pyiron_atomistics.atomistics.master.parallel.AtomisticParallelMaster

pyiron_atomistics.atomistics.master.parallel.AtomisticParallelMaster#

class pyiron_atomistics.atomistics.master.parallel.AtomisticParallelMaster(project, job_name)[source]#

Bases: ParallelMaster, AtomisticGenericJob

__init__(project, job_name)[source]#

Methods

`__init__`(project, job_name)
`animate_structure`([spacefill, show_cell, ...])	Animates the job if a trajectory is present
`animate_structures`([spacefill, show_cell, ...])	Animate a series of atomic structures.
`append`(job)	Append a job to the GenericMaster - just like you would append an element to a list.
`calc_md`([temperature, pressure, ...])
`calc_minimize`([ionic_energy_tolerance, ...])	param ionic_energy_tolerance: Maximum energy difference between 2 steps
`calc_static`()	Returns:
`check_if_job_exists`([job_name, project])	Check if a job already exists in an specific project.
`check_setup`()	Checks whether certain parameters (such as plane wave cutoff radius in DFT) are changed from the pyiron standard values to allow for a physically meaningful results.
`child_hdf`(job_name)	Find correct HDF for new children.
`clear_job`()	Convenience function to clear job info after suspend.
`collect_logfiles`()	Collect the log files of the external executable and store the information in the HDF5 file.
`collect_output`()	Collect the output files of the external executable and store the information in the HDF5 file.
`collect_structures`([filter_function])	Collects a copy of all structures in a compact `StructureStorage`.
`compress`([files_to_compress, files_to_remove])	Compress the output files of a job object.
`continue_with_final_structure`([job_type, ...])	param job_type:
`continue_with_restart_files`([job_type, job_name])	param job_type:
`convergence_check`()	Check if and all child jobs of the calculation are converged.
`copy`()	Copy the GenericJob object which links to the job and its HDF5 file
`copy_file_to_working_directory`(file)	Copy a specific file to the working directory before the job is executed.
`copy_template`([project, new_job_name])	Copy the content of the job including the HDF5 file but without the output data to a new location
`copy_to`([project, new_job_name, input_only, ...])	Copy the content of the job including the HDF5 file to a new location.
`create_child_job`(job_name)	Internal helper function to create the next child job from the reference job template - usually this is called as part of the create_jobs() function.
`create_job`(job_type, job_name[, ...])	Create one of the following jobs: - 'StructureContainer’: - ‘StructurePipeline’: - ‘AtomisticExampleJob’: example job just generating random number - ‘ExampleJob’: example job just generating random number - ‘Lammps’: - ‘KMC’: - ‘Sphinx’: - ‘Vasp’: - ‘GenericMaster’: - ‘ParallelMaster’: series of jobs run in parallel - ‘KmcMaster’: - ‘ThermoLambdaMaster’: - ‘RandomSeedMaster’: - ‘MeamFit’: - ‘Murnaghan’: - ‘MinimizeMurnaghan’: - ‘ElasticMatrix’: - ‘ConvergenceEncutParallel’: - ‘ConvergenceKpointParallel’: - ’PhonopyMaster’: - ‘DefectFormationEnergy’: - ‘LammpsASE’: - ‘PipelineMaster’: - ’TransformationPath’: - ‘ThermoIntEamQh’: - ‘ThermoIntDftEam’: - ‘ScriptJob’: Python script or jupyter notebook job container - ‘ListMaster': list of jobs
`create_pipeline`(step_lst[, delete_existing_job])	Create a job pipeline
`db_entry`()	Generate the initial database entry
`decompress`()	Decompress the output files of a compressed job object.
`drop_status_to_aborted`()	Change the job status to aborted when the job was intercepted.
`first_child_name`()	Get the name of the first child job
`from_dict`(obj_dict)	Populate the object from the serialized object.
`from_hdf`([hdf, group_name])	Restore the GenericMaster from an HDF5 file
`from_hdf_args`(hdf)	Read arguments for instance creation from HDF5 file
`get`(name[, default])	Internal wrapper function for __getitem__() - self[name]
`get_calculate_function`()	Generate calculate() function
`get_child_cores`()	Calculate the currently active number of cores, by summarizing all childs which are neither finished nor aborted.
`get_final_structure`()	Get the final structure calculated from the job.
`get_from_table`(path, name)	Get a specific value from a pandas.Dataframe
`get_input_parameter_dict`()	Get an hierarchical dictionary of input files.
`get_job_id`([job_specifier])	get the job_id for job named job_name in the local project path from database
`get_neighbors`([start, stop, stride, ...])	Get the neighbors for a given section of the trajectory
`get_neighbors_snapshots`([snapshot_indices, ...])	Get the neighbors only for the required snapshots from the trajectory
`get_output_parameter_dict`()
`get_structure`([frame, wrap_atoms, ...])	Retrieve structure from object.
`get_workdir_file`(filename)	Checks if a given file exists within the job's working directory and returns the absolute path to it.
`gui`()	Returns:
`inspect`(job_specifier)	Inspect an existing pyiron object - most commonly a job - from the database
`instantiate`(obj_dict[, version])	Create a blank instance of this class.
`interactive_close`()	Not implemented for MetaJobs.
`interactive_fetch`()	Not implemented for MetaJobs.
`interactive_flush`([path, include_last_step])	Not implemented for MetaJobs.
`interactive_ref_job_initialize`()	To execute the reference job in interactive mode it is necessary to initialize it.
`is_compressed`()	Check if the job is already compressed or not.
