qualia_core.postprocessing package

Submodules

Module contents

class qualia_core.postprocessing.PostProcessing[source]

Bases: ABC, Generic[T]

process_name(name: str) → str[source]

process_framework(framework: LearningFramework[T]) → LearningFramework[Any][source]

process_mem_params(mem_params: int) → Callable[[LearningFramework[T], T], int][source]

process_model(model: T, model_conf: ModelConfigDict, framework: LearningFramework[T]) → tuple[T, ModelConfigDict][source]

class qualia_core.postprocessing.QualiaCodeGen(quantize: str, long_width: int | None = None, outdir: str | None = None, metrics: list[str] | None = None)[source]

Bases: Converter[Any]

deployers: ModuleType | None = <module 'qualia_core.deployment.qualia_codegen' from '/home/runner/work/qualia-core/qualia-core/src/qualia_core/deployment/qualia_codegen/__init__.py'>: Suggested deployers

convert_model_to_modelgraph(model: nn.Module | keras.Model) → ModelGraph | None[source]

convert_modelgraph_to_c(modelgraph: ModelGraph, output_path: Path) → str | None[source]

convert_metrics_to_cpp(metrics: list[str], output_path: Path) → str | None[source]

convert(framework: LearningFramework[nn.Module | keras.Model], model: nn.Module | keras.Model, model_name: str, representative_dataset: numpy.typing.NDArray[Any]) → QualiaCodeGen | None[source]

property h: str | None

property name: str | None

property directory: Path | None

process_mem_params(mem_params: int) → Callable[[LearningFramework[nn.Module | keras.Model], nn.Module | keras.Model], int][source]