calamari_ocr.ocr.predict

class calamari_ocr.ocr.predict.params.PredictionCharacter(char: str = '', label: int = 0, probability: float = 0)

Bases: object

char: str = ''

label: int = 0

probability: float = 0

__init__(char: str = '', label: int = 0, probability: float = 0) → None

classmethod from_dict(kvs: Optional[Union[dict, list, str, int, float, bool]], *, infer_missing=False) → A

classmethod from_json(s: Union[str, bytes, bytearray], *, parse_float=None, parse_int=None, parse_constant=None, infer_missing=False, **kw) → A

classmethod schema(*, infer_missing: bool = False, only=None, exclude=(), many: bool = False, context=None, load_only=(), dump_only=(), partial: bool = False, unknown=None) → SchemaF[A]

to_dict(encode_json=False) → Dict[str, Optional[Union[dict, list, str, int, float, bool]]]

to_json(*, skipkeys: bool = False, ensure_ascii: bool = True, check_circular: bool = True, allow_nan: bool = True, indent: Optional[Union[int, str]] = None, separators: Optional[Tuple[str, str]] = None, default: Optional[Callable] = None, sort_keys: bool = False, **kw) → str

class calamari_ocr.ocr.predict.params.PredictionPosition(chars: List[calamari_ocr.ocr.predict.params.PredictionCharacter] = <factory>, local_start: int = 0, local_end: int = 0, global_start: int = 0, global_end: int = 0)

Bases: object

chars: List[PredictionCharacter]

local_start: int = 0

local_end: int = 0

global_start: int = 0

global_end: int = 0

__init__(chars: ~typing.List[~calamari_ocr.ocr.predict.params.PredictionCharacter] = <factory>, local_start: int = 0, local_end: int = 0, global_start: int = 0, global_end: int = 0) → None

classmethod from_dict(kvs: Optional[Union[dict, list, str, int, float, bool]], *, infer_missing=False) → A

classmethod from_json(s: Union[str, bytes, bytearray], *, parse_float=None, parse_int=None, parse_constant=None, infer_missing=False, **kw) → A

classmethod schema(*, infer_missing: bool = False, only=None, exclude=(), many: bool = False, context=None, load_only=(), dump_only=(), partial: bool = False, unknown=None) → SchemaF[A]

to_dict(encode_json=False) → Dict[str, Optional[Union[dict, list, str, int, float, bool]]]

to_json(*, skipkeys: bool = False, ensure_ascii: bool = True, check_circular: bool = True, allow_nan: bool = True, indent: Optional[Union[int, str]] = None, separators: Optional[Tuple[str, str]] = None, default: Optional[Callable] = None, sort_keys: bool = False, **kw) → str

class calamari_ocr.ocr.predict.params.Prediction(id: str = '', sentence: str = '', labels: List[int] = <factory>, positions: List[calamari_ocr.ocr.predict.params.PredictionPosition] = <factory>, logits: Union[<built-in function array>, NoneType] = None, total_probability: float = 0, avg_char_probability: float = 0, is_voted_result: bool = False, line_path: str = '', voter_predictions: Union[List[ForwardRef('Prediction')], NoneType] = None)

Bases: object

id: str = ''

sentence: str = ''

labels: List[int]

positions: List[PredictionPosition]

logits: Optional[array] = None

total_probability: float = 0

avg_char_probability: float = 0

is_voted_result: bool = False

line_path: str = ''

voter_predictions: Optional[List[Prediction]] = None

__init__(id: str = '', sentence: str = '', labels: ~typing.List[int] = <factory>, positions: ~typing.List[~calamari_ocr.ocr.predict.params.PredictionPosition] = <factory>, logits: ~typing.Optional[~numpy.array] = None, total_probability: float = 0, avg_char_probability: float = 0, is_voted_result: bool = False, line_path: str = '', voter_predictions: ~typing.Optional[~typing.List[~calamari_ocr.ocr.predict.params.Prediction]] = None) → None

classmethod from_dict(kvs: Optional[Union[dict, list, str, int, float, bool]], *, infer_missing=False) → A

classmethod from_json(s: Union[str, bytes, bytearray], *, parse_float=None, parse_int=None, parse_constant=None, infer_missing=False, **kw) → A

classmethod schema(*, infer_missing: bool = False, only=None, exclude=(), many: bool = False, context=None, load_only=(), dump_only=(), partial: bool = False, unknown=None) → SchemaF[A]

to_dict(encode_json=False) → Dict[str, Optional[Union[dict, list, str, int, float, bool]]]

to_json(*, skipkeys: bool = False, ensure_ascii: bool = True, check_circular: bool = True, allow_nan: bool = True, indent: Optional[Union[int, str]] = None, separators: Optional[Tuple[str, str]] = None, default: Optional[Callable] = None, sort_keys: bool = False, **kw) → str

class calamari_ocr.ocr.predict.params.Predictions(predictions: List[calamari_ocr.ocr.predict.params.Prediction] = <factory>, line_path: str = '')

