Tau

class mlptrain.loss.tau.Tau(x, error: float | None = None)

Bases: LossValue

class mlptrain.loss.tau.TauCalculator(e_lower: float = 0.1, e_thresh: float | None = None, max_time: float = 1000.0, time_interval: float = 50.0, temp: float = 300.0, dt: float = 0.5)

Bases: LossFunction

__call__(configurations: ConfigurationSet, mlp: MLPotential, **kwargs) Tau

Calculate τ_acc from a set of initial configurations

Parameters:

mlp – Machine learnt potential

Returns:

τ_acc

Return type:

(Tau)

__init__(e_lower: float = 0.1, e_thresh: float | None = None, max_time: float = 1000.0, time_interval: float = 50.0, temp: float = 300.0, dt: float = 0.5)

τ_acc prospective error metric in fs

error possible in the system

Parameters:
  • e_thresh – (float | None) E_t total cumulative error in eV. τ_acc is defined at the time in the simulation where this threshold is exceeded. If None then: e_thresh = 10 * e_lower

  • max_time – (float) Maximum time in femto-seconds for τ_acc

  • time_interval – (float) Interval between which |E_true - E_GAP| is calculated. Must be at least one timestep

  • temp – (float) Temperature of the simulation to perform

  • dt – (float) Timestep of the simulation in femto-seconds