Struct ModularPredictor 

pub struct ModularPredictor<R, P, I, O> {
    pub image_reader: R,
    pub preprocessor: P,
    pub inference_engine: I,
    pub postprocessor: O,
}

A modular predictor that composes granular components.

This struct demonstrates how to build a complete predictor using the granular traits. It provides the same interface as StandardPredictor but with composable components.

Fields

image_reader: R

Image reader component

preprocessor: P

Preprocessor component

inference_engine: I

Inference engine component

postprocessor: O

Postprocessor component

Implementations

impl<R, P, I, O> ModularPredictor<R, P, I, O>

where R: ImageReader, P: Preprocessor, I: InferenceEngine<Input = P::Output>, O: Postprocessor<InferenceOutput = I::Output, PreprocessOutput = P::Output>,

pub fn new( image_reader: R, preprocessor: P, inference_engine: I, postprocessor: O, ) -> Self

Create a new modular predictor with the given components.

Trait Implementations

impl<R: Debug, P: Debug, I: Debug, O: Debug> Debug for ModularPredictor<R, P, I, O>

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

Formats the value using the given formatter.

impl<R, P, I, O> StandardPredictor for ModularPredictor<R, P, I, O>

where R: ImageReader, P: Preprocessor, I: InferenceEngine<Input = P::Output>, O: Postprocessor<InferenceOutput = I::Output, PreprocessOutput = P::Output, Config = P::Config>,

type Config = <P as Preprocessor>::Config

The configuration type for this predictor.

type Result = <O as Postprocessor>::Result

The result type of this predictor.

type PreprocessOutput = <P as Preprocessor>::Output

The output type of the preprocessing step.

type InferenceOutput = <I as InferenceEngine>::Output

The output type of the inference step.

fn read_images<'a>( &self, paths: impl Iterator<Item = &'a str>, ) -> Result<Vec<RgbImage>, OCRError>

Reads images from the given paths.

fn preprocess( &self, images: Vec<RgbImage>, config: Option<&Self::Config>, ) -> Result<Self::PreprocessOutput, OCRError>

Preprocesses the given images.

fn infer( &self, input: &Self::PreprocessOutput, ) -> Result<Self::InferenceOutput, OCRError>

Performs inference on the preprocessed input.

fn postprocess( &self, output: Self::InferenceOutput, preprocessed: &Self::PreprocessOutput, batch_data: &BatchData, raw_images: Vec<RgbImage>, config: Option<&Self::Config>, ) -> OcrResult<Self::Result>

Postprocesses the inference output.

fn empty_result(&self) -> OcrResult<Self::Result>

Returns an empty result for the predictor type.

fn predict( &self, images: Vec<RgbImage>, config: Option<Self::Config>, ) -> OcrResult<Self::Result>

Performs prediction directly from in-memory images.

fn create_dummy_batch_data(&self, count: usize) -> BatchData

Creates dummy batch data for in-memory processing.