Welcome to researchscripts’s documentation!

agg2lammps

Write ‘aggcar’ (positions + velocities) to a lammps file, for visualizing heat dissipation

analysis

analysis.analyzeFragments(datadir, **kwargs)

Pass in name and shallow kwargs if needed for utils.readStruct function

analysis.coordLabeller(atoms, fullCoordinations={'F': 1, 'H': 1, 'N': 3, 'Si': 4}, minAngles={'F': 360, 'H': 360, 'N': 109.5, 'Si': 90}, maxBonds_per_element={'F': 1, 'H': 1, 'N': 4, 'Si': 6}, angle_tolerance=0.15, bond_tolerance=0.25, minz=0)

Takes a structure, returns two dictionaries, the keys of which are identical: the index of the atom for which the statistic is calculated relativeCoordinations: -1 if atom i is undercoordinated, 1 if overcoordinated, 0 else bonds: list of bonds for atom i Defaults for angle and bond tolerance based on amorphous SiN

analysis.getBondSubset(data, elem)

Requires data pd df that has columns ‘geom’, ‘coordlabels’, ‘wantedIndices’ Returns pd Series with the subsetted bond counts of (A, elem) where A should be set by the input wantedIndices

analysis.plotBondDistributionComposition(data, a='Si', b='N')

Requires data, a pd df with ‘geom’, ‘coordlabels’, and (optionally) ‘wantedIndices’ Gives overall distribution and percent makeup of elements A and B in that distribution

cart2dir

Performs some kind of parsing on POSCAR file!…?

Erik won’t tell me what this does, so all I can say is that it parses the file passed in and writes some modified version to an outfile.

(it translates cartesian positions to direct positions :P)

Args:

POSCAR - filepath for POSCAR file

Returns:

None; writes results to “{POSCAR}.out” based on the POSCAR filepath provided as an argument.

energyanimator

energyanimator.main(filename='OSZICAR', outputname='energy', interval=16.666666666666668)

Calls OSZICARsearch to generate the potential and kinetic energy arrays; sums the two, and creates an animation

energyanimator.oszicarSearch(filename='OSZICAR', term='E0')

Searches an OSZICAR for the specified statistic, and returns a numpy array of values for that statistic from each ionic loop.

gen2pos

Convert from .gen to POSCAR

genvis

View .gen file

input_generation

input_generation.gridVasp(adsorbate, slab, directory='tempout/', spacing=0.2)

Takes in slab, adsorbate, and desired output directory. Writes numbered POSCAR files to output directory with adsorbate placed in grid pattern

Args:

adsorbate: either path (str) for POS/CONTCAR or Atoms obj slab: either path (str) for POS/CONTCAR or Atoms obj directory: path (str) for desired output location spacing: spacing between grid points (eg, 0.2 is a 5x5 grid)

Returns:

None

input_generation.randomGrid(adsorbate, slab, h=2, outputDir='tempout/', numDirs=10, runsPerDir=17, shallow=False, kind='gen')

Produces a collection of structures with adsorbate randomly placed on given slab. Writes (.gen or vasp) outputs to desired directory

Args:

adsorbate: Atoms obj of adsorbate slab: Atoms obj of slab h: height of adsorbate (above max slab position) outputDir: Path (str) for desired output location. numDirs: Number of batches. Defaults to 10. runsPerDir: Number of sims per batch. Defaults to 17. shallow: If all runs to be at one directory level. Defaults to False. kind: output desired, from (gen, vasp). Defaults to gen.

Returns:

None

input_generation.randomGridMultiple(n, adsorbate, slab, h=2, outputDir='tempout/', numDirs=10, runsPerDir=17, shallow=False, kind='gen', minDist=1)

Produces a collection of structures (of size numDirs * runsperDir; each has n adsorbates randomly placed on given slab. Writes (.gen or vasp) outputs to desired directory

Args:

n: number of adsorbates per slab adsorbate: Atoms obj of adsorbate slab: Atoms obj of slab h: height of adsorbate (above max slab position) outputDir: Path (str) for desired output location. numDirs: Number of batches. Defaults to 10. runsPerDir: Number of sims per batch. Defaults to 17. shallow: If all runs to be at one directory level. Defaults to False. kind: output desired, from (gen, vasp). Defaults to gen. minDist: minimum distance between placed adsorbates (in Å)

