Event Sequence Analysis Library¶

Esal (“easel”) is a library for the descriptive statistical analysis and manipulation of event sequences and timelines. Esal is intended to be used for exploring event sequence data and preparing data for modeling, but does not do any modeling itself. Conceptually, Esal is a representation for a dataset of sequences / timelines and an associated set of meaningful operations (selection, counting, transformation).

Documentation¶

https://esal.readthedocs.io

Features¶

This project is in the early design and implementation stages. The implemented features are:

Event objects
Event sequence data structure for the efficient querying of event sequences
Interval objects
Allen’s interval algebra (constant-time implementation that uses the minimum number of comparisons)

The planned features are:

Selection and counting of sequences
Selection and counting of events
Sampling events and sequences
Temporal statistics
Reading/Writing various relational and flat representations

Some of the planned features have already been implemented but haven’t yet been organized into an official API. Feel free to look through the code. Suggestions are welcome.

Requirements¶

Python 3

Install¶

pip3 install [–user] https://github.com/cloudbopper/esal/archive/<name>.zip#egg=esal

Replace <name> with the name of the tag, branch, or commit you want to install, e.g. “master” or “v0.2.0”. If you don’t have a pip3, replace it with python3 -m pip. For more information, see the Pip documentation.

License¶

Esal is free, open source software. It is released under the MIT license. See the LICENSE file for details.

Concepts¶

See the package documentation for a conceptual overview of events and sequences.

Contact¶

Open an issue to report a bug or ask a question. To contribute, use the regular fork and pull request work flow.