Bases: object

predictions: List[Prediction]

line_path: str = ''

__init__(predictions: ~typing.List[~calamari_ocr.ocr.predict.params.Prediction] = <factory>, line_path: str = '') → None

classmethod from_dict(kvs: Optional[Union[dict, list, str, int, float, bool]], *, infer_missing=False) → A

classmethod from_json(s: Union[str, bytes, bytearray], *, parse_float=None, parse_int=None, parse_constant=None, infer_missing=False, **kw) → A

classmethod schema(*, infer_missing: bool = False, only=None, exclude=(), many: bool = False, context=None, load_only=(), dump_only=(), partial: bool = False, unknown=None) → SchemaF[A]

to_dict(encode_json=False) → Dict[str, Optional[Union[dict, list, str, int, float, bool]]]

to_json(*, skipkeys: bool = False, ensure_ascii: bool = True, check_circular: bool = True, allow_nan: bool = True, indent: Optional[Union[int, str]] = None, separators: Optional[Tuple[str, str]] = None, default: Optional[Callable] = None, sort_keys: bool = False, **kw) → str

class calamari_ocr.ocr.predict.params.PredictorParams(device: tfaip.device.device_config.DeviceConfigParams = <factory>, pipeline: tfaip.data.databaseparams.DataPipelineParams = <factory>, silent: bool = True, progress_bar: bool = True, run_eagerly: bool = False, include_targets: bool = False, include_meta: bool = False)

Bases: PredictorParams

silent: bool = True

__init__(device: ~tfaip.device.device_config.DeviceConfigParams = <factory>, pipeline: ~tfaip.data.databaseparams.DataPipelineParams = <factory>, silent: bool = True, progress_bar: bool = True, run_eagerly: bool = False, include_targets: bool = False, include_meta: bool = False) → None

classmethod from_dict(kvs: Optional[Union[dict, list, str, int, float, bool]], *, infer_missing=False) → A

classmethod from_json(s: Union[str, bytes, bytearray], *, parse_float=None, parse_int=None, parse_constant=None, infer_missing=False, **kw) → A

classmethod schema(*, infer_missing: bool = False, only=None, exclude=(), many: bool = False, context=None, load_only=(), dump_only=(), partial: bool = False, unknown=None) → SchemaF[A]

to_dict(encode_json=False) → Dict[str, Optional[Union[dict, list, str, int, float, bool]]]

to_json(*, skipkeys: bool = False, ensure_ascii: bool = True, check_circular: bool = True, allow_nan: bool = True, indent: Optional[Union[int, str]] = None, separators: Optional[Tuple[str, str]] = None, default: Optional[Callable] = None, sort_keys: bool = False, **kw) → str

class calamari_ocr.ocr.predict.params.PredictionResult(prediction, codec, text_postproc, out_to_in_trans: Callable[[int], int], ground_truth=None)

Bases: object

__init__(prediction, codec, text_postproc, out_to_in_trans: Callable[[int], int], ground_truth=None)

The output of a networks prediction (PredictionProto) with additional information

It stores all required information for decoding (codec) and interpreting the output.

Parameters:

prediction (PredictionProto) – prediction the DNN
codec (Codec) – codec required to decode the prediction
text_postproc (TextPostprocessor) – text processor to apply to the decodec prediction to receive the actual prediction sentence

class calamari_ocr.ocr.predict.predictor.Predictor(params: PredictorParams, data: DataBase, **kwargs)

Bases: Predictor

static from_checkpoint(params: PredictorParams, checkpoint: str, auto_update_checkpoints=True)

class calamari_ocr.ocr.predict.predictor.MultiPredictor(voter_params, *args, **kwargs)

Bases: MultiModelPredictor

classmethod from_paths(checkpoints: List[str], auto_update_checkpoints=True, predictor_params: Optional[PredictorParams] = None, voter_params: Optional[VoterParams] = None, **kwargs) → MultiModelPredictor

Create a MultiModePredictor. The data of the first model (in paths) will be used as the defining scenario and data (i.e. the post-processing). All data pre-procs must be identical.

Parameters:

paths – paths to the scenario_params (see ScenarioBase.params_from_path)
params – PredictorParams
scenario – Type of the ScenarioBase
use_first_params – Only False is supported ATM.
model_paths – Paths to the actual models saved dirs (optional), by default based on paths
models – Already instantiated models (optional), by default created based on paths
predictor_args – Additional args for instantiating the Predictor (that are not part of PredictorParams)

Returns:

An instantiated and ready to use MultiModelPredictor

__init__(voter_params, *args, **kwargs)

create_voter(data_params: DataParams) → MultiModelVoter