CTIP is a general, extensible tool for running batches of configurable jobs in parallel on a variety of environments from your local machine to a remote server.
By relying on the end-user to specify what to run and how to run it, CTIP is a very general solution for optimizing and experimenting on programs that use configuration files to determine behavior. CTIP provides the infrastructure for generating and passing configurations to user-adapted classes that setup and run a given program. It also provides tools for tracking job status when jobs are run locally.
See the INSTALL file for more detailed information.
Via pip:
$ pip install ctip
Configure ctip installation by setting paths to directories containing all user
defined Experiments
and Environments
:
$ ctip set experiment-dir ~/research/ctip-exp
$ ctip set environment-dir ~/research/ctip-env
For complete documentation of ctip, browse the docs folder or go to the future and visit a website I will make with all the documentation.
For a concise summary of all available commands, run:
$ ctip help
Run configurations for the Doodle
experiment in the Local
environment as
specified in a genfile named doodle_configs.gen
:
$ ctip run gen doodle_configs.gen --experiment Doodle --environment Local
Note: During development, avoid constant re-installation by using
python ctip-runner.py
instead of the ctip
command.
Please see the CHANGELOG for information about what has changed recently or the TODO list to see what's comming.
From the root ctip directory run:
$ py.test
If you discover any security related issues, please email me at [email protected] instead of using the issue tracker.
With that being said, any communication ctip does with the outside world is through user provided Environment classes and thus most security vulnerabilities will likely be the responsibility of the end-user.
Also sql injection through the optional 'where' clause is not only possible but is considered a feature. The only database malicious queries effect is the ctip database which only ever holds information you put in it... so go ahead.
Thanks to the Hintze Lab at Michigan State University for being a willing guinea pig.
The MIT License (MIT). Please see the LICENSE file for more information.