Returns:

None

krr_utils

krr_utils.convertAdsorbateToHe(struct, centerIndex, molIndices, height=None)

Preprocess final relaxed adsorption structures; replace adsorbate with He

Args:

• struct: total structure (Atoms object)
• centerIndex: index of central atom (where He will be) (int)
• molIndices: list of indices to delete from the slab
• height(float) : height of He to be placed

Returns:

• output: Atoms object with He representing the location of the adsorbate

krr_utils.getSOAPs(geometries, species, rcut=5, nmax=10, lmax=9, sigma=0.1, periodic=True, crossover=True, sparse=False)

Takes a Series of geometries with one He present,

returns SOAP representation of the chemical environment of He for each item

Assumes any given structure in geometries has the same collection of elements

as all the other structures

Assumes any given structure in geometries has the same number of atoms as all

the other structures

Input:

geometries: Series of Atoms objects; each must contain exactly 1 He atom rcut, nmax, lmax, sigma, periodic, crossover, sparse: SOAP parameters

Output:

output: Series of SOAP matrices, each corresponding to the appropriate index

krr_utils.predictz(surf, x, y, zmodel, species)

surf: bare substrate x, y: position at which to place adsorbate zmodel: model object (which takes in a dataframe that’s n*p, p = #SOAP features)

returns a predicted z value, based on sum of regionalMaxHeight and zmodel outcome

krr_utils.regionalMaxHeight(surf, x, y, R=2.2)

Returns the max height of any atom in a region of radius R around x,y R is default 2.2 from some experiments I did

krr_utils.rulez(surf, x, y)

surf: bare substrate x, y: position at which to place adsorbate

returns predicted z value

latticeconstmod

Modify lattice constants without modifying atom positions, useful for adding vacuum

measure_damage_depth

Meausre depth of damage induced by ion bombardment

pbcprettify

Rewrap periodic images that had jumped in a trajectory for nicer visualization I think

periodicprocess

Not sure what this does honestly

placeadsorbates

placeadsorbates.getSOAPs(geometries, rcut=5, nmax=10, lmax=9, sigma=0.1, periodic=True, crossover=True, sparse=False)

Takes a Series of geometries with one He present,

returns SOAP representation of the chemical environment of He for each item

Assumes any given structure in geometries has the same collection of elements

as all the other structures

Assumes any given structure in geometries has the same number of atoms as all

the other structures

Input:

geometries: Series of Atoms objects; each must contain exactly 1 He atom rcut, nmax, lmax, sigma, periodic, crossover, sparse: SOAP parameters

Output:

output: Series of SOAP matrices, each corresponding to the appropriate index

placeadsorbates.main(basename)

Perform ML-based isotherm seeding.

Args:

basename: name of base slab

placeadsorbates.predictz(surf, x, y, zmodel='zmodel.pkl')

surf: bare substrate x, y: position at which to place adsorbate zmodel: Path object pointing to the model pickle

postprocessing

postprocessing.postprocessAggregated(simindices, directory='../')

Takes in a list of indices, corresponding to the bombardment trials to analyze Looks for files named aggregated_{bomb,quench,eq}$i in directory specified.

postprocessing.postprocessResults(directory='../')

Takes in a list of indices, corresponding to the bombardment trials to analyze Looks for files named results$i{bomb,quench,eq}.csv in directory specified. Returns list of 3 dfs; each one has elements and keys

random_velos

Rewrite a given set of CONTCARs to give bombarding ion random velocity direction

refstructs

A collection of reference structures stored as Atoms objects.

