pixano_inference.configs.deployment

Deployment configuration of a single model: the object every model is constructed with.

These classes are plain pydantic models. They belong to the model API, not to the Ray Serve runtime, so a model package imports them without loading the server stack.

AutoscalingConfig(**data)

Bases: BaseModel

Autoscaling configuration for Ray Serve deployments.

Attributes:

Name	Type	Description
min_replicas	int	Minimum number of replicas. Can be 0 for scale-to-zero.
max_replicas	int	Maximum number of replicas.
target_num_ongoing_requests_per_replica	int	Target number of ongoing requests per replica before scaling up.
downscale_delay_s	float	Delay in seconds before scaling down.
upscale_delay_s	float	Delay in seconds before scaling up.

Source code in pydantic/main.py

def __init__(self, /, **data: Any) -> None:
    """Create a new model by parsing and validating input data from keyword arguments.

    Raises [`ValidationError`][pydantic_core.ValidationError] if the input data cannot be
    validated to form a valid model.

    `self` is explicitly positional-only to allow `self` as a field name.
    """
    # `__tracebackhide__` tells pytest and some other tools to omit this function from tracebacks
    __tracebackhide__ = True
    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
    if self is not validated_self:
        warnings.warn(
            'A custom validator is returning a value other than `self`.\n'
            "Returning anything other than `self` from a top level model validator isn't supported when validating via `__init__`.\n"
            f'See the `model_validator` docs (https://pydantic.dev/docs/validation/{version_short()}/concepts/validators/#model-validators) for more details.',
            stacklevel=2,
        )

ModelDeploymentConfig(**data)

Bases: BaseModel

Configuration for deploying a single model.

Attributes:

Name	Type	Description
name	str	Unique model name. Optional for HuggingFace models (auto-derived from path).
capability	str	Capability string (e.g. "segmentation").
model_class	str	Registered class name (e.g. "MyDetector"), provided by an installed model package.
model_params	dict	Parameters passed to model init via config.
resources	ResourceConfig	Resource configuration for the deployment.
autoscaling	AutoscalingConfig	Autoscaling configuration for the deployment.
max_batch_size	int	Maximum batch size for inference (1 disables batching).
batch_wait_timeout_s	float	Timeout for waiting to fill batch.
max_ongoing_requests	int	Max concurrent requests Serve routes to one replica before queueing / triggering autoscaling.
health_check_period_s	float	How often Serve calls the replica health check.
timeout_s	float | None	Per-request inference timeout. None uses the capability default.

Source code in pydantic/main.py

def __init__(self, /, **data: Any) -> None:
    """Create a new model by parsing and validating input data from keyword arguments.

    Raises [`ValidationError`][pydantic_core.ValidationError] if the input data cannot be
    validated to form a valid model.

    `self` is explicitly positional-only to allow `self` as a field name.
    """
    # `__tracebackhide__` tells pytest and some other tools to omit this function from tracebacks
    __tracebackhide__ = True
    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
    if self is not validated_self:
        warnings.warn(
            'A custom validator is returning a value other than `self`.\n'
            "Returning anything other than `self` from a top level model validator isn't supported when validating via `__init__`.\n"
            f'See the `model_validator` docs (https://pydantic.dev/docs/validation/{version_short()}/concepts/validators/#model-validators) for more details.',
            stacklevel=2,
        )

ResourceConfig(**data)

Bases: BaseModel

Resource configuration for a model deployment.

Attributes:

Name	Type	Description
num_gpus	float	Number of GPUs per replica.
num_cpus	float	Number of CPUs per replica.
memory_mb	int | None	Memory limit in MB. None means no limit.

Source code in pydantic/main.py

def __init__(self, /, **data: Any) -> None:
    """Create a new model by parsing and validating input data from keyword arguments.

    Raises [`ValidationError`][pydantic_core.ValidationError] if the input data cannot be
    validated to form a valid model.

    `self` is explicitly positional-only to allow `self` as a field name.
    """
    # `__tracebackhide__` tells pytest and some other tools to omit this function from tracebacks
    __tracebackhide__ = True
    validated_self = self.__pydantic_validator__.validate_python(data, self_instance=self)
    if self is not validated_self:
        warnings.warn(
            'A custom validator is returning a value other than `self`.\n'
            "Returning anything other than `self` from a top level model validator isn't supported when validating via `__init__`.\n"
            f'See the `model_validator` docs (https://pydantic.dev/docs/validation/{version_short()}/concepts/validators/#model-validators) for more details.',
            stacklevel=2,
        )