Skip to content

Latest commit

 

History

History
38 lines (27 loc) · 1.45 KB

README.md

File metadata and controls

38 lines (27 loc) · 1.45 KB

SPA

SMC for Processor Analysis.

SPA is a statistics framework which allows users to use Statistical Model Checking (SMC)-based techniques to analyze numerical results. The technique is described in the paper Rigorous Evaluation of Computer Processors with Statistical Model Checking.

Installation

Conda: Installation Without Source (Recommended)

The file spa_env_no_source.yml provides a Conda environment file to set up the environment containing the SPA library. The following code will

  1. Download the required environment file
  2. Create a new Conda environment called spa
$ wget https://raw.githubusercontent.com/filipmazurek/spa-library/v0.0.1/spa_env_no_source.yml
$ conda env create -f spa_env_no_source.yml
$ conda activate spa

Conda: Installation From Source

This is recommended for further development of the SPA library.

We provided a Conda environment file to set up the correct environment.

$ git clone --depth 1 --branch v0.0.1 [email protected]:filipmazurek/spa-library.git
$ cd spa-library/
$ conda env create -f environment.yml
$ conda activate spa

This will install the SPA library in editable mode, so that changes in the repository will reflect in the installation.

Usage

Please refer to examples/spa-example.py for recommended usage of the SPA library, particularly the function property_use(). The main use of this library is to create confidence intervals.