NearestTraceReducer

class NearestTraceReducer(step_time: float, time_constant: float, amplitude: int | float | complex, target: int | float | bool | complex, tolerance: int | float | None = None, *, duration: float = 0.0, inclusive: bool = False, inplace: bool = False)

Bases: FoldReducer

Stores the trace over time, considering the latest match.

\[\begin{split}x(t) = \begin{cases} A & \lvert h(t) - h^* \rvert \leq \epsilon \\ x(t - \Delta t) \exp \left(-\frac{\Delta t}{\tau_x}\right) & \lvert h(t) - h^* \rvert > \epsilon \end{cases}\end{split}\]

For the trace (state) \(x\) and observation \(h\).

Parameters:

• step_time (float) – length of the discrete step time, \(\Delta t\).

• time_constant (float) – time constant of exponential decay, \(\tau_x\).

• amplitude (int | float | complex) – value to set trace to for matching elements, \(A\).

• target (int | float | bool | complex) – target value test for when determining if an input is a match, \(h^*\).

• tolerance (int | float | None, optional) – allowable absolute difference to still count as a match, \(\epsilon\). Defaults to None.

• duration (float, optional) – length of time over which results should be stored, in the same units as \(\Delta t\). Defaults to 0.0.

• inclusive (bool, optional) – if the duration should be inclusive. Defaults to False.

• inplace (bool, optional) – if write operations should be performed in-place. Defaults to False.

Important

Because the input tensor to fold() is treated as an event condition, it will have its datatype cast to that of the reducer itself.

property dt: float

Length of the simulation time step, in milliseconds.

Parameters:

value (float) – new simulation time step length.

Returns:

length of the simulation time step.

Return type:

float

fold(obs: Tensor, state: Tensor | None) → Tensor

Application of nearest trace.

Parameters:

• obs (torch.Tensor) – observation to incorporate into state.

• state (torch.Tensor | None) – state from the prior time step, None if no prior observations.

Returns:

state for the current time step.

Return type:

torch.Tensor

interpolate(prev_data: Tensor, next_data: Tensor, sample_at: Tensor, step_time: float) → Tensor

Exponential decay interpolation between observations.

Parameters:

• prev_data (torch.Tensor) – most recent observation prior to sample time.

• next_data (torch.Tensor) – most recent observation subsequent to sample time.

• sample_at (torch.Tensor) – relative time at which to sample data.

• step_time (float) – length of time between the prior and subsequent observations.

Returns:

interpolated data at sample time.

Return type:

torch.Tensor