You're reading the documentation of the 0.3. For the latest released version, please have a look at 0.4.

merlin.measurement.autodiff module

Automatic differentiation handling for sampling.

class merlin.measurement.autodiff.AutoDiffProcess(sampling_method='multinomial')

Bases: object

Handles automatic differentiation backend and sampling noise integration.

Parameters:

sampling_method (str) – Sampling method used to initialize the internal sampling process.

autodiff_backend(needs_gradient, apply_sampling, shots)

Determine sampling configuration based on gradient requirements.

Parameters:

• needs_gradient (bool) – Whether gradients are currently required.

• apply_sampling (bool) – Whether sampling was requested.

• shots (int) – Requested number of shots.

Returns:

Effective sampling flag and effective number of shots.

Return type:

tuple[bool, int]