`is_finished`()	Check if the ParallelMaster job is finished - by checking the job status and the submission status.
`is_master_id`(job_id)	Check if the job ID job_id is the master ID for any child job
`is_self_archived`()	Check if the HDF5 file of the Job is compressed as tar-archive
`iter_jobs`([convert_to_object])	Iterate over the jobs within the ListMaster
`iter_structures`([wrap_atoms])	Iterate over all structures in this object.
`job_file_name`(file_name[, cwd])	combine the file name file_name with the path of the current working directory
`kill`()	Kill the job.
`list_all`()	Returns dictionary of :method:`.list_groups()` and :method:`.list_nodes()`.
`list_childs`()	List child jobs as JobPath objects - not loading the full GenericJob objects for each child
`list_files`()	List files inside the working directory
`list_groups`()	Return a list of names of all nested groups.
`list_nodes`()	Return a list of names of all nested nodes.
`load`(job_specifier[, convert_to_object])	Load an existing pyiron object - most commonly a job - from the database
`map`(function, parameter_lst)	Create `MapMaster` with the current job as reference job.
`move_to`(project)	Move the content of the job including the HDF5 file to a new location
`output_to_pandas`([sort_by, h5_path])	Convert output of all child jobs to a pandas Dataframe object.
`pop`([i])	Pop a job from the GenericMaster - just like you would pop an element from a list
`refresh_job_status`()	Refresh job status by updating the job status with the status from the database if a job ID is available.
`refresh_submission_status`()	Refresh the submission status - if a job ID job_id is set then the submission status is loaded from the database.
`relocate_hdf5`([h5_path])	Relocate the hdf file.
`remove`([_protect_childs])	Remove the job - this removes the HDF5 file, all data stored in the HDF5 file an the corresponding database entry.
`remove_and_reset_id`([_protect_childs])	Remove the job and reset its ID.
`remove_child`()	internal function to remove command that removes also child jobs.
`rename`(new_job_name)	Rename the job - by changing the job name
`reset_job_id`([job_id])	Reset the job id sets the job_id to None as well as all connected modules like JobStatus and SubmissionStatus.
`restart`([job_name, job_type])	Restart a new job created from an existing calculation.
`run`([delete_existing_job, repair, debug, ...])	This is the main run function, depending on the job status ['initialized', 'created', 'submitted', 'running', 'collect','finished', 'refresh', 'suspended'] the corresponding run mode is chosen.
`run_if_interactive`()	For jobs which executables are available as Python library, those can also be executed with a library call instead of calling an external executable.
`run_if_interactive_non_modal`()	Not implemented for MetaJobs.
`run_if_modal`()	The run if modal function is called by run to execute the simulation, while waiting for the output.
`run_if_refresh`()	Internal helper function the run if refresh function is called when the job status is 'refresh'.
`run_if_scheduler`()	The run if queue function is called by run if the user decides to submit the job to and queing system.
`run_static`()	The run_static function is executed within the GenericJob class and depending on the run_mode of the Parallelmaster and its child jobs a more specific run function is selected.
`run_time_to_db`()	Internal helper function to store the run_time in the database
`save`()	Save the object, by writing the content to the HDF5 file and storing an entry in the database.
`save_output`([output_dict, shell_output])	Store output of the calculate function in the HDF5 file.
`self_archive`()	Compress HDF5 file of the job object to tar-archive
`self_unarchive`()	Decompress HDF5 file of the job object from tar-archive
`set_child_id_func`(child_id_func)	Add an external function to derive a list of child IDs - experimental feature
`set_input_to_read_only`()	This function enforces read-only mode for the input classes, but it has to be implemented in the individual classes.
`show_hdf`()	Display the output of the child jobs in a human readable print out
`signal_intercept`(sig)	Abort the job and log signal that caused it.
`store_structure`()	Create `StructureContainer` job with the initial structure of the job and sets that jobs `parent_id` from this job.
`suspend`()	Suspend the job by storing the object and its state persistently in HDF5 file and exit it.
`to_dict`()	Reduce the object to a dictionary.
`to_hdf`([hdf, group_name])	Store the GenericMaster in an HDF5 file
`to_object`([object_type])	Load the full pyiron object from an HDF5 file
`trajectory`([stride, center_of_mass, ...])	Returns a Trajectory instance containing the necessary information to describe the evolution of the atomic structure during the atomistic simulation.
`transfer_from_remote`()	Transfer the job from a remote location to the local machine.
`transform_structures`(modify)	Return a modified object by applying a function to each object lazily.
`update_master`([force_update])	After a job is finished it checks whether it is linked to any metajob - meaning the master ID is pointing to this jobs job ID.
`validate_ready_to_run`()	Returns:
`view_structure`([snapshot, spacefill, show_cell])	param snapshot: Snapshot of the trajectory one wants
`write_input`()	Call routines that generate the code specific input files Returns:
`write_traj`(filename[, file_format, ...])	Writes the trajectory in a given file file_format based on the ase.io.write function.

