Struct OrtInfer 

pub struct OrtInfer { /* private fields */ }

Core ONNX Runtime inference engine with support for pooling and configurable sessions.

Implementations

impl OrtInfer

pub fn new( model_path: impl AsRef<Path>, input_name: Option<&str>, ) -> Result<OrtInfer, OCRError>

Creates a new OrtInfer instance with default ONNX Runtime settings and a single session.

pub fn from_config( common: &ModelInferenceConfig, model_path: impl AsRef<Path>, input_name: Option<&str>, ) -> Result<OrtInfer, OCRError>

Creates a new OrtInfer instance from ModelInferenceConfig, applying ORT session configuration.

pub fn with_output_name( model_path: impl AsRef<Path>, input_name: Option<&str>, output_name: Option<&str>, ) -> Result<OrtInfer, OCRError>

Creates a new OrtInfer instance with a specified output tensor name.

impl OrtInfer

pub fn model_path(&self) -> &Path

Returns the model path associated with this inference engine.

pub fn model_name(&self) -> &str

Returns the model name associated with this inference engine.

pub fn infer_4d( &self, x: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, ) -> Result<ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, OCRError>

pub fn infer_2d( &self, x: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, ) -> Result<ArrayBase<OwnedRepr<f32>, Dim<[usize; 2]>>, OCRError>

pub fn infer_3d( &self, x: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, ) -> Result<ArrayBase<OwnedRepr<f32>, Dim<[usize; 3]>>, OCRError>

pub fn infer_2d_i64( &self, x: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, ) -> Result<ArrayBase<OwnedRepr<i64>, Dim<[usize; 2]>>, OCRError>

Runs inference and returns a 2D int64 tensor.

This is useful for models that output token IDs (e.g., formula recognition). The output shape is typically [batch_size, sequence_length].

pub fn infer_dual_3d( &self, x: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, ) -> Result<(ArrayBase<OwnedRepr<f32>, Dim<[usize; 3]>>, ArrayBase<OwnedRepr<f32>, Dim<[usize; 3]>>), OCRError>

Runs inference for models with dual 3D outputs.

This is used for models like SLANet that output two 3D tensors. The first output is typically structure/token predictions, and the second is bounding box predictions or similar auxiliary outputs.

Returns

A tuple of two 3D tensors: (first_output, second_output)

pub fn infer_4d_layout( &self, x: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, scale_factor: Option<ArrayBase<OwnedRepr<f32>, Dim<[usize; 2]>>>, im_shape: Option<ArrayBase<OwnedRepr<f32>, Dim<[usize; 2]>>>, ) -> Result<ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, OCRError>

Runs inference with multiple inputs for layout detection models.

Layout detection models typically require:

• image: The preprocessed image tensor [N, 3, H, W]
• scale_factor: Scale factors used during preprocessing [N, 2] (for PicoDet)
• im_shape: Original image shape [N, 2] (for PP-DocLayout)

impl OrtInfer

pub fn input_name(&self) -> &str

Returns the input tensor name.

pub fn get_session( &self, idx: usize, ) -> Result<MutexGuard<'_, Session>, OCRError>

Gets a session from the pool.

pub fn primary_input_shape(&self) -> Option<Vec<i64>>

Attempts to retrieve the primary input tensor shape from the first session.

Returns a vector of dimensions if available. Dynamic dimensions (e.g., -1) are returned as-is.

Trait Implementations

impl Debug for OrtInfer

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

Formats the value using the given formatter.

impl From<OrtInfer> for OrtInfer2D

fn from(inner: OrtInfer) -> OrtInfer2D

Converts to this type from the input type.

impl From<OrtInfer> for OrtInfer3D

fn from(inner: OrtInfer) -> OrtInfer3D

Converts to this type from the input type.

impl From<OrtInfer> for OrtInfer4D

fn from(inner: OrtInfer) -> OrtInfer4D

Converts to this type from the input type.