pulsatrix::ResizeTransform Class Reference

Every transform in this file operates on sample.fields[0], assumed to be an image Tensor of shape (1, channels, height, width) – the convention MnistDatasetAdapter, CsvDataset, and ImageDecoder all already share (image/features first, label last). More...

#include <image_transforms.hpp>

Inheritance diagram for pulsatrix::ResizeTransform:

Collaboration diagram for pulsatrix::ResizeTransform:

Public Member Functions
	ResizeTransform (int64_t target_height, int64_t target_width, DeviceBackend *backend)
Sample	apply (Sample sample) const override
	Applies this transform to a sample.
Public Member Functions inherited from pulsatrix::Transform
virtual	~Transform ()=default

Detailed Description

Every transform in this file operates on sample.fields[0], assumed to be an image Tensor of shape (1, channels, height, width) – the convention MnistDatasetAdapter, CsvDataset, and ImageDecoder all already share (image/features first, label last).

Nearest-neighbor resize to (target_height, target_width).

Note

Needs a DeviceBackend to allocate the differently-shaped output Tensor – unlike NormalizeTransform/HorizontalFlipTransform, which modify a same-shaped buffer in place and need no backend at all.

Constructor & Destructor Documentation

ResizeTransform()

pulsatrix::ResizeTransform::ResizeTransform ( int64_t  target_height, int64_t  target_width, DeviceBackend *  backend  )

inline

Member Function Documentation

apply()

Sample pulsatrix::ResizeTransform::apply ( Sample  sample ) const

overridevirtual

Applies this transform to a sample.

Parameters

sample

Input sample.

Returns

The transformed sample.

Implements pulsatrix::Transform.

---

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

• include/pulsatrix/image_transforms.hpp