How to use Yambo:advanced usage

From The Yambo Project
Jump to: navigation, search

This module presents examples of more advanced usage of the Yambo command line interface.

Prerequisites

Previous modules

You will need:

  • The SAVE databases for bulk hBN (Download here)
  • The yambo executable

Lowercase options: input file generation

Yambo reads various parameters from existing database files and/or input files and use them to suggest values or ranges. Let's illustrate this by calculating the static screening for bulk hBN.

$ cd YAMBO_TUTORIALS/hBN/YAMBO
$ yambo        (initialize, if you haven't already done so)
$ yambo -b

and look at the generated input file. Inside you will see a suggested range of bands, which Yambo guesses from the core ns.db1 database.

% BndsRnXd
   1 | 100 |                 # [Xd] Polarization function bands
%

Change these values to 8 | 9 | , save and exit the editor. Now launch the input file generator again:

$ yambo -b

and notice that your chosen values remain: they are simply read from the yambo.in file.

Now run the calculation. To keep things neat, add the -J TEST1 option, which directs all output to a directory TEST1.

$ yambo -J TEST1

After a few seconds, the code finishes. If you now look inside the TEST1 folder, you will see several new databases:

$ ls TEST1
ndb.dip_iR_and_P ndb.em1s [...]

Delete the input file yambo.in, and generate the input file for a different task: optical response in the independent particle approximation.

$ rm yambo.in
$ yambo -J TEST1 -o c

Looking inside the input file you will find:

% BndsRnXd
 8 |  9 |                   # [Xd] Polarization function bands
%

Here, Yambo has read from the ndb.dip_iR_and_P database and suggests to use the previous values. If at this stage, you decide to change these values (increase the range), Yambo will recompute the relevant database.

Uppercase options: command line options

Here is an example of how to combine several uppercase command line options. Starting from the usual folder containing the SAVE, do

$ mkdir TEST2
$ cd TEST2
$ yambo -F yambo.in_IP -I ../ -C q100 -J 3Ry -V RL -o c      and change to
FFTGvecs= 3            Ry    # [FFT] Plane-waves
$ yambo -F yambo.in_IP -I ../ -C q100 -J 3Ry -V RL 

You will notice

<---> [WF] Performing Wave-Functions I/O from ..//SAVE
<---> Dipoles: P and iR (T): |########################################| [100%] --(E) --(X)

with the ndb.dip_iR_and_P databases written in 3Ry, and output files inside the folder q100. Now select a different q-component and run again:

$ yambo -F yambo.in_IP -I ../ -C q001 -J 3Ry -V RL -o c      and change to
% LongDrXd
0.000000 | 0.000000 | 1.000000 |        # [Xd] [cc] Electric Field
%
$ yambo -F yambo.in_IP -I ../ -C q001 -J 3Ry -V RL 

For this run, yambo reads the database from the 3Ry folder, does a quick calculation, and writes the outputs inside q001.

The overall result is very neatly organised. Remember to check the report file to verify which databases are being WR = written to and RD = read from. Of course, it is up to the user to decide which options they like best.

Links