turi::neural_net::image_augmenter::pad_options Struct Reference

#include <ml/neural_net/image_augmentation.hpp>

Public Attributes
float	min_aspect_ratio = 0.8f
float	max_aspect_ratio = 1.25f
float	min_area_fraction = 1.f
float	max_area_fraction = 2.f
size_t	max_attempts = 50

Detailed Description

Parameters governing random padding.

Definition at line 132 of file image_augmentation.hpp.

Member Data Documentation

max_area_fraction

float turi::neural_net::image_augmenter::pad_options::max_area_fraction = 2.f

Given a sampled aspect ratio, determines the upper bound of the uniformly sampled height.

Definition at line 150 of file image_augmentation.hpp.

max_aspect_ratio

float turi::neural_net::image_augmenter::pad_options::max_aspect_ratio = 1.25f

Upper bound for the uniformly sampled aspect ratio (width/height)

Definition at line 138 of file image_augmentation.hpp.

max_attempts

size_t turi::neural_net::image_augmenter::pad_options::max_attempts = 50

The maximum number of random aspect ratios to sample, looking for one that satisfies the constraints on area.

Definition at line 156 of file image_augmentation.hpp.

min_area_fraction

float turi::neural_net::image_augmenter::pad_options::min_area_fraction = 1.f

Given a sampled aspect ratio, determines the lower bound of the uniformly sampled height.

Definition at line 144 of file image_augmentation.hpp.

min_aspect_ratio

float turi::neural_net::image_augmenter::pad_options::min_aspect_ratio = 0.8f

Lower bound for the uniformly sampled aspect ratio (width/height)

Definition at line 135 of file image_augmentation.hpp.

---

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

• ml/neural_net/image_augmentation.hpp