Attributes

`calculate_kwargs`	Generate keyword arguments for the calculate() function.
`child_ids`	list of child job ids - only meta jobs have child jobs - jobs which list the meta job as their master
`child_names`	Dictionary matching the child ID to the child job name
`child_project`	project which holds the created child jobs
`content`
`database_entry`
`exclude_groups_hdf`	Get the list of groups which are excluded from storing in the hdf5 file
`exclude_nodes_hdf`	Get the list of nodes which are excluded from storing in the hdf5 file
`executable`	Get the executable used to run the job - usually the path to an external executable.
`executor_type`
`files`	Allows to browse the files in a job directory.
`files_to_compress`
`files_to_remove`
`id`	Unique id to identify the job in the pyiron database - use self.job_id instead
`input`
`job_id`	Unique id to identify the job in the pyiron database
`job_info_str`	Short string to describe the job by it is job_name and job ID - mainly used for logging
`job_name`	Get name of the job, which has to be unique within the project
`job_object_dict`	internal cache of currently loaded jobs
`job_type`	['ExampleJob', 'ParallelMaster', 'ScriptJob',
`logger`	Get the logger object to monitor the external execution and internal pyiron warnings.
`master_id`	Get job id of the master job - a meta job which groups a series of jobs, which are executed either in parallel or in serial.
`name`	Get name of the job, which has to be unique within the project
`number_jobs_total`	Get number of total jobs
`number_of_structures`	maximum iteration_step + 1 that can be passed to `get_structure()`.
`parent_id`	Get job id of the predecessor job - the job which was executed before the current one in the current job series
`path`	Absolute path of the HDF5 group starting from the system root - combination of the absolute system path plus the absolute path inside the HDF5 file starting from the root group.
`project`	Project instance the jobs is located in
`project_hdf5`	Get the ProjectHDFio instance which points to the HDF5 file the job is stored in
`queue_id`	Get the queue ID, the ID returned from the queuing system - it is most likely not the same as the job ID.
`ref_job`	Get the reference job template from which all jobs within the ParallelMaster are generated.
`restart_file_dict`	A dictionary of the new name of the copied restart files
`restart_file_list`	Get the list of files which are used to restart the calculation from these files.
`server`	Get the server object to handle the execution environment for the job.
`status`	Execution status of the job, can be one of the following [initialized, appended, created, submitted, running,
`structure`	Returns:
`version`	Get the version of the hamiltonian, which is also the version of the executable unless a custom executable is used.
`working_directory`	Get the working directory of the job is executed in - outside the HDF5 file. The working directory equals the path but it is represented by the filesystem: /absolute/path/to/the/file.h5/path/inside/the/hdf5/file becomes: /absolute/path/to/the/file_hdf5/path/inside/the/hdf5/file.

animate_structure(spacefill: bool = True, show_cell: bool = True, stride: int = 1, center_of_mass: bool = False, particle_size: float = 0.5, camera: str = 'orthographic', atom_indices: list | ndarray = None, snapshot_indices: list | ndarray = None, repeat: int | Tuple[int, int, int] = None)#

Animates the job if a trajectory is present

Parameters:

spacefill (bool) – If True, then atoms are visualized in spacefill stype
show_cell (bool) – True if the cell boundaries of the structure is to be shown
stride (int) –
show animation every stride [::stride] use value >1 to make animation faster

default=1
particle_size (float) – Scaling factor for the spheres representing the atoms. (The radius is determined by the atomic number)
center_of_mass (bool) – False (default) if the specified positions are w.r.t. the origin
camera (str) – camera perspective, choose from “orthographic” or “perspective”
atom_indices (list/numpy.ndarray) – The atom indices for which the trajectory should be generated
snapshot_indices (list/numpy.ndarray) – The snapshots for which the trajectory should be generated
repeat (int/3-tuple of int) – Repeat the structures by this before animating

Returns:

nglview IPython widget

Return type:

animation

animate_structures(spacefill: bool = True, show_cell: bool = True, center_of_mass: bool = False, particle_size: float = 0.5, camera: str = 'orthographic')#

Animate a series of atomic structures.

Parameters:

spacefill (bool) – If True, then atoms are visualized in spacefill stype
show_cell (bool) – True if the cell boundaries of the structure is to be shown
particle_size (float) – Scaling factor for the spheres representing the atoms. (The radius is determined by the atomic number)
center_of_mass (bool) – False (default) if the specified positions are w.r.t. the origin
camera (str) – camera perspective, choose from “orthographic” or “perspective”

Returns:

nglview IPython widget

Return type:

animation

append(job)#

Append a job to the GenericMaster - just like you would append an element to a list.

Parameters:: job (GenericJob) – job to append

calc_minimize(ionic_energy_tolerance=0, ionic_force_tolerance=0.0001, e_tol=None, f_tol=None, max_iter=1000, pressure=None, n_print=1)#

Parameters:

ionic_energy_tolerance (float) – Maximum energy difference between 2 steps
ionic_force_tolerance (float) – Maximum force magnitude that each of atoms is allowed to have
e_tol (float) – same as ionic_energy_tolerance (deprecated)
f_tol (float) – same as ionic_force_tolerance (deprecated)
max_iter (int) – Maximum number of force evluations
pressure (float/list) – Targetpressure values
n_print (int) – Print period

Returns:

calc_static()#: Returns:

property calculate_kwargs: dict#

Generate keyword arguments for the calculate() function. A new simulation code only has to extend the get_input_parameter_dict() function which by default specifies an hierarchical dictionary with files_to_write and files_to_copy.

Example:

>>> calculate_function = job.get_calculate_function()
>>> shell_output, parsed_output, job_crashed = calculate_function(**job.calculate_kwargs)
>>> job.save_output(output_dict=parsed_output, shell_output=shell_output)

Returns:: keyword arguments for the calculate() function
Return type:: dict

check_if_job_exists(job_name: str | None = None, project: ProjectHDFio | pyiron_base.project.generic.Project | None = None)#

Check if a job already exists in an specific project.

Parameters:

job_name (str) – Job name (optional)
project (ProjectHDFio, Project) – Project path (optional)

Returns:

True / False

Return type:

(bool)

check_setup() → None#: Checks whether certain parameters (such as plane wave cutoff radius in DFT) are changed from the pyiron standard values to allow for a physically meaningful results. This function is called manually or only when the job is submitted to the queueing system.

child_hdf(job_name)#

Find correct HDF for new children. Depending on self.server.new_hdf this creates a new hdf file or creates the group in the file of this job.

Parameters:: job_name (str) – name of the new job
Returns:: HDF file for new child job, can be assigned to its project_hdf5
Return type:: ProjectHDFio

property child_ids#

list of child job ids - only meta jobs have child jobs - jobs which list the meta job as their master

Returns:: list of child job ids
Return type:: list

property child_names#

Dictionary matching the child ID to the child job name

Returns:: {child_id: child job name }
Return type:: dict

property child_project#

project which holds the created child jobs

Type:: Project

clear_job() → None#: Convenience function to clear job info after suspend. Mimics deletion of all the job info after suspend in a local test environment.

collect_logfiles()#: Collect the log files of the external executable and store the information in the HDF5 file. This method is currently not implemented for the ParallelMaster.

collect_output()#: Collect the output files of the external executable and store the information in the HDF5 file. This method has to be implemented in the individual meta jobs derived from the ParallelMaster.

collect_structures(filter_function=None) → StructureStorage#

Collects a copy of all structures in a compact StructureStorage.

