Struct SemanticModel 

pub struct SemanticModel {}

Wrapper around an ONNX Runtime session for the embedding model.

Implementations

impl SemanticModel

pub fn load() -> Result<Self>

Load the model from the OS cache directory.

If the cache file is missing or has a mismatched SHA256, this extracts bundled model bytes into the cache directory before creating an ORT session. When no bundled bytes are available, this attempts a one-time download of model/tokenizer assets from stable URLs.

Errors

Returns an error if the model cannot be loaded or the cache is invalid.

pub fn model_cache_path() -> Result<PathBuf>

Return the OS-appropriate cache path for the model file.

Uses dirs::cache_dir() / bones / models / minilm-l6-v2-int8.onnx.

Errors

Returns an error if the OS cache directory cannot be determined.

pub fn is_cached_valid(path: &Path) -> bool

Check if cached model matches expected SHA256.

pub fn extract_to_cache(path: &Path) -> Result<()>

Extract bundled model bytes to cache directory.

Errors

Returns an error if no bundled bytes are available, or if the file system operations fail, or if the extracted model fails SHA256 verification.

pub fn embed(&self, text: &str) -> Result<Vec<f32>>

Run inference for a single text input.

Errors

Returns an error if the runtime is unavailable or inference fails.

pub fn embed_batch(&self, texts: &[&str]) -> Result<Vec<Vec<f32>>>

Batch inference for efficiency.

Errors

Returns an error if the runtime is unavailable or batch inference fails.