ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Updated i-pi example

Browse Source
This commit is contained in:
Michele Ceriotti
2024-06-04 23:48:17 +02:00
parent f5253eb926
commit 8a94faee58
2 changed files with 45 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
examples/PACKAGES/ipi/README → examples/PACKAGES/ipi/README.md
View File

@ -1,17 +1,17 @@
i-PI path integral interface examples i-PI path integral interface examples
===================================== =====================================
This folder contains a couple of examples to run LAMMPS as a client, This folder contains an example to run LAMMPS as a client,
exchanging information on the atomic configurations, energy and forces exchanging information on the atomic configurations, energy and forces
with the i-PI Python interface with the [http://ipi-code.org](i-PI Python interface). These
[http://epfl-cosmo.github.io/gle4md/index.html?page=ipi]. These
examples require a working copy of i-PI and compiling LAMMPS in a UNIX examples require a working copy of i-PI and compiling LAMMPS in a UNIX
environment. Note that a copy of i-PI is provided with LAMMPS, in the environment.
tools/i-pi directory. i-PI can be installed from source or from `pip`. Please refer to the
documentation for up-to-date installation instruction.
Note that the i-PI examples listed here are designed for the public V1.0 Note that the i-PI examples listed here have been tested to work with the
version of i-PI. Refer to the LAMMPS examples distributed with i-PI if you 3.0 version of i-PI. Refer to the LAMMPS examples distributed with i-PI if you
are using a development version. are using a different version.
Path integral simulation of graphene Path integral simulation of graphene
------------------------------------ ------------------------------------
@ -25,7 +25,7 @@ and the format of the output.
How to run i-PI How to run i-PI
--------------- ---------------
You should have a relatively recent (>=2.5) version of Python and Numpy, You should have a relatively recent (>=3.5) version of Python and Numpy,
and the public version of i-PI. You can then run i-PI by executing and the public version of i-PI. You can then run i-PI by executing
```bash ```bash
@ -33,7 +33,7 @@ and the public version of i-PI. You can then run i-PI by executing
``` ```
In a separate terminal, then, you should run LAMMPS compiled to provide In a separate terminal, then, you should run LAMMPS compiled to provide
fix_ipi functionalities. `fix_ipi` functionalities.
```bash ```bash
$LAMMPS -in in.graphene $LAMMPS -in in.graphene
@ -41,6 +41,3 @@ fix_ipi functionalities.
You can run multiple instances of LAMMPS if you want to exploit the You can run multiple instances of LAMMPS if you want to exploit the
parallelism over the path integral beads. parallelism over the path integral beads.

64
examples/PACKAGES/ipi/i-pi_input.xml
View File

@ -1,31 +1,37 @@
<simulation verbosity="medium"> <simulation verbosity='medium'>
<initialize nbeads='8'> <output prefix='simulation'>
<file mode='xyz' units="angstrom"> i-pi_positions.xyz </file> <properties stride='5' filename='out' flush="10" > [ step, time{picosecond}, conserved{kelvin}, temperature{kelvin}, kinetic_cv{kelvin}, potential{kelvin}, pressure_cv{megapascal}] </properties>
<cell mode='manual' units="angstrom"> [ 51.8,0,0,0, 49.84,0,0,0, 200 ] </cell>
<velocities mode='thermal' units='kelvin'> 300 </velocities>
</initialize>
<output prefix='graphene'>
<properties stride='5' filename='out' flush="10" > [ step, time{picosecond}, conserved{electronvolt}, temperature{kelvin}, kinetic_cv{electronvolt}, potential{electronvolt}, pressure_cv{megapascal}] </properties>
<properties stride='5' filename='iso' flush="10" > [ isotope_tdfep(1.167;C), isotope_scfep(1.167;0) ] </properties> <properties stride='5' filename='iso' flush="10" > [ isotope_tdfep(1.167;C), isotope_scfep(1.167;0) ] </properties>
<trajectory stride='20' filename='pos' flush="100"> positions{angstrom}</trajectory> <properties stride='1' filename='kc'> [ kinetic_cv{kelvin}(C) ] </properties>
<checkpoint stride='2000' /> <trajectory stride='5' filename='pos' flush="100">positions{angstrom}</trajectory>
</output> <trajectory stride='5' filename='kin' flush="100"> kinetic_cv </trajectory>
<total_steps>1000</total_steps> <checkpoint stride='20000'/>
<total_time>128000</total_time> </output>
<prng><seed>8417</seed></prng> <total_steps>1000000</total_steps>
<forces> <prng>
<socket mode="unix"> <seed>31415</seed>
<address>graphene</address> </prng>
</socket> <ffsocket name='lammps' mode='unix'>
</forces> <address> graphene </address>
<ensemble mode='nvt'> </ffsocket>
<thermostat mode='gle'> <system>
<A shape='(5,5)'> <initialize nbeads='8'>
[ 1.124524713863e-3, 1.648702679619e-6, 6.970075857471e-5, -2.202066291263e-4, 1.401342873485e-3, -1.681700567912e-6, 5.197673899653e-10, 4.365423872046e-6, -1.200041116490e-6, 2.564577183580e-6, -8.965478630849e-5, -4.365423872046e-6, 8.218704940997e-6, 3.114246791997e-5, -6.044142906315e-5, -6.272281358913e-5, 1.200041116490e-6, -3.114246791997e-5, 1.612301941566e-4, 6.958958085115e-5, 1.318373360752e-3, -2.564577183580e-6, 6.044142906315e-5, -6.958958085115e-5, 1.872119364197e-3] <file mode='xyz'> i-pi_positions.xyz </file>
</A> <velocities mode='thermal' units='kelvin'> 300 </velocities>
</thermostat> </initialize>
<timestep units="femtosecond">1.0</timestep> <forces>
<temperature units='kelvin'>300</temperature> <force forcefield='lammps'> </force>
<fixcom> True </fixcom> </forces>
</ensemble> <motion mode='dynamics'>
<dynamics mode='nvt'>
<timestep units='femtosecond'> 1.0 </timestep>
<thermostat mode='langevin'>
<tau units='femtosecond'> 100 </tau>
</thermostat>
</dynamics>
</motion>
<ensemble>
<temperature units='kelvin'> 300 </temperature>
</ensemble>
</system>
</simulation> </simulation>