than the experimental run. Additionally, there was no visual aid or guidance for choosing appropriate binning parameters a priori.
To help mitigate these shortcomings, we developed the Ping-Map, a python-based graphical user interface (GUI), communicating over ZMQ with an efficient C++ backend, which is capable of sorting and binning data frames from raw files, even as they are being written to.
Thus, the post-processing can be done during data acquisition, or after, on the order of seconds to minutes. The GUI allows for interactive selection of bin parameters and provides visual constructs that illustrate the progress of the binning, as well as intermediate
bin-averaged data frames and difference maps thereof, for real-time user-interaction.
The modularized backend allows for the binned data to be fed into further downstream post-processing, such as azint (see “Numerically Consistent Azimuthal Integration for 2D pixel detectors for orientated data" – submitted by Stuart Ansell).
[Note to organizers: if the latter is rejected, please remove the last sentence of this abstract.]
