DPGEN2 configurations

Op configs

RunDPTrain

init_model_start_pref_v:
type: float, optional, default: 0.0
argument path: init_model_start_pref_v

The start virial prefactor in loss when init-model

init_model_start_pref_f:
type: int | float, optional, default: 100
argument path: init_model_start_pref_f

The start force prefactor in loss when init-model

init_model_start_pref_e:
type: float, optional, default: 0.1
argument path: init_model_start_pref_e

The start energy prefactor in loss when init-model

init_model_start_lr:
type: float, optional, default: 0.0001
argument path: init_model_start_lr

The start learning rate when init-model

init_model_numb_steps:
type: int, optional, default: 400000, alias: init_model_stop_batch
argument path: init_model_numb_steps

The number of training steps when init-model

init_model_old_ratio:
type: float, optional, default: 0.9
argument path: init_model_old_ratio

The frequency ratio of old data over new data

init_model_policy:
type: str, optional, default: no
argument path: init_model_policy

The policy of init-model training. It can be

  • ‘no’: No init-model training. Traing from scratch.

  • ‘yes’: Do init-model training.

  • ‘old_data_larger_than:XXX’: Do init-model if the training data size of the previous model is larger than XXX. XXX is an int number.

RunLmp

command:
type: str, optional, default: lmp
argument path: command

The command of LAMMPS

RunVasp

out:
type: str, optional, default: data
argument path: out

The output dir name of labeled data. In deepmd/npy format provided by dpdata.

log:
type: str, optional, default: vasp.log
argument path: log

The log file name of VASP

command:
type: str, optional, default: vasp
argument path: command

The command of VASP

Alloy configs

fmt:
type: str, optional, default: lammps/lmp
argument path: fmt

The format of file content

atom_pert_dist:
type: float, optional, default: 0.0
argument path: atom_pert_dist

The distance of atomic position perturbation

cell_pert_frac:
type: float, optional, default: 0.0
argument path: cell_pert_frac

The faction of cell perturbation

concentration:
type: list | NoneType, optional, default: None
argument path: concentration

The concentration of each element. If None all elements have the same concentration

numb_confs:
type: int, optional, default: 1
argument path: numb_confs

The number of configurations to generate

replicate:
type: list | NoneType, optional, default: None
argument path: replicate

The number of replicates in each direction

type_map:
type: list
argument path: type_map

The type map of the system

lattice:
type: list | tuple
argument path: lattice

The lattice. Should be a list providing [ “lattice_type”, lattice_const ], or a list providing [ “/path/to/dpdata/system”, “fmt” ]. The two styles are distinguished by the type of the second element.

Task group configs

task_group_configs:
type: dict
argument path: task_group_configs

Depending on the value of type, different sub args are accepted.

type:
type: str (flag key)
argument path: task_group_configs/type
possible choices: lmp-md, lmp-template

the type of the task group

When type is set to lmp-md (or its alias lmp-npt):

temps:
type: list, alias: Ts
argument path: task_group_configs[lmp-md]/temps

A list of temperatures in K. Also used to initialize the temperature

press:
type: list, optional, alias: Ps
argument path: task_group_configs[lmp-md]/press

A list of pressures in bar.

ens:
type: str, optional, default: nve, alias: ensemble
argument path: task_group_configs[lmp-md]/ens

The ensemble. Allowd options are ‘nve’, ‘nvt’, ‘npt’, ‘npt-a’, ‘npt-t’. ‘npt-a’ stands for anisotrpic box sampling and ‘npt-t’ stands for triclinic box sampling.

dt:
type: float, optional, default: 0.001
argument path: task_group_configs[lmp-md]/dt

The time step

nsteps:
type: int, optional, default: 100
argument path: task_group_configs[lmp-md]/nsteps

The number of steps

trj_freq:
type: int, optional, default: 10, aliases: t_freq, trj_freq, traj_freq
argument path: task_group_configs[lmp-md]/trj_freq

The number of steps

tau_t:
type: float, optional, default: 0.05
argument path: task_group_configs[lmp-md]/tau_t

The time scale of thermostat

tau_p:
type: float, optional, default: 0.5
argument path: task_group_configs[lmp-md]/tau_p

