管道选项

基类: str, Enum

运行模型推断的设备

属性：

• AUTO –
• CPU –
• CUDA –
• MPS –

AUTO = 'auto'

CPU = 'cpu'

CUDA = 'cuda'

MPS = 'mps'

基类: BaseSettings

方法：

• check_alternative_envvars –

  从 "alternative" 环境变量 OMP_NUM_THREADS 设置 num_threads。

• validate_device –

属性：

• device (Union[str, AcceleratorDevice]) –
• model_config –
• num_threads (int) –

device: Union[str, AcceleratorDevice] = 'auto'

model_config = SettingsConfigDict(
    env_prefix="DOCLING_",
    env_nested_delimiter="_",
    populate_by_name=True,
)

num_threads: int = 4

check_alternative_envvars(data: Any) -> Any

validate_device(value)

基类: OcrOptions

EasyOCR引擎的选项。

属性：

• bitmap_area_threshold (float) –
• confidence_threshold (float) –
• download_enabled (bool) –
• force_full_page_ocr (bool) –
• kind (Literal['easyocr']) –
• lang (List[str]) –
• model_config –
• model_storage_directory (可选[str]) –
• recog_network (可选[str]) –
• use_gpu (可选[bool]) –

bitmap_area_threshold: float = 0.05

confidence_threshold: float = 0.5

download_enabled: bool = True

force_full_page_ocr: bool = False

kind: Literal['easyocr'] = 'easyocr'

lang: List[str] = ['fr', 'de', 'es', 'en']

model_config = ConfigDict(
    extra="forbid", protected_namespaces=()
)

model_storage_directory: Optional[str] = None

recog_network: Optional[str] = 'standard'

use_gpu: Optional[bool] = None

基类: str, Enum

有效OCR引擎的枚举。

属性：

• EASYOCR –
• OCRMAC –
• RAPIDOCR –
• TESSERACT –
• TESSERACT_CLI –

EASYOCR = 'easyocr'

OCRMAC = 'ocrmac'

RAPIDOCR = 'rapidocr'

TESSERACT = 'tesseract'

TESSERACT_CLI = 'tesseract_cli'

基类: OcrOptions

Mac OCR 引擎的选项。

属性：

• bitmap_area_threshold (float) –
• force_full_page_ocr (bool) –
• framework (str) –
• kind (Literal['ocrmac']) –
• lang (List[str]) –
• model_config –
• 识别 (str) –

bitmap_area_threshold: float = 0.05

force_full_page_ocr: bool = False

framework: str = 'vision'

kind: Literal['ocrmac'] = 'ocrmac'

lang: List[str] = ['fr-FR', 'de-DE', 'es-ES', 'en-US']

model_config = ConfigDict(extra='forbid')

recognition: str = 'accurate'

基础: BaseModel

OCR 选项。

属性：

• bitmap_area_threshold (float) –
• force_full_page_ocr (bool) –
• kind (str) –
• lang (List[str]) –

bitmap_area_threshold: float = 0.05

force_full_page_ocr: bool = False

kind: str

lang: List[str]

基类: str, Enum

有效PDF后端的枚举。

属性：

• DLPARSE_V1 –
• DLPARSE_V2 –
• PYPDFIUM2 –

DLPARSE_V1 = 'dlparse_v1'

DLPARSE_V2 = 'dlparse_v2'

PYPDFIUM2 = 'pypdfium2'

基类: PipelineOptions

PDF管道的选项。

属性：

• accelerator_options (AcceleratorOptions) –
• artifacts_path (可选[联合[Path, str]]) –
• create_legacy_output (bool) –
• do_code_enrichment (bool) –
• do_formula_enrichment (bool) –
• do_ocr (bool) –
• do_picture_classification (bool) –
• do_picture_description (bool) –
• do_table_structure (bool) –
• document_timeout (可选[float]) –
• enable_remote_services (bool) –
• generate_page_images (bool) –
• generate_picture_images (bool) –
• generate_table_images (bool) –
• images_scale (float) –
• ocr_options (联合[EasyOcrOptions, TesseractCliOcrOptions, TesseractOcrOptions, OcrMacOptions, RapidOcrOptions]) –
• picture_description_options (已注释[联合[PictureDescriptionApiOptions, PictureDescriptionVlmOptions], 字段(区分符=kind)]) –
• table_structure_options (TableStructureOptions) –

accelerator_options: AcceleratorOptions = (
    AcceleratorOptions()
)

artifacts_path: Optional[Union[Path, str]] = None

create_legacy_output: bool = True

do_code_enrichment: bool = False

do_formula_enrichment: bool = False

do_ocr: bool = True

do_picture_classification: bool = False

do_picture_description: bool = False

do_table_structure: bool = True

document_timeout: Optional[float] = None

enable_remote_services: bool = False

generate_page_images: bool = False

generate_picture_images: bool = False

generate_table_images: bool = Field(
    default=False,
    deprecated="Field `generate_table_images` is deprecated. To obtain table images, set `PdfPipelineOptions.generate_page_images = True` before conversion and then use the `TableItem.get_image` function.",
)

images_scale: float = 1.0

ocr_options: Union[
    EasyOcrOptions,
    TesseractCliOcrOptions,
    TesseractOcrOptions,
    OcrMacOptions,
    RapidOcrOptions,
] = Field(EasyOcrOptions(), discriminator="kind")

picture_description_options: Annotated[
    Union[
        PictureDescriptionApiOptions,
        PictureDescriptionVlmOptions,
    ],
    Field(discriminator=kind),
] = smolvlm_picture_description

table_structure_options: TableStructureOptions = (
    TableStructureOptions()
)