This can be used to force lazily applied modifications with transform_structures() or simply to obtain a known object type from a generic HasStructure object.

Parameters:: filter_function (function) – include structure only if this function returns True for it
Returns:: a copy of all (filtered) structures
Return type:: StructureStorage

compress(files_to_compress: List[str] | None = None, files_to_remove: List[str] | None = None) → None#

Compress the output files of a job object.

Parameters:: files_to_compress (list)

continue_with_final_structure(job_type=None, job_name=None)#

Parameters:

job_type
job_name

Returns:

continue_with_restart_files(job_type=None, job_name=None)#

Parameters:

job_type
job_name

Returns:

convergence_check() → bool#

Check if and all child jobs of the calculation are converged. May need be extended in the base classes depending on the specific application

Returns:: If the calculation is converged
Return type:: (bool)

copy()#

Copy the GenericJob object which links to the job and its HDF5 file

Returns:: New object pointing to the same job
Return type:: GenericJob

copy_file_to_working_directory(file: str) → None#

Copy a specific file to the working directory before the job is executed.

Parameters:: file (str) – path of the file to be copied.

copy_template(project: ProjectHDFio | JobCore | None = None, new_job_name: None = None) → GenericJob#

Copy the content of the job including the HDF5 file but without the output data to a new location

Parameters:

project (JobCore/ProjectHDFio/Project/None) – The project to copy the job to. (Default is None, use the same project.)
new_job_name (str) – The new name to assign the duplicate job. Required if the project is None or the same project as the copied job. (Default is None, try to keep the same name.)

Returns:

GenericJob object pointing to the new location.

Return type:

GenericJob

copy_to(project: ProjectHDFio | JobCore | None = None, new_job_name: str | None = None, input_only: bool = False, new_database_entry: bool = True, delete_existing_job: bool = False, copy_files: bool = True)#

Copy the content of the job including the HDF5 file to a new location.

Parameters:

project (JobCore/ProjectHDFio/Project/None) – The project to copy the job to. (Default is None, use the same project.)
new_job_name (str) – The new name to assign the duplicate job. Required if the project is None or the same project as the copied job. (Default is None, try to keep the same name.)
input_only (bool) – [True/False] Whether to copy only the input. (Default is False.)
new_database_entry (bool) – [True/False] Whether to create a new database entry. If input_only is True then new_database_entry is False. (Default is True.)
delete_existing_job (bool) – [True/False] Delete existing job in case it exists already (Default is False.)
copy_files (bool) – If True copy all files the working directory of the job, too

Returns:

GenericJob object pointing to the new location.

Return type:

GenericJob

create_child_job(job_name)#

Internal helper function to create the next child job from the reference job template - usually this is called as part of the create_jobs() function.

Parameters:: job_name (str) – name of the next job
Returns:: next job
Return type:: GenericJob

create_job(job_type: str, job_name: str, delete_existing_job: bool = False) → GenericJob#

Create one of the following jobs: - ‘StructureContainer’: - ‘StructurePipeline’: - ‘AtomisticExampleJob’: example job just generating random number - ‘ExampleJob’: example job just generating random number - ‘Lammps’: - ‘KMC’: - ‘Sphinx’: - ‘Vasp’: - ‘GenericMaster’: - ‘ParallelMaster’: series of jobs run in parallel - ‘KmcMaster’: - ‘ThermoLambdaMaster’: - ‘RandomSeedMaster’: - ‘MeamFit’: - ‘Murnaghan’: - ‘MinimizeMurnaghan’: - ‘ElasticMatrix’: - ‘ConvergenceEncutParallel’: - ‘ConvergenceKpointParallel’: - ’PhonopyMaster’: - ‘DefectFormationEnergy’: - ‘LammpsASE’: - ‘PipelineMaster’: - ’TransformationPath’: - ‘ThermoIntEamQh’: - ‘ThermoIntDftEam’: - ‘ScriptJob’: Python script or jupyter notebook job container - ‘ListMaster’: list of jobs

Parameters:

job_type (str) – job type can be [‘StructureContainer’, ‘StructurePipeline’, ‘AtomisticExampleJob’, ‘ExampleJob’, ‘Lammps’, ‘KMC’, ‘Sphinx’, ‘Vasp’, ‘GenericMaster’, ‘ParallelMaster’, ‘KmcMaster’, ‘ThermoLambdaMaster’, ‘RandomSeedMaster’, ‘MeamFit’, ‘Murnaghan’, ‘MinimizeMurnaghan’, ‘ElasticMatrix’, ‘ConvergenceEncutParallel’, ‘ConvergenceKpointParallel’, ’PhonopyMaster’, ‘DefectFormationEnergy’, ‘LammpsASE’, ‘PipelineMaster’, ’TransformationPath’, ‘ThermoIntEamQh’, ‘ThermoIntDftEam’, ‘ScriptJob’, ‘ListMaster’]
job_name (str) – name of the job
delete_existing_job (bool) – delete an existing job - default false

Returns:

job object depending on the job_type selected

Return type:

GenericJob

create_pipeline(step_lst, delete_existing_job=False)#

Create a job pipeline

Parameters:: step_lst (list) – List of functions which create calculations
Return type:: FlexibleMaster

db_entry()#

Generate the initial database entry

Returns:: db_dict
Return type:: (dict)

decompress() → None#: Decompress the output files of a compressed job object.

drop_status_to_aborted() → None#: Change the job status to aborted when the job was intercepted.

property exclude_groups_hdf: list#

Get the list of groups which are excluded from storing in the hdf5 file

Returns:: groups(list)

property exclude_nodes_hdf: list#

Get the list of nodes which are excluded from storing in the hdf5 file

Returns:: nodes(list)

property executable: Executable#

Get the executable used to run the job - usually the path to an external executable.

