Differences with torchtext.data.functional.load_sp_model

View Source On Gitee

torchtext.data.functional.load_sp_model

torchtext.data.functional.load_sp_model(
    spm
)

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

mindspore.dataset.text.SentencePieceTokenizer

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

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

Differences

PyTorch: Load a sentencepiece model.

MindSpore: Construct a SentencePiece tokenizer, including load a sentencepiece model.

Categories

Subcategories

PyTorch

MindSpore

Difference

Parameter

Parameter1

spm

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
model = load_sp_model("sentencepiece.bpe.model")

# MindSpore
import mindspore.dataset.text as text
model = text.SentencePieceTokenizer("sentencepiece.bpe.model", out_type=text.SPieceTokenizerOutType.STRING)