基础: PictureDescriptionBaseOptions

属性：

• batch_size (int) –
• bitmap_area_threshold (float) –
• headers (Dict[str, str]) –
• kind (Literal['api']) –
• params (Dict[str, Any]) –
• prompt (str) –
• 来源 (str) –
• scale (float) –
• timeout (float) –
• url (AnyUrl) –

batch_size: int = 8

bitmap_area_threshold: float = 0.2

headers: Dict[str, str] = {}

kind: Literal['api'] = 'api'

params: Dict[str, Any] = {}

prompt: str = 'Describe this image in a few sentences.'

provenance: str = ''

scale: float = 2

timeout: float = 20

url: AnyUrl = AnyUrl(
    "http://localhost:8000/v1/chat/completions"
)

基础: BaseModel

属性：

• batch_size (int) –
• bitmap_area_threshold (float) –
• kind (str) –
• scale (float) –

batch_size: int = 8

bitmap_area_threshold: float = 0.2

kind: str

scale: float = 2

基础: PictureDescriptionBaseOptions

属性：

• batch_size (int) –
• bitmap_area_threshold (float) –
• generation_config (Dict[str, Any]) –
• kind (Literal['vlm']) –
• prompt (str) –
• repo_cache_folder (str) –
• repo_id (str) –
• scale (float) –

batch_size: int = 8

bitmap_area_threshold: float = 0.2

generation_config: Dict[str, Any] = dict(
    max_new_tokens=200, do_sample=False
)

kind: Literal['vlm'] = 'vlm'

prompt: str = 'Describe this image in a few sentences.'

repo_cache_folder: str

repo_id: str

scale: float = 2

基础: BaseModel

基础管道选项。

属性：

• accelerator_options (AcceleratorOptions) –
• create_legacy_output (bool) –
• document_timeout (可选[float]) –
• enable_remote_services (bool) –

accelerator_options: AcceleratorOptions = (
    AcceleratorOptions()
)

create_legacy_output: bool = True

document_timeout: Optional[float] = None

enable_remote_services: bool = False

基类: OcrOptions

RapidOCR引擎的选项。

属性：

• bitmap_area_threshold (float) –
• cls_model_path (可选[str]) –
• det_model_path (可选[str]) –
• force_full_page_ocr (bool) –
• kind (Literal['rapidocr']) –
• lang (List[str]) –
• model_config –
• print_verbose (bool) –
• rec_keys_path (可选[str]) –
• rec_model_path (可选[str]) –
• text_score (float) –
• use_cls (可选[布尔]) –
• use_det (可选[布尔值]) –
• use_rec (可选[bool]) –

bitmap_area_threshold: float = 0.05

cls_model_path: Optional[str] = None

det_model_path: Optional[str] = None

force_full_page_ocr: bool = False

kind: Literal['rapidocr'] = 'rapidocr'

lang: List[str] = ['english', 'chinese']

model_config = ConfigDict(extra='forbid')

print_verbose: bool = False

rec_keys_path: Optional[str] = None

rec_model_path: Optional[str] = None

text_score: float = 0.5

use_cls: Optional[bool] = None

use_det: Optional[bool] = None

use_rec: Optional[bool] = None

基类: str, Enum

TableFormer模型的模式。

属性：

• ACCURATE –
• FAST –

ACCURATE = 'accurate'

FAST = 'fast'

基础: BaseModel

表结构的选项。

属性：

• do_cell_matching (bool) –
• mode (TableFormerMode) –

do_cell_matching: bool = True

mode: TableFormerMode = FAST

基类: OcrOptions

TesseractCli引擎的选项。

属性：

• bitmap_area_threshold (float) –
• force_full_page_ocr (bool) –
• kind (Literal['tesseract']) –
• lang (List[str]) –
• model_config –
• path (可选[str]) –
• tesseract_cmd (str) –

bitmap_area_threshold: float = 0.05

force_full_page_ocr: bool = False

kind: Literal['tesseract'] = 'tesseract'

lang: List[str] = ['fra', 'deu', 'spa', 'eng']

model_config = ConfigDict(extra='forbid')

path: Optional[str] = None

tesseract_cmd: str = 'tesseract'

基类: OcrOptions

用于Tesseract引擎的选项。

属性：

• bitmap_area_threshold (float) –
• force_full_page_ocr (bool) –
• kind (Literal['tesserocr']) –
• lang (List[str]) –
• model_config –
• path (可选[str]) –

bitmap_area_threshold: float = 0.05

force_full_page_ocr: bool = False

kind: Literal['tesserocr'] = 'tesserocr'

lang: List[str] = ['fra', 'deu', 'spa', 'eng']

model_config = ConfigDict(extra='forbid')

path: Optional[str] = None