Source code for torchtext.datasets.sst2

# Copyright (c) Facebook, Inc. and its affiliates.
import os

from torchtext._internal.module_utils import is_module_available
from import (

if is_module_available("torchdata"):
    from torchdata.datapipes.iter import IterableWrapper, FileOpener

    # we import HttpReader from _download_hooks so we can swap out public URLs
    # with interal URLs when the dataset is used within Facebook
    from torchtext._download_hooks import HttpReader

URL = ""

MD5 = "9f81648d4199384278b86e315dac217c"

    "train": 67349,
    "dev": 872,
    "test": 1821,

_PATH = ""


    "train": os.path.join("SST-2", "train.tsv"),
    "dev": os.path.join("SST-2", "dev.tsv"),
    "test": os.path.join("SST-2", "test.tsv"),

[docs]@_create_dataset_directory(dataset_name=DATASET_NAME) @_wrap_split_argument(("train", "dev", "test")) def SST2(root, split): """SST2 Dataset For additional details refer to Number of lines per split: - train: 67349 - dev: 872 - test: 1821 Args: root: Directory where the datasets are saved. Default: os.path.expanduser('~/.torchtext/cache') split: split or splits to be returned. Can be a string or tuple of strings. Default: (`train`, `dev`, `test`) :returns: DataPipe that yields tuple of text and/or label (1 to 4). The `test` split only returns text. :rtype: Union[(int, str), (str,)] """ # TODO Remove this after removing conditional dependency if not is_module_available("torchdata"): raise ModuleNotFoundError( "Package `torchdata` not found. Please install following instructions at ``" ) url_dp = IterableWrapper([URL]) cache_compressed_dp = url_dp.on_disk_cache( filepath_fn=lambda x: os.path.join(root, os.path.basename(x)), hash_dict={os.path.join(root, os.path.basename(URL)): MD5}, hash_type="md5", ) cache_compressed_dp = HttpReader(cache_compressed_dp).end_caching( mode="wb", same_filepath_fn=True ) cache_decompressed_dp = cache_compressed_dp.on_disk_cache( filepath_fn=lambda x: os.path.join(root, _EXTRACTED_FILES[split]) ) cache_decompressed_dp = ( FileOpener(cache_decompressed_dp, mode="b") .read_from_zip() .filter(lambda x: _EXTRACTED_FILES[split] in x[0]) ) cache_decompressed_dp = cache_decompressed_dp.end_caching( mode="wb", same_filepath_fn=True ) data_dp = FileOpener(cache_decompressed_dp, mode="b") # test split for SST2 doesn't have labels if split == "test": parsed_data = data_dp.parse_csv(skip_lines=1, delimiter="\t").map( lambda t: (t[1].strip(),) ) else: parsed_data = data_dp.parse_csv(skip_lines=1, delimiter="\t").map( lambda t: (t[0].strip(), int(t[1])) ) return parsed_data


Access comprehensive developer documentation for PyTorch

View Docs


Get in-depth tutorials for beginners and advanced developers

View Tutorials


Find development resources and get your questions answered

View Resources