Differences with torchtext.data.functional.sentencepiece_tokenizer

torchtext.data.functional.sentencepiece_tokenizer

torchtext.data.functional.sentencepiece_tokenizer(
    sp_model
)

For more information, see torchtext.data.functional.sentencepiece_tokenizer.

mindspore.dataset.text.SentencePieceTokenizer

class mindspore.dataset.text.SentencePieceTokenizer(
    mode,
    out_type
)

For more information, see mindspore.dataset.text.SentencePieceTokenizer.

Differences

PyTorch: Returns a generator that converts text into string based on the input sentencepiece tokenizer model.

MindSpore: According to the incoming sentencepiece model, the input text is segmented and marked; the output type is string or int type.

Categories	Subcategories	PyTorch	MindSpore	Difference
Parameter	Parameter1	sp_model	mode	MindSpore support SentencePieceVocab object or path of SentencePiece model
	Parameter2	-	out_type	The output type of tokenizer

Code Example

from download import download

url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/sentencepiece.bpe.model"
download(url, './sentencepiece.bpe.model', replace=True)

# PyTorch
from torchtext.data.functional import load_sp_model, sentencepiece_tokenizer

list_a = "sentencepiece encode as pieces"
model = load_sp_model("./sentencepiece.bpe.model")
sp_id_generator = sentencepiece_tokenizer(model)
print(list(sp_id_generator([list_a])))
# Out: [['▁sentence', 'piece', '▁en', 'code', '▁as', '▁pieces']]

# MindSpore
import mindspore.dataset.text as text

sp_id_generator = text.SentencePieceTokenizer("./sentencepiece.bpe.model", out_type=text.SPieceTokenizerOutType.STRING)
print(list(sp_id_generator(list_a)))
# Out: ['▁sentence', 'piece', '▁en', 'code', '▁as', '▁pieces']