tanh_ramp_pwc

signals.tanh_ramp_pwc(duration, segment_count, end_value, start_value=None, ramp_duration=None, center_time=None, *, name=None)

Create a Pwc representing a hyperbolic tangent ramp.

Parameters

  • duration (float) – The duration of the signal, TT.
  • segment_count (int) – The number of segments in the PWC.
  • end_value (float or complex or Tensor) – The asymptotic value of the ramp towards t+t \to +\infty, a+a_+. It must either be a scalar or contain a single element.
  • start_value (float or complex or Tensor or None , optional) – The asymptotic value of the ramp towards tt \to -\infty, aa_-. If passed, it must either be a scalar or contain a single element. Defaults to minus end_value.
  • ramp_duration (float or Tensor or None , optional) – The characteristic time for the hyperbolic tangent ramp, trampt_\mathrm{ramp}. If passed, it must either be a scalar or contain a single element. Defaults to T/6T/6.
  • center_time (float or Tensor or None , optional) – The time at which the ramp has its greatest slope, t0t_0. If passed, it must either be a scalar or contain a single element. Defaults to T/2T/2.
  • name (str or None , optional) – The name of the node.

Returns

The sampled hyperbolic tangent ramp.

Return type

Pwc

SEE ALSO

Graph.signals.linear_ramp_pwc : Create a Pwc representing a linear ramp.

boulderopal.signals.tanh_ramp : Create a Signal object representing a hyperbolic tangent ramp.

Graph.signals.tanh_ramp_stf : Corresponding operation with Stf output.

Graph.tanh : Calculate the element-wise hyperbolic tangent of an object.

Notes

The hyperbolic tangent ramp is defined as

Tanh(t)=a++a2+a+a2tanh(tt0tramp), \mathop{\mathrm{Tanh}}(t) = \frac{a_+ + a_-}{2} + \frac{a_+ - a_-}{2} \tanh\left( \frac{t - t_0}{t_\mathrm{ramp}} \right) ,

where the function’s asymptotic values a±a_\pm are defined by:

a±:=limt±Tanh(t), a_\pm := \lim_{t\to\pm\infty} \mathop{\mathrm{Tanh}}(t) ,

and t0t_0 is related to trampt_\mathrm{ramp} by:

dTanh(t)dtt=t0=(a+a)2tramp. \left.\frac{{\rm d}\mathop{\mathrm{Tanh}}(t)}{{\rm d}t}\right|_{t=t_0} = \frac{ (a_+ - a_-)}{2 t_\mathrm{ramp}} .

Note that if t0t_0 is close to the edges of the PWC, for example t02trampt_0 \lesssim 2 t_\mathrm{ramp}, then the first and last values of the PWC will differ from the expected asymptotic values.

With the default values of start_value (aa_-), ramp_duration (trampt_\mathrm{ramp}), and center_time (t0t_0), the ramp expression simplifies to

Tanh(t)=Atanh(tT/2T/6), \mathop{\mathrm{Tanh}}(t) = A \tanh\left( \frac{t - T/2}{T/6} \right),

where A=a+A = a_+ is the end value (the start value is then A-A). This defines a symmetric ramp (around (T/2,0)(T/2, 0)) between 0.995A-0.995 A (at t=0t=0) and 0.995A0.995 A (at t=Tt=T).

Examples

Define a simple tanh PWC ramp.

>>> graph.signals.tanh_ramp_pwc(
...     duration=5.0, segment_count=50, end_value=1, name="tanh_ramp"
... )
<Pwc: name="tanh_ramp", operation_name="discretize_stf", value_shape=(), batch_shape=()>
>>> result = bo.execute_graph(graph=graph, output_node_names="tanh_ramp")
>>> result["output"]["tanh_ramp"]
{
    'durations': array([0.1, 0.1, ..., 0.1, 0.1]),
    'values': array([-0.99442601, -0.99291942, ..., 0.99291942,  0.99442601]),
    'time_dimension': 0
}

Define a flat-top pulse from two hyperbolic tangent ramps.

>>> ramp = graph.signals.tanh_ramp_pwc(
...     duration=3,
...     segment_count=60,
...     end_value=1,
...     ramp_duration=0.25,
...     center_time=0.5,
... )
>>> flat_top_pulse = 0.5 * (ramp + graph.time_reverse_pwc(ramp))
>>> flat_top_pulse.name="flat_top_pulse"
>>> result = bo.execute_graph(graph=graph, output_node_names="flat_top_pulse")
>>> result["output"]["flat_top_pulse"]
{
    'durations': array([0.05, 0.05, ..., 0.05, 0.05]),
    'values': array([0.02188127, 0.03229546, ..., 0.03229546, 0.02188127]),
    'time_dimension': 0
}

Define a hyperbolic tangent ramp with optimizable parameters.

>>> end_value = graph.optimizable_scalar(
...     lower_bound=0, upper_bound=3e6, name="end_value"
... )
>>> ramp_duration = graph.optimizable_scalar(
...     lower_bound=0.1e-6, upper_bound=0.3e-6, name="ramp_duration"
... )
>>> center_time = graph.optimizable_scalar(
...     lower_bound=0.25e-6, upper_bound=0.75e-6, name="center_time"
... )
>>> graph.signals.tanh_ramp_pwc(
...     duration=1e-6,
...     segment_count=32,
...     end_value=end_value,
...     start_value=0.0,
...     ramp_duration=ramp_duration,
...     center_time=center_time,
...     name="tanh_ramp",
... )
<Pwc: name="tanh_ramp", operation_name="discretize_stf", value_shape=(), batch_shape=()>

Was this useful?