m*2T

Transport properties from multi-valley density of states in the relaxation time approximation.

Directory structure

• docs: documentation of the code
• Mstar2t: Link to the repository of the Julia package implementing the computing unit of m*2T
  • src: source code
    • transport: Julia package which contains the methods to compute the transport coefficients
    • scattering: Julia package that implements different models for the relaxation time
    • plot: Julia package to plot the results of the calculations as a function of temperature and chemical potential
    • server: Julia package to implement the computing unit as a server
    • other: collection of utility functions
  • examples: collection of scripts using the Mstar2t Julia package
  • perf: script to test the performance of the code
• Interface: Python implementation of the client interface
  • GUI: graphical user interface source code
  • CLI: command-line interface source code

Requirements

• Julia >= 1.6
• python >= 3.7

Dependencies

Computing unit (Mstar2t.jl Julia package)

• ArgParse == 1.1
• CSV == 0.10
• CairoMakie == 0.10
• Colors == 0.4
• DataFrames == 1.4
• Distributions == 0.25
• Einsum == 0.4
• FastGaussQuadrature == 0.5
• GLMakie == 0.8
• HTTP == 1.7
• HypergeometricFunctions == 0.3
• JSON == 0.21
• LaTeXStrings == 1.3
• LinearAlgebra == 3.4
• Parameters == 0.12
• PlotUtils == 1.3
• PolyLog == 2.3
• Polynomials == 3.2
• QuadGK == 2.6
• SpecialFunctions == 2.1

Interface

• numpy
• pandas
• requests
• matplotlib
• colorama
• argparse
• PySide2==5.15

Installation

Since Mstar2t.jl is a standalone Julia package, it is included as a submodule of the software. Therefore, when cloning the repository, the flag --recurse-submodules must be included to download all the files:

HTTPS:

$ git clone --recurse-submodules https://github.com/marcofornari/etrasport.git

SSH:

$ git clone --recurse-submodules git@github.com:marcofornari/etrasport.git

Computing unit

$ cd Mstar2t
$ julia

Enter the Julia package manager (typing ]) and run the following commands:

(v1.6) pkg> activate .
(Mstar2t) pkg> instantiate

Interface

Unix/Mac:

$ cd Interface
$ python3.7 -m venv Interface
$ source Interface/bin/activate
(Interface) $ pip install --upgrade pip
(Interface) $ python -m pip install -r requirements.txt

Windows:

> cd Interface
> python3.7 -m venv Interface
> .\Interface\Scripts\activate
(Interface) > pip install --upgrade pip
(Interface) > python -m pip install -r requirements.txt

Usage

There are two options to run m*2t: GUI or Command Line Interface (CLI). Both ways are designed to automatically set up the correct environment for the computing server.

Usage (with GUI)

(Interface) $ cd Interface/GUI
(Interface) $ python run_gui.py

Usage (with CLI)

First run the server (computing unit)

(Interface) $ cd Interface/CLI
(Interface) $ python run_cli.py

On a new shell, send simulations requests to the server by running:

(Interface) $ cd Interface/CLI
(Interface) $ python compute.py -i <input_file> --<tensor_name> --<plot>

Note: before running a calculation, edit the results fullpath argument in the input_file. This path identifies the location where the results are exported and must be in the same machine in which the server is running.

Examples

GUI

(CLI Python interface)

(Interface) $ python compute.py -i data.txt --conductivity --seebeck --concentration --tplot
(Interface) $ python compute.py -i data.txt -esn --tplot

Help (CLI Python interface)

(Interface) $ python compute.py --help

usage: compute.py [-h] -i INPUTFILE [--conductivity] [--seebeck] [--thermal] [--concentration] [--tplot] [--muplot]

optional arguments:
  -h, --help            show this help message and exit
  -i INPUTFILE, --inputfile INPUTFILE path to input file
  --conductivity, -e    compute electrical conductivity
  --seebeck, -s         compute Seebeck coefficient
  --thermal, -k         compute thermal conductivity
  --concentration, -n   compute carrier concentration
  --tplot               Temperature plot of the tensors
  --muplot              Fermi level plot of the results

Troubleshooting

If the environment gets corrupted and when launching the server you get a LoadError: ArgumentError exception similar to this:

ERROR: LoadError: ArgumentError: <PackageName> is required but does not seem to be installed:
 - Run `Pkg.instantiate()` to install all recorded dependencies.

you may try the following steps (see below):

1. Remove the Manifest.toml file inside Mstar2t;
2. Open the julia REPL;
3. Enter the julia package manager (typing ]);
4. Activate the Mstar2t env;
5. Run resolve;
6. Run instantiate.

$ cd Mstar2t
$ rm Manifest.toml 
$ julia               _

_       _ _(_)_     |  Documentation: https://docs.julialang.org
  (_)     | (_) (_)    |
   _ _   _| |_  __ _   |  Type "?" for help, "]?" for Pkg help.
  | | | | | | |/ _` |  |
  | | |_| | | | (_| |  |  Version 1.6.0-rc2 (2021-03-11)
 _/ |\__'_|_|_|\__'_|  |  Official https://julialang.org/ release
|__/                   |

(@v1.6) pkg> activate .
(Mstar2t) pkg> resolve
(Mstar2t) pkg> instantiate

This should clean and rebuild the environment with the correct dependencies.