Struct ToBatch 

pub struct ToBatch;

A struct for converting image data into batched tensor format.

This struct provides methods for validating input data and converting images into a batched tensor format suitable for processing in the OCR pipeline.

Implementations

impl ToBatch

pub fn new() -> Self

Creates a new ToBatch instance.

Returns

A new ToBatch instance.

pub fn validate_inputs( &self, imgs: &[Vec<f32>], shapes: &[(usize, usize, usize)], ) -> Result<(), OCRError>

Validates the input images and their shapes.

This method checks that the images and shapes arrays have the same length, that all images have the correct number of elements for their shapes, and that all dimensions are greater than zero.

Arguments

• imgs - A slice of vectors of f32 values representing the images.
• shapes - A slice of tuples representing the shapes of the images (channels, height, width).

Returns

A Result indicating success or an OCRError if validation fails.

pub fn apply( &self, imgs: &[Vec<f32>], shapes: &[(usize, usize, usize)], ) -> Result<Vec<f32>, OCRError>

Applies the batch conversion to the input images and shapes.

This method validates the inputs, then converts the images into a batched tensor format. If all images have the same dimensions, it uses a more efficient contiguous copying method. Otherwise, it uses a method that handles mixed dimensions.

Arguments

• imgs - A slice of vectors of f32 values representing the images.
• shapes - A slice of tuples representing the shapes of the images (channels, height, width).

Returns

A Result containing a vector of f32 values representing the batched tensor, or an OCRError if the operation fails.

Trait Implementations

impl Debug for ToBatch

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

Formats the value using the given formatter.

impl Default for ToBatch

fn default() -> ToBatch

Returns the “default value” for a type.