Returns:: exectuable path
Return type:: (str/pyiron_base.job.executable.Executable)

property files: FileBrowser#

Allows to browse the files in a job directory.

By default this object prints itself as a listing of the job directory and the files inside.
>>> job.files
/path/to/my/job:
    pyiron.log
    error.out
Access to the names of files is provided with list()
>>> job.files.list()
['pyiron.log', 'error.out', 'INCAR']
Access to the contents of files is provided by indexing into this object, which returns a list of lines in the file
>>> job.files['error.out']
["Oh no

“, “Something went wrong! “]

The tail() method prints the last lines of a file to stdout
>>> job.files.tail('error.out', lines=1)
Something went wrong!
For files that have valid python variable names can also be accessed by attribute notation
>>> job.files.INCAR 
File('INCAR')

first_child_name()#

Get the name of the first child job

Returns:: name of the first child job
Return type:: str

from_dict(obj_dict)#

Populate the object from the serialized object.

Parameters:

obj_dict (dict) – data previously returned from to_dict()
version (str) – version tag written together with the data

from_hdf(hdf=None, group_name=None)#

Restore the GenericMaster from an HDF5 file

Parameters:

hdf (ProjectHDFio) – HDF5 group object - optional
group_name (str) – HDF5 subgroup name - optional

classmethod from_hdf_args(hdf: ProjectHDFio) → dict#

Read arguments for instance creation from HDF5 file

Parameters:: hdf (ProjectHDFio) – HDF5 group object

get(name: str, default: Any | None = None) → Any#

Internal wrapper function for __getitem__() - self[name]

Parameters:

key (str, slice) – path to the data or key of the data object
default (any, optional) – return this if key cannot be found

Returns:

data or data object

Return type:

dict, list, float, int

Raises:

ValueError – key cannot be found and default is not given

get_calculate_function() → callable#

Generate calculate() function

Example:

>>> calculate_function = job.get_calculate_function()
>>> shell_output, parsed_output, job_crashed = calculate_function(**job.calculate_kwargs)
>>> job.save_output(output_dict=parsed_output, shell_output=shell_output)

Returns:: calculate() functione
Return type:: callable

get_child_cores()#

Calculate the currently active number of cores, by summarizing all childs which are neither finished nor aborted.

Returns:: number of cores used
Return type:: (int)

get_final_structure()#

Get the final structure calculated from the job.

Returns:: Atoms

get_from_table(path: str, name: str) → dict | list | float | int#

Get a specific value from a pandas.Dataframe

Parameters:

path (str) – relative path to the data object
name (str) – parameter key

Returns:

the value associated to the specific parameter key

Return type:

dict, list, float, int

get_input_parameter_dict() → dict#

Get an hierarchical dictionary of input files. On the first level the dictionary is divided in file_to_create and files_to_copy. Both are dictionaries use the file names as keys. In file_to_create the values are strings which represent the content which is going to be written to the corresponding file. In files_to_copy the values are the paths to the source files to be copied.

Returns:: hierarchical dictionary of input files
Return type:: dict

get_job_id(job_specifier: int | str | None = None) → int | None#

get the job_id for job named job_name in the local project path from database

Parameters:: job_specifier (str, int) – name of the job or job ID
Returns:: job ID of the job
Return type:: int

get_neighbors(start=0, stop=-1, stride=1, num_neighbors=12, **kwargs)#

Get the neighbors for a given section of the trajectory

Parameters:

start (int) – Start point of the slice of the trajectory to be sampled
stop (int) – End point of of the slice of the trajectory to be sampled
stride (int) – Samples the snapshots evert stride steps
num_neighbors (int) – The cutoff for the number of neighbors
**kwargs (dict) – Additional arguments to be passed to the get_neighbors() routine (eg. cutoff_radius, norm_order , etc.)

Returns:

NeighborsTraj instances: containing the neighbor information.

Return type:

pyiron_atomistics.atomistics.structure.neighbors.NeighborsTrajectory

get_neighbors_snapshots(snapshot_indices=None, num_neighbors=12, **kwargs)#

Get the neighbors only for the required snapshots from the trajectory

Parameters:

snapshot_indices (list/numpy.ndarray) – Snapshots for which the the neighbors need to be computed (eg. [1, 5, 10,…, 100]
num_neighbors (int) – The cutoff for the number of neighbors
**kwargs (dict) – Additional arguments to be passed to the get_neighbors() routine (eg. cutoff_radius, norm_order , etc.)

Returns:

NeighborsTraj instances: containing the neighbor information.

Return type:

pyiron_atomistics.atomistics.structure.neighbors.NeighborsTrajectory

get_structure(frame=-1, wrap_atoms=True, iteration_step=None)#

Retrieve structure from object. The number of available structures depends on the job and what kind of calculation has been run on it, see number_of_structures.

Parameters:: frame (int, object) – index of the structure requested, if negative count from the back; if

:param _translate_frame() is overridden: :param frame will pass through it: :param iteration_step: deprecated alias for frame :type iteration_step: int :param wrap_atoms: True if the atoms are to be wrapped back into the unit cell :type wrap_atoms: bool

Returns:: the requested structure
Return type:: pyiron_atomistics.atomistics.structure.atoms.Atoms
Raises:: IndexError – if not -number_of_structures <= iteration_step < number_of_structures

get_workdir_file(filename: str) → None#

Checks if a given file exists within the job’s working directory and returns the absolute path to it.

ToDo: Move this to pyiron_base since this is more generic.

