Multi-Class Label Processor

Processor for multi-class classification tasks. Converts string or integer labels into integer indices, with one output per label.

Super classes

RHealth::Processor -> RHealth::FeatureProcessor -> MultiClassLabelProcessor

Public fields

label_vocab

A named integer vector mapping labels to indices.

Methods

Public methods

• MultiClassLabelProcessor$new()

• MultiClassLabelProcessor$fit()

• MultiClassLabelProcessor$process()

• MultiClassLabelProcessor$size()

• MultiClassLabelProcessor$print()

• MultiClassLabelProcessor$clone()

Inherited methods

• RHealth::Processor$load()
• RHealth::Processor$save()

---

Method new()

Initialize with empty label vocabulary.

Usage

MultiClassLabelProcessor$new()

---

Method fit()

Fit the label vocabulary from the given field of all samples.

Usage

MultiClassLabelProcessor$fit(samples, field)

Arguments

samples

A list of named lists representing the dataset.

field

The name of the field containing labels.

---

Method process()

Convert a label into a torch long integer tensor (scalar).

Usage

MultiClassLabelProcessor$process(value)

Arguments

value

The raw label value.

Returns

An int64 torch tensor.

---

Method size()

Return number of classes (vocabulary size).

Usage

MultiClassLabelProcessor$size()

Returns

Integer

---

Method print()

Print a summary of the processor.

Print a summary of the processor.

Usage

MultiClassLabelProcessor$print(...)

Arguments

...

Ignored.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

MultiClassLabelProcessor$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.