Skip to content

Latest commit

 

History

History
165 lines (127 loc) · 7.96 KB

README.md

File metadata and controls

165 lines (127 loc) · 7.96 KB

MESA Python Operational Runtime Toolkit

A Python interface for the MESA stellar evolution code

Create, modify, run and share your MESA projects!

Features:

  • With MESA installed, anyone can run your MESA project using this Python toolkit. You only need to share your python project.

  • MESA-PORT can be used to create, clean, make, run and resume your MESA project.

  • This module also allows you to manipulate parameters in your inlist files. Your inputs will automatically be converted to the right data type and format for fortran.

  • Install MESA on Linux and macOS (ARM/M-series and Intel) with just this python package!

  • MESA-PORT can also run GYRE stellar oscillation code! See Usage.


Installation

pip install git+https://github.com/gautam-404/MESA-PORT.git

Usage

Note

The complete module and submodules are documented here. Below are some examples to get you started. There are a lot more features and method arguments available that are not covered here.

The ProjectOps class:

This class handle MESA operations. An object of this class allows you to create, clea, make, run, resume and delete your MESA project.

  • Creating a new MESA work directory:
    from mesaport import  ProjectOps
    
    ## Instantiate a new project
    proj = ProjectOps(name='work')   
    ## Default project name is 'work'. 
    
    ## Create a new project
    proj.create(overwrite=False, clean=False)    
    ## CLI is shown if no arguments are passed

Tip

Instead of single-star evolution, you can create a binary system or an astero project. This is done by passing boolean True for binary or astero arguments while instantiating the ProjectOps class.

  • Take control of your project; make, clean, run, resume and delete.
    proj.clean()
    proj.make()
    proj.run(silent=True, trace=None)                                    
    proj.resume("photo_name", silent=True, trace=None)
    proj.delete()                     

Tip

A list of MESA parameters can be passed to the trace argument to print their evolution in terminal along with age. Eg., trace=["log_L", "log_Teff"]

The MesaAccess class:

  • This class gives you access to the parameters in your inlist files. You can write, read, remove and set default values for inlist parameters.

    from mesaport import  MesaAccess
    
    star = MesaAccess("your_project")
    
    ## Write
    star.set(parameters, values)              
    ## Inputs parameters can be a string or a list of strings
    ## Input values can be a single value or a list of values
    ## If a list of values is passed, the length of the list must be equal to the length of the parameters list.
    
    ## While using the 'set' method, you can also pass a dictionary.
    star.set({"parameter1":value1, "parameter2":value2, "parameter3":value3})
    
    ## Read
    value = star.get(parameters)   
    ## Inputs parameters can be a string or a list of strings
    
    ## Delete
    star.delete(parameters)
    ## Inputs parameters can be a string or a list of strings
    
    ## Set to default
    star.setDefualt(parameters)
    • In addition to the above, you can also use the MesaAccess class object to load your custom inlists and other input files such as history_columns, profile_columns, run_star_extras, run_binary_extras, inlist_astero_search_controls and inlist_pgstar files.
    ### Path arguments can be a path or the name of a file in 'my_project' directory ###
    
    star.load_Extras("path/to/custom/run_star_extras_file")          ## Load custom run_star_extras.f90
    ## run proj.make() after loading extras. this allows the compiler to compile with the run_star_extras file
    star.load_InlistProject("/path/to/custom/inlist")                ## Load custom inlist_project 
    star.load_InlistPG("/path/to/custom/inlist")                     ## Load custom inlist_pgstar    
    star.load_HistoryColumns("path/to/custom/history_columns_file")  ## Load custom history_columns
    star.load_ProfileColumns("path/to/custom/profile_columns_file")  ## Load custom profile_columns
    star.load_InlistAsteroSearch("path/to/inlist")                   ## Load custom inlist_astero_search_controls
  • When working with a binary project, MesaAccess class object can be instantiated for the primary star, secondary star and the binary system. This allows you to manipulate input parameters for each star and the binary system separately.

    binary = MesaAccess("your_project", binary=True, target='binary')        ## For the binary system
    primary = MesaAccess("your_project", binary=True, target='primary')      ## For the primary star
    secondary = MesaAccess("your_project", binary=True, target='secondary')  ## For the secondary star
    
    ## Parameters can be accessed using the same methods as above
    ## For example:
    binary.set("binary_mass_ratio", 0.5)
    primary.set("profile_interval", 50)
    secondary.set("history_interval", 1)
    
    ## Load custom input files 
    primary.load_InlistProject("/path/to/custom/inlist")          ## Load custom 'inlist1'
    secondary.load_InlistProject("/path/to/custom/inlist")        ## Load custom 'inlist2'
    binary.load_InlistProject("/path/to/custom/inlist")           ## Load custom 'inlist_project' for the binary system
    binary.load_Extras("path/to/custom/run_binary_extras_file")   ## Load custom run_binary_extras.f90

Running GYRE

  • Stellar pulsation frequencies can be computed for MESA models using the GYRE code. This can be automated with MESA-PORT using the runGyre method of the ProjectOps class.
    proj.runGyre("gyre.in")  
    ## "gyre.in" can either be a path to your GYRE input file
    ## or it can also be the name of a file in your_project or your_project/LOGS directory
    
    ## If you want to run a GYRE input file for all profile data files in your LOGS directory, 
    ## pass files="all" as an argument.
    proj.runGyre("gyre.in", files="all")
    
    ## If you want to run GYRE for specific GYRE files, pass file names as an argument.
    proj.runGyre("gyre/input", files=["profile1.data.GYRE", "profile2.data.GYRE"])
  • You can also just pass the path to the directory containing the MESA profiles. It does not have to be the LOGS directory of a project.
    proj.runGyre("gyre.in", wdir="/path/to/profiles/dir", files="all")
  • GYRE can also be run for the primary or the secondary star in a binary system.
    proj.runGyre("gyre.in", target="primary")  ## Target can be "primary" or "secondary"

The Installer class:

  • This class allows you to install MESA on your Linux and macOS (ARM/M-series and Intel) systems.
    from mesaport import Installer
    
    Installer(version="latest", parentDir='where/to/install', cleanAfter=False )     
    ## version = "latest" will install the latest version available for your system.
    ## Available versions: 
    #     Linux: "23.05.1", "22.05.1", "15140" and "12778".
    #     macOS-Intel: "23.05.1", "22.05.1", "15140" and "12778".  
    #     macOS-ARM: "23.05.1", "22.05.1".
    
    ## The `cleanAfter` argument is set to False by default to allow re-running installation without removing downloaded files, this saves time when debugging a failed MESA build.
    It is possible to instantiate the Installer class without passing any arguments. The CLI will prompt you for the missing arguments. This is the easiest way to use the installer.
    CLI example

Warning

MESA-PORT is a work in progress. Please report any issues or bugs you encounter.