Includes:

• mef
• cf4
• amorphous
• xtl_n
• xtl_si
• xtl2x2
• xtl2x2_sifterm
• heavy_bomb
• bulk222
• annealed

seed_monolayer

Designed to work on the cluster, seeding a monolayer of some given adsdorbate based on KRR models

structure

structure.getslab(struct)

Input:

struct: structre from which we will trim unbound species (.gen file)

Output:

baseslab: structure with unbound species trimmed

structure.getslabs(data, directory, useInputs=False)

Utility for getting and writing slab files from readData (utils.py) function data is the df from readData function or any df with (struct, in) and (struct, out) columns

translatepos

Translate a POSCAR some amount in x,y,z directions for easier visualization

transmute

transmute.transmute(threshold, transmute=True, target='C', slabElems=['Si', 'N'], surfDepth=6, simple=True, numStructs=-1, numOut=-1, **kwargs)

Find undercoordinated surface/near-surface atoms and transmute them to C

Args:

• threshold (k): maxiumum number of bonds to be considered “undercoordinated” (should be 1 or 2)
• transmute: whether to transmute elements or not
• slabElems: elements to consider as part of the ‘core slab’; only these will be transmuted
• surfDepth: depth of atoms to consider below surface; surface defined as max z coord among slabElems
• simple: Use basic bond counting as implemented in ASE, or advanced with coordLabeller from dependencies (simple=False is UNTESTED!!!)
• numStructs: Number of structures to consider
• numOut: Number of structures to output
• kwargs: to be passed to the readStructs function from dependencies

transmute.write_transmuted(outdir, numout, numrep)

Write results of transmute function into some directory Args: numout: the top numout structures to output numrep: the number of times to replicate each of the above structures, with random adjustment of last atom position (should be Ar for bomb) files written in input$i-$j.gen format

trim

Designed to work on the cluster, removing ‘ejected species’ after each bomb or quench step

trimandseed

Designed to work on the cluster, removing ‘ejected species’ after each equilibration step seeding Ar

utils

utils.KE(v_tot)

Returns KE of Ar+ in eV given total velocity

utils.calculateVASPtemp(velos, nfixed, POSCAR)

Take in a 3N matrix of velocities, subtract of nfixed from DOF POSCAR (Atoms object) used for getting masses from atoms Return T as calculated by VASP

utils.flatten(l)

flattens a 2d list

utils.normalize(y, x)

Takes y, x of data and returns normalized y

utils.readData(outdir, indir, useRemoval=True, useFrags=True, useBonds=True, wrapStructs=True)

Specifically made for data with BOTH input and output structures Read structures; append removal data, fragment data, and bond data optionally Wrap structures optionally

Returns:

df of the above with columns (struct, in), (struct, out), optionally, (frags, <fragnames>), (bonds, <bondnames>), (removal, <removedelems>) where each <x> corresponds to a column label for a specific subgroup (ie, NH3, N-H, or NH3 resp.)

utils.readStructs(datadir, shallow=True, name='output')

Currently designed for output from single layer directory trees. Reads in final adsorption geometries and energy data, returns dataframe with geometry and energy data

Input:

datadir: string that points to directory containing the following:

• convergence: each line i has convergence status of run i
• energies: each line i has total energy and ads energy from run i
• output{indices}.gen: final geometries for each index

slabEnergy: energy of slab adsorbateEnergy: energy of the adsorbate in the system

Returns:

output: pd Dataframe with:

• index: indices for runs that worked
• geometry: final geometry of run
• total energy: raw energy from file
• adsorption energy: energy as adjusted by adsorbate_energy

utils.v_from_KE(E)

Returns v(z) of Ar+ in eV given KE

visualization

visualization.getabBondcountStructure(data, idx, element)

Gets a struture with ‘charges’ equal to nbonds between a (fixed) and b(element) data needs geom, coordlabels, and (optionally) wantedIndices columns geom is Atoms object of structure coordlabels is a raw output from the coordlabeller function (relcoords and raw bonds) element is desired secondary element (primary element determined by input) Calls view() on resulting geometry Returns the structure

visualization.plotElemDist(data, targetElem='C', latticeElems=['Si', 'N', 'H'], nbins=25, stacked=False)

Plot distribution of element within slab, data should be arraylike collection of stuctures

visualization.show_atoms_grid(data, rotation='-0x, 0y, 0z', save=False, filename='grid_configs')

Where data is list of Atoms objects

visualization.viewStructs(name, directory, kind='gen')

View collection of structures as a “trajectory”

Args:

• name (str): substring unique to structures (.gen, POSCAR, slab, etc)
• directory (str): Directory where the structures live
• kind: kind of output froim list of (vasp, gen)

Opens viewer with loaded trajectory (if remote, need X server)

Indices and tables

• Index
• Module Index
• Search Page