The time scale of barostat

pka_e:
type: NoneType | float, optional, default: None
argument path: task_group_configs[lmp-md]/pka_e

The energy of primary knock-on atom

neidelay:
type: int | NoneType, optional, default: None
argument path: task_group_configs[lmp-md]/neidelay

The delay of updating the neighbor list

no_pbc:
type: bool, optional, default: False
argument path: task_group_configs[lmp-md]/no_pbc

Not using the periodic boundary condition

use_clusters:
type: bool, optional, default: False
argument path: task_group_configs[lmp-md]/use_clusters

Calculate atomic model deviation

relative_f_epsilon:
type: NoneType | float, optional, default: None
argument path: task_group_configs[lmp-md]/relative_f_epsilon

Calculate relative force model deviation

relative_v_epsilon:
type: NoneType | float, optional, default: None
argument path: task_group_configs[lmp-md]/relative_v_epsilon

Calculate relative virial model deviation

When type is set to lmp-template:

lmp_template_fname:
type: str, aliases: lmp_template, lmp
argument path: task_group_configs[lmp-template]/lmp_template_fname

The file name of lammps input template

plm_template_fname:
type: NoneType | str, optional, default: None, aliases: plm_template, plm
argument path: task_group_configs[lmp-template]/plm_template_fname

The file name of plumed input template

revisions:
type: dict, optional, default: {}
argument path: task_group_configs[lmp-template]/revisions
traj_freq:
type: int, optional, default: 10, aliases: t_freq, trj_freq, trj_freq
argument path: task_group_configs[lmp-template]/traj_freq

The frequency of dumping configurations and thermodynamic states

Step configs

executor:
type: dict | NoneType, optional, default: None
argument path: executor

The executor of the step.

Depending on the value of type, different sub args are accepted.

type:
type: str (flag key)
argument path: executor/type
possible choices: lebesgue_v2, dispatcher

The type of the executor.

When type is set to lebesgue_v2:

extra:
type: dict, optional
argument path: executor[lebesgue_v2]/extra

The ‘extra’ key in the lebesgue executor. Note that we do not check if ‘the dict provided to the ‘extra’ key is valid or not.

scass_type:
type: str, optional
argument path: executor[lebesgue_v2]/extra/scass_type

The machine configuraiton.

program_id:
type: str, optional
argument path: executor[lebesgue_v2]/extra/program_id

The ID of the program.

job_type:
type: str, optional, default: container
argument path: executor[lebesgue_v2]/extra/job_type

The type of job.

template_cover_cmd_escape_bug:
type: bool, optional, default: True
argument path: executor[lebesgue_v2]/extra/template_cover_cmd_escape_bug

The key for hacking around a bug in Lebesgue.

When type is set to dispatcher:

parallelism:
type: int | NoneType, optional, default: None
argument path: parallelism

The parallelism for the step

continue_on_success_ratio:
type: NoneType | float, optional, default: None
argument path: continue_on_success_ratio

Only in the sliced OP case. Continue the workflow if a certain ratio of the sliced jobs are successful.

continue_on_num_success:
type: int | NoneType, optional, default: None
argument path: continue_on_num_success

Only in the sliced OP case. Continue the workflow if a certain number of the sliced jobs are successful.

continue_on_failed:
type: bool, optional, default: False
argument path: continue_on_failed

If continue the the step is failed (FatalError, TransientError, A certain number of retrial is reached…).

template_config:
type: dict, optional, default: {'image': 'dptechnology/dpgen2:latest'}
argument path: template_config

The configs passed to the PythonOPTemplate.

image:
type: str, optional, default: dptechnology/dpgen2:latest
argument path: template_config/image

The image to run the step.

timeout:
type: int | NoneType, optional, default: None
argument path: template_config/timeout

The time limit of the OP. Unit is second.

retry_on_transient_error:
type: NoneType | bool, optional, default: None
argument path: template_config/retry_on_transient_error

Retry the step if a TransientError is raised.

timeout_as_transient_error:
type: bool, optional, default: False
argument path: template_config/timeout_as_transient_error

Treat the timeout as TransientError.

envs:
type: dict | NoneType, optional, default: None
argument path: template_config/envs

The environmental variables.