mdsuite.experiment package

Submodules

Module contents

MDSuite: A Zincwarecode package.

License

This program and the accompanying materials are made available under the terms of the Eclipse Public License v2.0 which accompanies this distribution, and is available at https://www.eclipse.org/legal/epl-v20.html

SPDX-License-Identifier: EPL-2.0

Copyright Contributors to the Zincwarecode Project.

Contact Information

email: zincwarecode@gmail.com github: https://github.com/zincware web: https://zincwarecode.com/

Citation

If you use this module please cite us with:

Summary

class mdsuite.experiment.Experiment(project, name, time_step=None, temperature=None, units: Optional[Union[str, Units]] = None, cluster_mode=False)[source]

Bases: ExperimentDatabase

The central experiment class fundamental to all analysis.

project = mdsuite.Project()
project.add_experiment(
    name="NaCl",
    timestep=0.002,
    temperature=1400.0,
    units="metal",
    simulation_data="NaCl_gk_i_q.lammpstraj"
    )
project.experiments.NaCl.run.RadialDistributionFunction(
    number_of_configurations=500
)
name

The name of the analysis being performed e.g. NaCl_1400K

Type:

str

storage_path

Path to where the tensor_values should be stored (best to have drive capable of storing large files)

Type:

str

temperature

The temperature of the simulation that should be used in some analysis. Necessary as it cannot be easily read in from the simulation tensor_values.

Type:

float

time_step

Time step of the simulation e.g 0.002. Necessary as it cannot be easily read in from the trajectory.

Type:

float

volume

Volume of the simulation box

Type:

float

species

A dictionary of the species in the experiment and their properties. Their properties includes index location in the trajectory file, mass of the species as taken from the PubChem database_path, and the charge taken from the same database_path. When using these properties, it is best that users confirm this information, with exception to the indices as they are read from the file and will be correct.

Type:

dict

number_of_atoms

The total number of atoms in the simulation

Type:

int

add_data(simulation_data: Union[str, Path, FileProcessor, list], force: bool = False, update_with_pubchempy: bool = True)[source]

Add data to experiment. This method takes a filename, file path or a file reader (or a list thereof). If given a filename, it will try to instantiate the appropriate file reader with its default arguments. If you have a custom data format with its own reader or want to use non-default arguments for your reader, instantiate the reader and pass it to this method. TODO reference online documentation of data loading in the error messages :param simulation_data: or list thereof

if str or pathlib.Path: path to the file that contains the simulation_data if mdsuite.file_io.file_read.FileProcessor: An already instantiated file reader from mdsuite.file_io if list : must be list of any of the above (can be mixed).

Parameters:

  • force (bool) – If true, a file will be read regardless of if it has already been seen. Default: False

  • update_with_pubchempy (bool) – Whether or not to look for the masses of the species in pubchempy. Default: True.

cls_transformation_run(transformation: Transformations, *args, **kwargs)[source]

Run the transformation.

The Transformation class is updated with this experiment and afterwards performs the transformation. Preliminary work in accordance to https://github.com/zincware/MDSuite/issues/404

Parameters:

transformation (Transformations) –

load_matrix(property_name: str = None, species: ~typing.Iterable[str] = None, select_slice: <numpy.lib.index_tricks.IndexExpression object at 0x7f811e5ba760> = None, path: ~typing.Iterable[str] = None)[source]

Load a desired property matrix.

Parameters:

  • property_name (str) – Name of the matrix to be loaded, e.g. ‘Unwrapped_Positions’, ‘Velocities’

  • species (Iterable[str]) – List of species to be loaded

  • select_slice (np.slice) – A slice to select from the database_path.

  • path (str) – optional path to the database_path.

Returns:

property_matrix – Tensor of the property to be studied. Format depends on kwargs.

Return type:

np.array, tf.Tensor

property run: RunComputation

Method to access the available calculators.

Returns:

class that has all available calculators as properties

Return type:

RunComputation

run_visualization(species: Optional[list] = None, molecules: bool = False, unwrapped: bool = False)[source]

Perform a trajectory visualization.

Parameters:

  • species (list) – A list of species of molecules to study.

  • molecules (bool) – If true, molecule groups will be used.

  • unwrapped (bool) – If true, unwrapped coordinates will be visualized.

Return type:

Displays a visualization app.

set_charge(element: str, charge: float)[source]

Set the charge/s of an element.

Parameters:

  • element (str) – Name of the element whose charge you want to change

  • charge (list) – New charge/s of the element

set_mass(element: str, mass: float)[source]

Set the mass/es of an element.

Parameters:

  • element (str) – Name of the element whose mass you want to change

  • mass (list) – New mass/es of the element

static units_to_si(units_system) → Units[source]

Returns a dictionary with equivalences from the unit experiment given by a string to SI. Along with some constants in the unit experiment provided (boltzmann, or other conversions). Instead, the user may provide a dictionary. In that case, the dictionary will be used as the unit experiment.

Parameters:

  • experiment (units_system (str) -- current unit) –

  • change (dimension (str) -- dimension you would like to) –

Returns:

units – dataclass that contains the conversion factors to SI

Return type:

Units