Parameters:: filename (str) – The name of the file
Returns:: The name absolute path of the file in the working directory
Return type:: str
Raises:: FileNotFoundError – Raised if the given file does not exist.

gui()#: Returns:

property id: int#

Unique id to identify the job in the pyiron database - use self.job_id instead

Returns:: job id
Return type:: int

inspect(job_specifier: str | int) → JobCore#

Inspect an existing pyiron object - most commonly a job - from the database

Parameters:: job_specifier (str, int) – name of the job or job ID
Returns:: Access to the HDF5 object - not a GenericJob object - use load() instead.
Return type:: JobCore

classmethod instantiate(obj_dict: dict, version: str = None) → Self#

Create a blank instance of this class.

This can be used when some values are already necessary for the objects __init__.

Parameters:

obj_dict (dict) – data previously returned from to_dict()
version (str) – version tag written together with the data

Returns:

a blank instance of the object that is sufficiently initialized to call _from_dict() on it

Return type:

object

interactive_close()#: Not implemented for MetaJobs.

interactive_fetch()#: Not implemented for MetaJobs.

interactive_flush(path='generic', include_last_step=True)#: Not implemented for MetaJobs.

interactive_ref_job_initialize()#: To execute the reference job in interactive mode it is necessary to initialize it.

is_compressed() → bool#

Check if the job is already compressed or not.

Returns:: [True/False]
Return type:: bool

is_finished()#

Check if the ParallelMaster job is finished - by checking the job status and the submission status.

Returns:: [True/False]
Return type:: bool

is_master_id(job_id: int) → bool#

Check if the job ID job_id is the master ID for any child job

Parameters:: job_id (int) – job ID of the master job
Returns:: [True/False]
Return type:: bool

is_self_archived() → bool#

Check if the HDF5 file of the Job is compressed as tar-archive

Returns:: [True/False]
Return type:: bool

iter_jobs(convert_to_object=True)#

Iterate over the jobs within the ListMaster

Parameters:: convert_to_object (bool) – load the full GenericJob object (default) or just the HDF5 / JobCore object
Returns:: Yield of GenericJob or JobCore
Return type:: yield

iter_structures(wrap_atoms=True)#

Iterate over all structures in this object.

Parameters:: wrap_atoms (bool) – True if the atoms are to be wrapped back into the unit cell; passed to get_structure()
Yields:: pyiron_atomistics.atomistitcs.structure.atoms.Atoms – every structure attached to the object

job_file_name(file_name: str, cwd: str | None = None) → str#

combine the file name file_name with the path of the current working directory

Parameters:

file_name (str) – name of the file
cwd (str) – current working directory - this overwrites self.project_hdf5.working_directory - optional

Returns:

absolute path to the file in the current working directory

Return type:

str

property job_id: int#

Unique id to identify the job in the pyiron database

Returns:: job id
Return type:: int

property job_info_str: str#

Short string to describe the job by it is job_name and job ID - mainly used for logging

Returns:: job info string
Return type:: str

property job_name: str#

Get name of the job, which has to be unique within the project

Returns:: job name
Return type:: str

property job_object_dict#

internal cache of currently loaded jobs

Returns:: Dictionary of currently loaded jobs
Return type:: dict

property job_type: str#

[‘ExampleJob’, ‘ParallelMaster’, ‘ScriptJob’,: ‘ListMaster’]

Returns:: Job type object
Return type:: JobTypeChoice
Type:: Job type object with all the available job types

kill() → None#

Kill the job.

This function is used to terminate the execution of the job. It checks if the job is currently running or submitted, and if so, it removes and resets the job ID. If the job is not running or submitted, a ValueError is raised.

Returns:: None

list_all()#

Returns dictionary of :method:`.list_groups()` and :method:`.list_nodes()`.

Returns:

results of :method:`.list_groups() under the key "groups"; results of :method:`.list_nodes()` und the: key “nodes”

Return type:

dict

list_childs() → list#

List child jobs as JobPath objects - not loading the full GenericJob objects for each child

Returns:: list of child jobs
Return type:: list

list_files() → list#

List files inside the working directory

Parameters:: extension (str) – filter by a specific extension
Returns:: list of file names
Return type:: list

list_groups()#

Return a list of names of all nested groups.

Returns:: group names
Return type:: list of str

list_nodes()#

Return a list of names of all nested nodes.

Returns:: node names
Return type:: list of str

load(job_specifier: str | int, convert_to_object: bool = True) → pyiron_base.job.generic.GenericJob | JobCore#

Load an existing pyiron object - most commonly a job - from the database

Parameters:

job_specifier (str, int) – name of the job or job ID
convert_to_object (bool) – convert the object to an pyiron object or only access the HDF5 file - default=True accessing only the HDF5 file is about an order of magnitude faster, but only provides limited functionality. Compare the GenericJob object to JobCore object.

Returns:

Either the full GenericJob object or just a reduced JobCore object

Return type:

GenericJob, JobCore

property logger#

Get the logger object to monitor the external execution and internal pyiron warnings.

Returns:: logger object
Return type:: logging.getLogger()

map(function, parameter_lst)#

Create MapMaster with the current job as reference job.

The job name is created as ‘map_{self.name}’

Parameters:

function (callable) – passed as modify_function to the map master
parameter_list (list) – passed as parameter_list to the map master

Returns:

newly created master job

Return type:

MapMaster

property master_id: int#

Get job id of the master job - a meta job which groups a series of jobs, which are executed either in parallel or in serial.

Returns:: master id
Return type:: int

move_to(project)#

Move the content of the job including the HDF5 file to a new location

Parameters:: project (ProjectHDFio) – project to move the job to
Returns:: JobCore object pointing to the new location.
Return type:: JobCore

property name: str#

Get name of the job, which has to be unique within the project

Returns:: job name
Return type:: str

property number_jobs_total#

Get number of total jobs

Returns:: number of total jobs
Return type:: int

property number_of_structures#

maximum iteration_step + 1 that can be passed to get_structure().

