Embeddings encodedbatch sequence

EncodedBatch.Sequence

One sequence's ids, without the padding.

public ReadOnlySpan<long> Sequence(int index)

Parameters — index is the zero-based sequence, in the order the texts were given.

Returns — ReadOnlySpan<long> of that sequence's true length — Lengths[index], not SequenceLength. A view into InputIds, copying nothing.

Example — the shorter of two sequences, three ids rather than five.

using Lodestar.Embeddings.Tokenization;

var vocab = new Dictionary<string, int>(StringComparer.Ordinal)
{
    ["[UNK]"] = 0, ["token"] = 1, ["##ize"] = 2, ["text"] = 3,
    ["[CLS]"] = 4, ["[SEP]"] = 5, ["[PAD]"] = 6,
};
var tokenizer = new WordPieceTokenizer(
    vocab, unkToken: "[UNK]", continuationPrefix: "##", maxCharsPerWord: 100, lowercase: true);
var encoder = new BatchEncoder(tokenizer, new EncodingOptions { Template = SpecialTokenTemplate.Bert });

EncodedBatch batch = encoder.EncodeBatch(["text", "tokenize text"]);

int width = batch.SequenceLength;          // => 5
int actual = batch.Sequence(0).Length;     // => 3

Remarks — it stops at the true length, which is the whole reason to use it rather than slicing InputIds yourself: the arithmetic is easy to get right and easy to get subtly wrong, and a slice that includes padding reads as real tokens.

It is a view, not a copy, so it is only valid while the batch is.

Applies to — net10.0, netstandard2.0.

See also — EncodedBatch.