turi::neural_net::image_augmenter Class Reference

#include <ml/neural_net/image_augmentation.hpp>

Classes
struct	crop_options
struct	options
struct	pad_options
struct	result

Public Member Functions
virtual const options &	get_options () const =0
virtual result	prepare_images (std::vector< labeled_image > source_batch)=0

Detailed Description

Pure virtual interface for objects that process/augment/mutate images and their associated annotations.

Definition at line 86 of file image_augmentation.hpp.

Member Function Documentation

get_options()

virtual const options& turi::neural_net::image_augmenter::get_options ( ) const

pure virtual

Returns the options parameterizing this instance.

Implemented in turi::neural_net::float_array_image_augmenter.

prepare_images()

virtual result turi::neural_net::image_augmenter::prepare_images ( std::vector< labeled_image >  source_batch )

pure virtual

Performs augmentation on a batch of images (and their annotations).

If the source batch is smaller than the batch size specified in the options, then the result is padded with zeroes as needed.

Implemented in turi::neural_net::float_array_image_augmenter.

---

The documentation for this class was generated from the following file:

• ml/neural_net/image_augmentation.hpp