training/flax/run_pseudo_labelling_pt.py [40:130]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    DatasetDict,
    IterableDatasetDict,
    load_dataset,
)
from huggingface_hub import HfFolder, Repository, create_repo, get_full_repo_name
from torch.utils.data import DataLoader
from tqdm import tqdm
from transformers import (
    HfArgumentParser,
    Seq2SeqTrainingArguments,
    WhisperConfig,
    WhisperFeatureExtractor,
    WhisperForConditionalGeneration,
    WhisperProcessor,
    WhisperTokenizerFast,
)
from transformers.models.whisper.english_normalizer import EnglishTextNormalizer
from transformers.utils import check_min_version
from transformers.utils.versions import require_version


# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.34.0.dev0")

require_version("datasets>=2.14.6", "To fix: `pip install --upgrade datasets`")

logger = get_logger(__name__)


@dataclass
class ModelArguments:
    """
    Arguments pertaining to which model/config/tokenizer we are going to distill from.
    """

    model_name_or_path: str = field(
        metadata={"help": "Path to pretrained Whisper model or model identifier from huggingface.co/models"}
    )
    config_name: Optional[str] = field(
        default=None,
        metadata={"help": "Pretrained config name or path if not the same as model_name"},
    )
    tokenizer_name: Optional[str] = field(
        default=None,
        metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"},
    )
    feature_extractor_name: Optional[str] = field(
        default=None,
        metadata={"help": "feature extractor name or path if not the same as model_name"},
    )
    processor_name: Optional[str] = field(
        default=None,
        metadata={"help": "processor name or path if not the same as model_name"},
    )
    cache_dir: Optional[str] = field(
        default=None,
        metadata={"help": "Where to store the pretrained models downloaded from huggingface.co"},
    )
    use_fast_tokenizer: bool = field(
        default=True,
        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
    )
    model_revision: str = field(
        default="main",
        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
    )
    subfolder: str = field(
        default="",
        metadata={
            "help": "In case the relevant files are located inside a subfolder of the model repo on huggingface.co, you can"
            "specify the folder name here."
        },
    )
    token: str = field(
        default=None,
        metadata={
            "help": (
                "The token to use as HTTP bearer authorization for remote files. If not specified, will use the token "
                "generated when running `huggingface-cli login` (stored in `~/.huggingface`)."
            )
        },
    )
    dtype: Optional[str] = field(
        default="float32",
        metadata={
            "help": (
                "The data type (dtype) in which to load the model weights. One of `float32` (full-precision), "
                "`float16` or `bfloat16` (both half-precision)."
            )
        },
    )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



training/run_pseudo_labelling.py [40:132]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    DatasetDict,
    IterableDatasetDict,
    load_dataset,
)
from huggingface_hub import HfFolder, create_repo, get_full_repo_name, snapshot_download, upload_folder
from torch.utils.data import DataLoader
from tqdm import tqdm
from soundfile import LibsndfileError
from datasets.arrow_dataset import table_iter
from transformers import (
    HfArgumentParser,
    Seq2SeqTrainingArguments,
    WhisperConfig,
    WhisperFeatureExtractor,
    WhisperForConditionalGeneration,
    WhisperProcessor,
    WhisperTokenizerFast,
)
from transformers.models.whisper.english_normalizer import BasicTextNormalizer, EnglishTextNormalizer
from transformers.utils import check_min_version
from transformers.utils.versions import require_version


# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.34.0.dev0")

require_version("datasets>=2.14.6", "To fix: `pip install --upgrade datasets`")

logger = get_logger(__name__)


@dataclass
class ModelArguments:
    """
    Arguments pertaining to which model/config/tokenizer we are going to distill from.
    """

    model_name_or_path: str = field(
        metadata={"help": "Path to pretrained Whisper model or model identifier from huggingface.co/models"}
    )
    config_name: Optional[str] = field(
        default=None,
        metadata={"help": "Pretrained config name or path if not the same as model_name"},
    )
    tokenizer_name: Optional[str] = field(
        default=None,
        metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"},
    )
    feature_extractor_name: Optional[str] = field(
        default=None,
        metadata={"help": "feature extractor name or path if not the same as model_name"},
    )
    processor_name: Optional[str] = field(
        default=None,
        metadata={"help": "processor name or path if not the same as model_name"},
    )
    cache_dir: Optional[str] = field(
        default=None,
        metadata={"help": "Where to store the pretrained models downloaded from huggingface.co"},
    )
    use_fast_tokenizer: bool = field(
        default=True,
        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
    )
    model_revision: str = field(
        default="main",
        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
    )
    subfolder: str = field(
        default="",
        metadata={
            "help": "In case the relevant files are located inside a subfolder of the model repo on huggingface.co, you can"
            "specify the folder name here."
        },
    )
    token: str = field(
        default=None,
        metadata={
            "help": (
                "The token to use as HTTP bearer authorization for remote files. If not specified, will use the token "
                "generated when running `huggingface-cli login` (stored in `~/.huggingface`)."
            )
        },
    )
    dtype: Optional[str] = field(
        default="float32",
        metadata={
            "help": (
                "The data type (dtype) in which to load the model weights. One of `float32` (full-precision), "
                "`float16` or `bfloat16` (both half-precision)."
            )
        },
    )
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -