HashingEmbedder class

A deterministic, offline embedder based on feature hashing of words, word bigrams and character trigrams.

It captures lexical overlap, not meaning — "car" and "automobile" are unrelated to it — so it is a baseline for tests, demos and fully offline apps, not a substitute for a neural model. Vectors are L2-normalised, so it pairs with the cosine metric.

Implemented types

• Embedder

Available extensions

• EmbedOne

Constructors

HashingEmbedder({int dimensions = 256})

Creates an embedder producing dimensions-dimensional vectors.

Properties

dimensions → int

Length of every vector this embedder returns.

final

hashCode → int

The hash code for this object.

no setter, inherited

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

close() → void

Releases network resources.

override

embed(List<String> texts, {EmbedPurpose purpose = EmbedPurpose.document}) → Future<List<Float32List>>

Embeds texts, returning one vector per text, in order.

override

embedOne(String text, {EmbedPurpose purpose = EmbedPurpose.document}) → Future<Float32List>

Available on Embedder, provided by the EmbedOne extension

Embeds a single text.

embedSync(String text) → Float32List

The vector for text, synchronously.

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited