Metadata-Version: 1.0
Name: vivarium_cluster_tools
Version: 1.0.6
Summary: A set of tools for running simulation using vivarium on cluster.
Home-page: https://stash.ihme.washington.edu/projects/CSTE/repos/vivarium_cluster_tools
Author: The vivarium developers
Author-email: vivarium.dev@gmail.com
License: UNKNOWN
Description: Vivarium Cluster Tools
        =======================
        
        This package includes a tool that allows vivarium users to run simulations based on multiple scenarios conveniently.
        It requires an access to IHME cluster. To install this package, create or edit a file called ~/.pip/pip.conf which looks like this:
        
            | [global]
            | extra-index-url = http://dev-tomflem.ihme.washington.edu/simple
            | trusted-host = dev-tomflem.ihme.washington.edu
        
        
        This file tells the pip package management system to check with IHME's internal
        pypi server for packages.
        
        You can then install this package with
        
            | > source activate <env-name>
            | > pip install vivarium-cluster-tools
        
        In addition, this tool needs redis client and cython.
        
            | > conda install redis cython
        
        How to use
        -------------
        
        The Vivarium ecosystem uses YAML configuration files throughout, including vivarium-cluster-tools.
        To run the multiple scenarios, you need to make a separate branch file (YAML) in addition to a usual
        vivarium model specification YAML file. You can specify how to vary a certain parameter in this branch file.
        For example,
        
        ::
        
            input_draw_count: 25
            random_seed_count: 10
        
            branches:
              - intervention_name:
                   intervention_coverage: [0, 1]
                   efficacy: [0, .5, 1]
        
        This branch shows that you want to have 25 different input draws, 10 different random seeds and 2 different
        intervention coverages and 3 different intervention efficacy. This implies 25*10*2*3=1500 different simulation
        scenarios. You can run all 1500 simulations by
        
        ::
        
            psimulate run /path/to/your/model_configuration /path/to/your/branch_file
        
        As an optional parameter, you can specify the name of project as well as result directory to save the outputs
        and logs. To specify those optional arguments,
        
        ::
        
            psimulate run /path/to/your/model_configuration /path/to/your/branch_file -p project_name -o /path/to/output/
        
        If your psimulate run has any failed jobs from the previous runs, you can restart failed jobs by specifying
        which output directory includes the partially completed jobs by,
        
        ::
        
            psimulate restart /path/to/the/previous/results/
        
        
        
Platform: UNKNOWN
