Byte-pair encoding
Algorithm for data compression and tokenization that iteratively replaces the most frequent pairs of consecutive bytes or symbols with a new representation
Nº Q860962 ★
Comum · Saberes
Byte-pair encoding
Algorithm for data compression and tokenization that iteratively replaces the most frequent pairs of consecutive bytes or symbols with a new representation
In computing, byte-pair encoding (BPE), or digram coding, is an algorithm, first described in 1994 by Philip Gage, for encoding strings of text into smaller strings by creating and using a translation table. A slightly modified version of the algorithm is used in large language model tokenizers.
Último preço
—
Preço mínimo
—
Mediana 7 d
—
Vendas 30 d
0
Faixa 30 d
—
Em circulação
0
Cotação
mediana
mín – máx
vendas
Sem vendas no período
Ver tabela
| Data | mediana | Mín | Máx | vendas |
|---|
Histórico de vendas
- Última venda
- —
- Média 30 d
- —
- Mínima 30 d
- —
- Máxima 30 d
- —
- Vendas 7 d
- 0
- Vendas 30 d
- 0
Ainda sem vendas.
Vendas anônimas: sem comprador nem vendedor. Os números contam só vendas entre jogadores.
Na Wikipédia
Texto em inglês Ainda não há artigo no seu idioma: trecho em inglês.
In computing, byte-pair encoding (BPE), or digram coding, is an algorithm, first described in 1994 by Philip Gage, for encoding strings of text into smaller strings by creating and using a translation table. A slightly modified version of the algorithm is used in large language model tokenizers. The original version of the algorithm focused on compression. It replaces the highest-frequency pair of bytes with a new byte that was not contained in the initial dataset. A lookup table of the replacements is required to rebuild the initial dataset. The modified version builds "tokens" (units of recognition) that match varying amounts of source text, from single characters (including single digits or single punctuation marks) to whole words (even long compound words).
Texto: Wikipédia em inglês, CC BY-SA 4.0. ·
