Function profile_cli

spacy/cli/profile.py:20–49

Profile which functions take the most time in a spaCy pipeline. Input should be formatted as one JSON object per line with a key "text". It can either be provided as a JSONL file, or be read from sys.sytdin. If no input file is specified, the IMDB dataset is loaded via Thinc. D

(
    # fmt: off
    ctx: typer.Context,  # This is only used to read current calling context
    model: str = Arg(..., help="Trained pipeline to load"),
    inputs: Optional[Path] = Arg(
        None,
        help="Location of input file. '-' for stdin.",
        exists=True,
        allow_dash=True,
    ),
    n_texts: int = Opt(
        10000, "--n-texts", "-n", help="Maximum number of texts to use if available"
    ),
    # fmt: on
)

@debug_cli.command("profile")
@app.command("profile", hidden=True)
def profile_cli(
    # fmt: off
    ctx: typer.Context,  # This is only used to read current calling context
    model: str = Arg(..., help="Trained pipeline to load"),
    inputs: Optional[Path] = Arg(
        None,
        help="Location of input file. '-' for stdin.",
        exists=True,
        allow_dash=True,
    ),
    n_texts: int = Opt(
        10000, "--n-texts", "-n", help="Maximum number of texts to use if available"
    ),
    # fmt: on
):
    """
    Profile which functions take the most time in a spaCy pipeline.
    Input should be formatted as one JSON object per line with a key "text".
    It can either be provided as a JSONL file, or be read from sys.sytdin.
    If no input file is specified, the IMDB dataset is loaded via Thinc.

    DOCS: https://spacy.io/api/cli#debug-profile
    """
    if ctx.parent.command.name == NAME:  # type: ignore[union-attr]    # called as top-level command
        msg.warn(
            "The profile command is now available via the 'debug profile' "
            "subcommand. You can run python -m spacy debug --help for an "
            "overview of the other available debugging commands."
        )
    profile(model, inputs=inputs, n_texts=n_texts)


def profile(model: str, inputs: Optional[Path] = None, n_texts: int = 10000) -> None: