bloqade.gemini.device.logical.result.GeminiLogicalResult

← Module overview

classGeminiLogicalResult¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult

Bases: Result, Generic[RetType]

Result view over stored Gemini logical shots.

class GeminiLogicalResult

Public import paths

• from bloqade.gemini import GeminiLogicalResult
• from bloqade.gemini.device import GeminiLogicalResult
• from bloqade.gemini.device.logical import GeminiLogicalResult
• from bloqade.gemini.device.logical.future import GeminiLogicalResult

Merge-oriented methods assume each selected task ID has the same subtask structure. Post-processing is applied to each selected subtask’s flat shot array.

Raw SLM result properties reconstruct their measurement mapping by compiling the stored logical kernel with the locally installed Bloqade Lanes compiler and Gemini physical architecture. The reconstructed mapping is valid only when those match the compiler and architecture used for remote execution.

Attributes

Name	Type	Description
storage	StorageBackend	Storage backend that holds shots and task metadata.
shot_filter	ShotFilter	Filter used when reading shots and deriving subtask scope. Defaults to the DETECTED frame type.

methodpostprocessing_functions¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.postprocessing_functions

def postprocessing_functions() -> dict[int, Callable | None]

Build legacy compact-shot postprocessors from stored programs.

This method preserves the original GeminiLogicalResult API for result stores whose bitstrings are already in compact physical measurement order. Raw 160-site SLM frames use _slm_postprocessing_functions instead.

Returns

dict[int, Callable | None]

source

propertymeasurements¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.measurements

measurements: Sequence[Sequence[Sequence[bool]]]

True indicates a projective measurement of the |1> state or atom loss; False indicates a projective measurement of the |0> state.

source

methodlogical_results¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.logical_results

Signature

def logical_results(verify: bool = True, postprocessing_functions: dict[int, Callable[[np.ndarray], RetType] | None] | None = None) -> list[RetType | np.ndarray]

Return legacy post-processed results grouped by merged subtask.

This method preserves the original compact-shot result API. Each postprocessor receives the stored shot array directly; raw 160-site SLM results should instead be accessed through return_values and the other canonical result properties.

Parameters

Name	Type	Default	Description
verify	bool	True	Validate that selected task IDs can be merged.
postprocessing_functions	dict[int, Callable[[np.ndarray], RetType] | None] | None	None	Optional mapping from program index to a function accepting that subtask's stored shot array. When omitted, legacy postprocessors are generated from the stored logical kernels.

Returns

list[RetType | np.ndarray]

source

propertyreturn_values¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.return_values

return_values: Sequence[Sequence[RetType]]

Return canonical logical values reconstructed from raw SLM shots.

Results are grouped as subtask -> shot -> kernel return value. Use logical_results only for the legacy compact-shot API.

source

propertydetectors¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.detectors

detectors: Sequence[Sequence[Sequence[bool]]]

source

propertyobservables¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.observables

observables: Sequence[Sequence[Sequence[bool]]]

source

propertyfilling_at_start¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.filling_at_start

filling_at_start: Sequence[Sequence[Sequence[bool]]]

True indicates that the atom was present during the sorted frame; False indicates that it was not.

source

methodpostselect_on_fully_filled¶source

bloqade.gemini.device.logical.result.GeminiLogicalResult.postselect_on_fully_filled

def postselect_on_fully_filled() -> GeminiLogicalResult[RetType]

Return a result view containing only fully filled shots.

The predicate reads the SORTED SLM frame and projects it through the stored kernel’s SLM-to-raw mapping. A shot is kept only when every mapped physical-measurement bit is true. The returned result preserves this result’s output frame selection, which is normally DETECTED.

Returns

GeminiLogicalResult[RetType]

source