Struct NodeModelConfig 

pub struct NodeModelConfig {
    pub name: String,
    pub provider: Option<Provider>,
    pub path: String,
    pub source: ModelSource,
    pub version: Option<i32>,
    pub confidence_threshold: Option<f32>,
}

A node’s reference to the on-disk model it should load - shared/reserved across every built-in node that has this concept (object-detector, image-classifier, rl), moved onto crate::node_config_base::NodeConfigBase (LAB-4384 follow-up) the same way relay_topics already was, rather than each of those nodes re-declaring its own near-identical ModelConfig struct.

This is deliberately a different type from ModelConfig above: that one is the model catalog’s own on-disk metadata (models/catalog/<name>/config.json - task, repo, filename, preprocessing, labels); this one is a node’s pointer to a model (which one, where its file lives, where to resolve it from), a completely different concept that happens to share the word “model”.

name/path default to the empty string rather than this type baking in any one node’s real default value (object-detector defaults to "yolov8n", rl to "find-target", image-classifier to "mobilenet-v2" - genuinely different per node). Each node resolves its own real default for an empty value itself (see each node’s own model() accessor on its Config type) - this mirrors exactly how #[serde(default)] on this struct’s own fields already behaved before this type was shared (calling Default::default(), not a per-node default function, when the field - or the whole model key - is merely absent).

confidence_threshold only has meaning for image-classifier today (object-detector/rl ignore it) - it lives here rather than as a image-classifier-only extra field so this struct stays the single shared shape for the “which model does this node use” concept across all three nodes rather than splintering into near-identical siblings again.

provider only has meaning for llm-command (object-detector/image-classifier/rl ignore it) - it used to be a separate field on NodeConfigBase alongside model, but “which backend resolves this model” is part of the same “which model does this node use” concept, so it consolidated into this one struct instead of two separate Option fields on the base.

Fields

name: String

Name of the model to load. Empty means “not set” - the owning node resolves its own default in that case.

provider: Option<Provider>

Which LLM backend to query (llm-command-specific; ignored by other nodes using this type). None means “not set” - the owning node resolves its own default in that case.

path: String

Filesystem path to the model file. Empty means “not set” - the owning node resolves its own default in that case.

source: ModelSource

Where to resolve the model from.

version: Option<i32>

Specific model version to use (registry-sourced models only); latest if omitted.

confidence_threshold: Option<f32>

Minimum confidence score a classification result must meet to be reported (image-classifier-specific; ignored by other nodes using this type).

Trait Implementations

impl Clone for NodeModelConfig

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for NodeModelConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for NodeModelConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for NodeModelConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for NodeModelConfig

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl PartialEq for NodeModelConfig

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for NodeModelConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for NodeModelConfig