Skip to content

Commit

Permalink
Merge pull request #42 from iguinn/main
Browse files Browse the repository at this point in the history
Added more information to README
  • Loading branch information
iguinn authored Jan 4, 2024
2 parents 6c310b8 + 9fb002e commit edc960c
Showing 1 changed file with 18 additions and 2 deletions.
20 changes: 18 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,14 @@
# dspeed

```
_____ _________________________________________________________________
|| ____ _____ ____ __ `,_
|| / __ \/ ___/ / __ \ ___ ___ ____/ / | `-_
[] || [] [] [] [] / / / /\__ \ / /_/ // _ \ / _ \ / __ / [] [] [] '-----`-,_
====||===============/ /_/ /___/ // ____// __// __// /_/ /====================== ``--,_
|| /_____//____//_/ \___/ \___/ \__,_/ ``--,
|| ________ ________ )
\____||___/.-. .-.\______________________________________________________/.-. .-.\_____,,--''
==========='-'=='-'========================================================'-'=='-'=============
```
[![PyPI](https://img.shields.io/pypi/v/dspeed?logo=pypi)](https://pypi.org/project/dspeed/)
![GitHub tag (latest by date)](https://img.shields.io/github/v/tag/legend-exp/dspeed?logo=git)
[![GitHub Workflow Status](https://img.shields.io/github/checks-status/legend-exp/dspeed/main?label=main%20branch&logo=github)](https://github.com/legend-exp/dspeed/actions)
Expand All @@ -10,3 +19,10 @@
![GitHub pull requests](https://img.shields.io/github/issues-pr/legend-exp/dspeed?logo=github)
![License](https://img.shields.io/github/license/legend-exp/dspeed)
[![Read the Docs](https://img.shields.io/readthedocs/dspeed?logo=readthedocs)](https://dspeed.readthedocs.io)

# DSPeed
DSPeed (pronounced dee-ess-speed) is a python-based package that performs bulk, high-performance digital signal processing (DSP) of time-series data such as digitized waveforms. This package is part of the [pygama](https://github.com/legend-exp/pygama) scientific computing suite.

DSPeed enables the user to define an arbitrary chain of vectorized signal processing routines that can be applied in bulk to waveforms and other data provided using the [LH5-format](https://legend-exp.github.io/legend-data-format-specs). These routines can include numpy ufuncs, custom functions accelerated with numba, or other arbitrary functions. DSPeed will carefully manage file I/O to optimize memory usage and performance. Processing chains are defined using highly portable JSON files.

See the [online documentation](https://dspeed.readthedocs.io/en/stable/) for more information!

0 comments on commit edc960c

Please sign in to comment.