Adaptive Buffer¶

class pathsim.utils.adaptivebuffer.AdaptiveBuffer(delay)[source]¶

Bases: object

A class that manages an adaptive buffer for delay modeling which is primarily used in the pathsim ‘Delay’ block but might have future applications aswell.

It implements a linear interpolation for arbitrary time lookup.

Parameters:: delay (float) – time delay in seconds

buffer_t¶

deque that collects the time data for buffering

Type:: deque

buffer_v¶

deque that collects the value data for buffering

Type:: deque

ns¶

safety for buffer truncation

Type:: int

add(t, value)[source]¶

adding a new datapoint to the buffer

Parameters:

t (float) – time to add
value (float, int, complex) – numerical value to add

interp(t)[source]¶

interpolate buffer at defined lookup time

Parameters:: t (float) – time for interpolation
Returns:: out – interpolated value
Return type:: float, array

get(t)[source]¶

lookup datapoint from buffer with delay at t_lookup = t - delay

Parameters:: t (float) – time for lookup with delay

clear()[source]¶: clear the buffer, reset everything

to_checkpoint(prefix)[source]¶

Serialize buffer state for checkpointing.

Parameters:: prefix (str) – NPZ key prefix
Returns:: npz_data – numpy arrays keyed by path
Return type:: dict

load_checkpoint(npz, prefix)[source]¶

Restore buffer state from checkpoint.

Parameters:

npz (dict-like) – numpy arrays from checkpoint NPZ
prefix (str) – NPZ key prefix