Type:: int

output_to_pandas(sort_by=None, h5_path='output')#

Convert output of all child jobs to a pandas Dataframe object.

Parameters:

sort_by (str) – sort the output using pandas.DataFrame.sort_values(by=sort_by)
h5_path (str) – select child output to include - default=’output’

Returns:

output as dataframe

Return type:

pandas.Dataframe

property parent_id: int#

Get job id of the predecessor job - the job which was executed before the current one in the current job series

Returns:: parent id
Return type:: int

property path: str#

Absolute path of the HDF5 group starting from the system root - combination of the absolute system path plus the absolute path inside the HDF5 file starting from the root group.

Returns:: absolute path
Return type:: str

pop(i=-1)#

Pop a job from the GenericMaster - just like you would pop an element from a list

Parameters:: i (int) – position of the job. (Default is last element, -1.)
Returns:: job
Return type:: GenericJob

property project: pyiron_base.project.generic.Project#

Project instance the jobs is located in

Returns:: project the job is located in
Return type:: Project

property project_hdf5: ProjectHDFio#

Get the ProjectHDFio instance which points to the HDF5 file the job is stored in

Returns:: HDF5 project
Return type:: ProjectHDFio

property queue_id: int#

Get the queue ID, the ID returned from the queuing system - it is most likely not the same as the job ID.

Returns:: queue ID
Return type:: int

property ref_job#

Get the reference job template from which all jobs within the ParallelMaster are generated.

Returns:: reference job
Return type:: GenericJob

refresh_job_status() → None#: Refresh job status by updating the job status with the status from the database if a job ID is available.

refresh_submission_status()#: Refresh the submission status - if a job ID job_id is set then the submission status is loaded from the database.

relocate_hdf5(h5_path: str | None = None)#: Relocate the hdf file. This function is needed when the child job is spawned by a parent job (cf. pyiron_base.jobs.master.generic)

remove(_protect_childs: bool = True) → None#

Remove the job - this removes the HDF5 file, all data stored in the HDF5 file an the corresponding database entry.

Parameters:: _protect_childs (bool) – [True/False] by default child jobs can not be deleted, to maintain the consistency - default=True

remove_and_reset_id(_protect_childs: bool = True) → None#

Remove the job and reset its ID.

Parameters:: _protect_childs (bool) – Flag indicating whether to protect child jobs (default is True).
Returns:: None

remove_child() → None#: internal function to remove command that removes also child jobs. Do never use this command, since it will destroy the integrity of your project.

rename(new_job_name: str) → None#

Rename the job - by changing the job name

Parameters:: new_job_name (str) – new job name

reset_job_id(job_id=None)#: Reset the job id sets the job_id to None as well as all connected modules like JobStatus and SubmissionStatus.

restart(job_name=None, job_type=None)#

Restart a new job created from an existing calculation. :param project: Project instance at which the new job should be created :type project: pyiron_atomistics.project.Project instance :param job_name: Job name :type job_name: str :param job_type: Job type :type job_type: str

Returns:: New job
Return type:: new_ham

property restart_file_dict: dict#: A dictionary of the new name of the copied restart files

property restart_file_list: list#

Get the list of files which are used to restart the calculation from these files.

Returns:: list of files
Return type:: list

run(delete_existing_job: bool = False, repair: bool = False, debug: bool = False, run_mode: str | None = None, run_again: bool = False) → None#

This is the main run function, depending on the job status [‘initialized’, ‘created’, ‘submitted’, ‘running’, ‘collect’,’finished’, ‘refresh’, ‘suspended’] the corresponding run mode is chosen.

Parameters:

delete_existing_job (bool) – Delete the existing job and run the simulation again.
repair (bool) – Set the job status to created and run the simulation again.
debug (bool) – Debug Mode - defines the log level of the subprocess the job is executed in.
run_mode (str) – [‘modal’, ‘non_modal’, ‘queue’, ‘manual’] overwrites self.server.run_mode
run_again (bool) – Same as delete_existing_job (deprecated)

run_if_interactive()#: For jobs which executables are available as Python library, those can also be executed with a library call instead of calling an external executable. This is usually faster than a single core python job.

run_if_interactive_non_modal()#: Not implemented for MetaJobs.

run_if_modal() → None#: The run if modal function is called by run to execute the simulation, while waiting for the output. For this we use subprocess.check_output()

run_if_refresh()#: Internal helper function the run if refresh function is called when the job status is ‘refresh’. If the job was suspended previously, the job is going to be started again, to be continued.

run_if_scheduler() → None | int#: The run if queue function is called by run if the user decides to submit the job to and queing system. The job is submitted to the queuing system using subprocess.Popen() :returns: Returns the queue ID for the job. :rtype: int

run_static()#: The run_static function is executed within the GenericJob class and depending on the run_mode of the Parallelmaster and its child jobs a more specific run function is selected.

run_time_to_db() → None#: Internal helper function to store the run_time in the database

save()#

Save the object, by writing the content to the HDF5 file and storing an entry in the database.

Returns:: Job ID stored in the database
Return type:: (int)

save_output(output_dict: dict | None = None, shell_output: str | None = None) → None#

Store output of the calculate function in the HDF5 file.

Parameters:

output_dict (dict) – hierarchical output dictionary to be stored in the HDF5 file.
shell_output (str) – shell output from calling the external executable to be stored in the HDF5 file.

self_archive() → None#: Compress HDF5 file of the job object to tar-archive

self_unarchive() → None#: Decompress HDF5 file of the job object from tar-archive

property server: Server#

Get the server object to handle the execution environment for the job.

