Package org.anchoranalysis.image.inference

Class ImageInferenceContext

Object

ImageInferenceContext

public final class ImageInferenceContext
extends Object

Attributes describing the situation in which inference on images is occurring.

These may be helpful for decoding output from the inference.

Author:

Owen Feehan

Constructor Summary

Constructors

Constructor	Description
ImageInferenceContext(DualScale<Dimensions> dimensions, ScaleFactor scaleFactor, Optional<List<String>> classLabels, VoxelsResizer resizer, ExecutionTimeRecorder executionTimeRecorder, Logger logger)	Creates a new ImageInferenceContext instance.

Method Summary

Modifier and Type	Method	Description
boolean	equals(Object o)
Optional<List<String>>	getClassLabels()	If available, labels of classes loaded from a text file at classLabelsPath.
DualScale<Dimensions>	getDimensions()	The size of the image for which we wish to segment, before and after any scaling for model inference.
ExecutionTimeRecorder	getExecutionTimeRecorder()	Allows execution-time for particular operations to be recorded.
Logger	getLogger()	Where to log information messages during inference.
VoxelsResizer	getResizer()	How to resize images or voxel-buffers.
ScaleFactor	getScaleFactor()	The inverse of the scaling-factor applied to reduce unscaledDimensions to the input-matrix used for inference.
int	hashCode()
final DualScale<Optional<ScaleFactor>>	scaleFactorUpscale()	The scaling-factors needed to upscale the model output to match the desired scale.
String	toString()

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Details

ImageInferenceContext

public ImageInferenceContext(DualScale<Dimensions> dimensions,
 ScaleFactor scaleFactor,
 Optional<List<String>> classLabels,
 VoxelsResizer resizer,
 ExecutionTimeRecorder executionTimeRecorder,
 Logger logger)

Creates a new ImageInferenceContext instance.

Parameters:

dimensions - The size of the image for which we wish to segment, before and after any scaling for model inference.

scaleFactor - The inverse of the scaling-factor applied to reduce unscaledDimensions to the input-matrix used for inference.

classLabels - If available, labels of classes loaded from a text file at classLabelsPath.

resizer - How to resize images or voxel-buffers.

executionTimeRecorder - Allows execution-time for particular operations to be recorded.

logger - Where to log information messages during inference.

Method Details

scaleFactorUpscale

public final DualScale<Optional<ScaleFactor>> scaleFactorUpscale()

The scaling-factors needed to upscale the model output to match the desired scale.

Returns:

each scaling-factor (if required) to scale an image (at the respective scale in DualScale by, so they become identically sized to the DualScale.atInputScale().

getDimensions

public DualScale<Dimensions> getDimensions()

The size of the image for which we wish to segment, before and after any scaling for model inference.

getScaleFactor

public ScaleFactor getScaleFactor()

The inverse of the scaling-factor applied to reduce unscaledDimensions to the input-matrix used for inference.

getClassLabels

public Optional<List<String>> getClassLabels()

If available, labels of classes loaded from a text file at classLabelsPath.

getResizer

public VoxelsResizer getResizer()

How to resize images or voxel-buffers.

getExecutionTimeRecorder

public ExecutionTimeRecorder getExecutionTimeRecorder()

Allows execution-time for particular operations to be recorded.

getLogger

public Logger getLogger()

Where to log information messages during inference.

equals

public boolean equals(Object o)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object

toString

public String toString()

Overrides:

toString in class Object