machine-translation-nar-ru-en-0001

Use Case and High-Level Description

This is a Russian-English machine translation model based on non-autoregressive Transformer topology.

Tokenization occurs using the SentencePieceBPETokenizer (see the demo code for implementation details) and the enclosed tokenizer_src and tokenizer_tgt folders.

Specification

Metric Value
GOps 23.17
MParams 69.29
Source framework PyTorch*

Accuracy

The quality metrics were calculated on the wmt19-ru-en dataset ("test" split in lower case).

Metric Value
BLEU 22.8 %

Performance

Input

name: tokens shape: [1, 192] description: sequence of tokens (integer values) representing the tokenized sentence. The sequence structure is as follows (<s>, </s> and <pad> should be replaced by corresponding token IDs as specified by the dictionary): <s> + tokenized sentence + </s> + (<pad> tokens to pad to the maximum sequence length of 192)

Output

name: pred shape: [1, 192] description: sequence of tokens (integer values) representing the tokenized translation. The sequence structure is as follows (<s>, </s> and <pad> should be replaced by corresponding token IDs as specified by the dictionary): <s> + tokenized sentence + </s> + (<pad> tokens to pad to the maximum sequence length of 192)

Legal Information

[*] Other names and brands may be claimed as the property of others.