Returns:: server object
Return type:: Server

set_child_id_func(child_id_func)#

Add an external function to derive a list of child IDs - experimental feature

Parameters:: child_id_func (Function) – Python function which returns the list of child IDs

set_input_to_read_only() → None#: This function enforces read-only mode for the input classes, but it has to be implemented in the individual classes.

show_hdf()#: Display the output of the child jobs in a human readable print out

signal_intercept(sig) → None#

Abort the job and log signal that caused it.

Expected to be called from pyiron_base.state.signal.catch_signals().

Parameters:: sig (int) – the signal that triggered the abort

property status: str#

Execution status of the job, can be one of the following [initialized, appended, created, submitted, running,: aborted, collect, suspended, refresh, busy, finished]

Returns:: status
Return type:: (str/pyiron_base.job.jobstatus.JobStatus)

store_structure()#

Create StructureContainer job with the initial structure of the job and sets that jobs parent_id from this job.

Returns:: job containing initial structure of this job
Return type:: StructureContainer

property structure#: Returns:

suspend() → None#: Suspend the job by storing the object and its state persistently in HDF5 file and exit it.

to_dict()#

Reduce the object to a dictionary.

Returns:: serialized state of this object
Return type:: dict

to_hdf(hdf=None, group_name=None)#

Store the GenericMaster in an HDF5 file

Parameters:

hdf (ProjectHDFio) – HDF5 group object - optional
group_name (str) – HDF5 subgroup name - optional

to_object(object_type: str | None = None, **qwargs) → pyiron_base.job.generic.GenericJob#

Load the full pyiron object from an HDF5 file

Parameters:

object_type – if the ‘TYPE’ node is not available in the HDF5 file a manual object type can be set - optional
**qwargs – optional parameters [‘job_name’, ‘project’] - to specify the location of the HDF5 path

Returns:

pyiron object

Return type:

GenericJob

trajectory(stride=1, center_of_mass=False, atom_indices=None, snapshot_indices=None, overwrite_positions=None, overwrite_cells=None)#

Returns a Trajectory instance containing the necessary information to describe the evolution of the atomic structure during the atomistic simulation.

Parameters:

stride (int) – The trajectories are generated with every ‘stride’ steps
center_of_mass (bool) – False (default) if the specified positions are w.r.t. the origin
atom_indices (list/ndarray) – The atom indices for which the trajectory should be generated
snapshot_indices (list/ndarray) – The snapshots for which the trajectory should be generated
overwrite_positions (list/ndarray) – List of positions that are meant to overwrite the existing trajectory. Useful to wrap coordinates for example
overwrite_cells (list/ndarray) – List of cells that are meant to overwrite the existing trajectory. Only used when overwrite_positions is defined. This must have the same length of overwrite_positions

Returns:

Trajectory instance

Return type:

pyiron_atomistics.atomistics.job.atomistic.Trajectory

transfer_from_remote() → None#

Transfer the job from a remote location to the local machine.

This method transfers the job from a remote location to the local machine. It performs the following steps: 1. Retrieves the job from the remote location using the queue adapter. 2. Transfers the job file to the remote location, with the option to delete the file on the remote location after transfer. 3. Updates the project database if it is disabled, otherwise updates the file table in the database with the job information.

Parameters:: None
Returns:: None

transform_structures(modify) → TransformStructure#

Return a modified object by applying a function to each object lazily.

Parameters:: modify (function) – applied to each structure, has to return the modified structure
Returns:: a container with the modified structures
Return type:: TransformStructure

update_master(force_update=True)#

After a job is finished it checks whether it is linked to any metajob - meaning the master ID is pointing to this jobs job ID. If this is the case and the master job is in status suspended - the child wakes up the master job, sets the status to refresh and execute run on the master job. During the execution the master job is set to status refresh. If another child calls update_master, while the master is in refresh the status of the master is set to busy and if the master is in status busy at the end of the update_master process another update is triggered.

Parameters:: force_update (bool) – Whether to check run mode for updating master

validate_ready_to_run()#: Returns:

property version: str#

Get the version of the hamiltonian, which is also the version of the executable unless a custom executable is used.

Returns:: version number
Return type:: str

view_structure(snapshot=-1, spacefill=True, show_cell=True)#

Parameters:

snapshot (int) – Snapshot of the trajectory one wants
spacefill (bool)
show_cell (bool)

Returns:

nglview IPython widget

Return type:

view

property working_directory: str#

Get the working directory of the job is executed in - outside the HDF5 file. The working directory equals the path but it is represented by the filesystem:

/absolute/path/to/the/file.h5/path/inside/the/hdf5/file

becomes:: /absolute/path/to/the/file_hdf5/path/inside/the/hdf5/file

Returns:: absolute path to the working directory
Return type:: str

write_input() → None#: Call routines that generate the code specific input files Returns:

write_traj(filename, file_format=None, parallel=True, append=False, stride=1, center_of_mass=False, atom_indices=None, snapshot_indices=None, overwrite_positions=None, overwrite_cells=None, **kwargs)#

Writes the trajectory in a given file file_format based on the ase.io.write function.

Parameters:

filename (str) – Filename of the output
file_format (str) – The specific file_format of the output
parallel (bool) – ase parameter
append (bool) – ase parameter
stride (int) – Writes trajectory every stride steps
center_of_mass (bool) – True if the positions are centered on the COM
atom_indices (list/numpy.ndarray) – The atom indices for which the trajectory should be generated
snapshot_indices (list/numpy.ndarray) – The snapshots for which the trajectory should be generated
overwrite_positions (list/numpy.ndarray) – List of positions that are meant to overwrite the existing trajectory. Useful to wrap coordinates for example
overwrite_cells (list/numpy.ndarray) – List of cells that are meant to overwrite the existing trajectory. Only used when overwrite_positions is defined. This must have the same length of overwrite_positions
**kwargs – Additional ase arguments

pyiron_atomistics.atomistics.master.parallel.AtomisticParallelMaster

Contents

pyiron_atomistics.atomistics.master.parallel.AtomisticParallelMaster#