nemoguardrails.library.factchecking.align_score.server

Module Contents

Classes

Name	Description
`AlignScoreRequest`	-

Functions

Name	Description
`alignscore_base`	-
`alignscore_large`	-
`get_alignscore`	-
`get_model`	Initialize a model.
`hello_world`	-
`start`	-

Data

API

class nemoguardrails.library.factchecking.align_score.server.AlignScoreRequest()

Bases: BaseModel

claim

str

evidence

str

nemoguardrails.library.factchecking.align_score.server.alignscore_base(
    request: nemoguardrails.library.factchecking.align_score.server.AlignScoreRequest
)

nemoguardrails.library.factchecking.align_score.server.alignscore_large(
    request: nemoguardrails.library.factchecking.align_score.server.AlignScoreRequest
)

nemoguardrails.library.factchecking.align_score.server.get_alignscore(
    model,
    evidence: str,
    claim: str
) -> dict

nemoguardrails.library.factchecking.align_score.server.get_model(
    model: str
)

Initialize a model.

Args model: The type of the model to be loaded, i.e. “base”, “large”.

nemoguardrails.library.factchecking.align_score.server.hello_world()

nemoguardrails.library.factchecking.align_score.server.start(
    port: int = typer.Option(default=5000, ...,
    models: typing.List[str] = typer.Option(default=['base...,
    initialize_only: bool = typer.Option(default=False,...
)

nemoguardrails.library.factchecking.align_score.server.app = FastAPI()

nemoguardrails.library.factchecking.align_score.server.cli_app = typer.Typer()

nemoguardrails.library.factchecking.align_score.server.device = os.environ.get('ALIGN_SCORE_DEVICE', 'cpu')

nemoguardrails.library.factchecking.align_score.server.models_path = os.environ.get('ALIGN_SCORE_PATH')