Optimizers API

OptimizationProperty

Defines one optimization variable.

Fields include:

• name
• type (MODEL_INPUT or NETLIST_VARIABLE)
• min_value
• max_value
• step
• optional unit
• optional linked_to

OptimizationType

• MODEL_INPUT: variable passed to ONNX surrogate inputs.
• NETLIST_VARIABLE: variable patched directly in parsed netlist content.

DesignGoal and DesignParameter

DesignGoal combines:

• parameter enum (DesignParameter),
• optional min/max bounds,
• optional frequency_range,
• optional weight.

Penalty is computed from normalized squared bound violations.

OptunaOptimizer

Default optimizer wrapper for Optuna.

Typical options:

• sampler: tpe, random, simulated_annealing, auto
• pruner: median, successive_halving, hyperband
• multi_objective: boolean

GradientDescentOptimizer

Used for local refinement workflows, including optional fine-tuning mode in COBRA.

Tip

Start with OptunaOptimizer and move to gradient descent only when you are refining near a known good region.