experipy 0.1.1

A framework for writing and running Computational Science experiments

A framework for writing and running Computational Science experiments.

Experipy provides a composable grammar for automatically writing scripts and a tool which can then execute them. Support for PBS script based queueing is (mostly) implemented.

A Simple Example

from experipy.exp       import Experiment
from experipy.grammar   import Executable

exp = Experiment(Executable("echo", ["Hello World"]),
                 expname="test",
                 destdir="results")
exp.run()

This will run the program echo with the argument Hello World in a directory in /tmp, writing the output and error, along with timing information, to the directory results. Directories will be created as needed. A complete example showing how to write an experiment for a Python script can be found in test/runtest.py.

Components

experipy.grammar

Contains abstractions for describing program executables, their argurments, and relationships between them such as pipelines.

experipy.system

Contains a number of standard tools such as cp, mkdir, and the Python interpreter described as elements within the grammar.

experipy.exp

Contains the Experiment class, instances of which accepts a composition in the grammar, and can generate and execute a shell script from it.

corsys

Tools used by the Corsys research group at UTK, will likely be separated at some point, but serves as a useful example (Heavily WIP)

Features In The Works

• Develop a means for defining configurations

• Expand experipy.system to include more standard command line tools

• Beef up PBS script options