Running quTARANG

Packages Required

One can run quTARANG on a single CPU as well as a single GPU.

You need the following libraries to run quTARANG:

pathlib, h5py, numpy (to run code on CPU), cupy (to run code on GPU).

You can install the above package by using the following commands:

pip install pathlib
pip install h5py
pip install numpy

cupy library is required for GPU run. In order to install cupy you must first install a compatible version of CUDA for your GPU. Then install the cupy library for your cuda version. For more details: https://docs.cupy.dev/en/stable/install.html

Running the code

To run a simulation:

1. Import the required libraries

   from quTARANG import Params, GPE
   import numpy as xp
   # import cupy as xp # For GPU run
   

2. Set the parameters

   Create an instance of the Params class and set the parameters according to your need. The parameters have been detailed in the documentation. Example:

   # Create an instance of the Params class for storing parameters.
   par = gpe.Params(N = [64, 64, 64],
                L = [16, 16, 16],
                g = 0.1,
                dt = 0.001,
                tmax = 5,
                rms = [True, 0, 100])
   

3. Initiate GPE class

   Create an instance of the GPE class by passing the Params instance created previously.

   # Create an instance of the GPE class.
   G = gpe.GPE(par)
   

4. Set initial conditon

   You can give initial condition in terms of wavefunction and potential by defining their functions and passing them to the function set_init.

   # Set wavefunction
   wfc = lambda x, y=0, z=0: (1/xp.pi**(1/4)) * xp.exp(-(x**2/2 + y**2/2 + z**2/2))
   
   # Set potential
   pot = lambda x, y=0, z=0: (x**2 + y**2 + z**2)/2
   
   G.set_init(wfc, pot)
   

   wfc function will be used to set the initial wavefunction and pot variable will be used to set the initial potential.

5. Start the simulation:

   G.evolve()
   

The results are stored as hdf5 files in the cwd or the path set by the user in the Params instance.