fontr package

FontR

Subpackages

• fontr.fontr package
  • Submodules
  • fontr.fontr.autoencoder module
    • Autoencoder
      • Autoencoder.__init__()
      • Autoencoder.base_step()
      • Autoencoder.configure_optimizers()
      • Autoencoder.decode()
      • Autoencoder.encode()
      • Autoencoder.forward()
      • Autoencoder.test_step()
      • Autoencoder.training_step()
      • Autoencoder.validation_step()
  • fontr.fontr.classifier module
    • Classifier
      • Classifier.__init__()
      • Classifier.base_step()
      • Classifier.configure_optimizers()
      • Classifier.forward()
      • Classifier.predict_step()
      • Classifier.test_step()
      • Classifier.training_step()
      • Classifier.validation_step()
  • fontr.fontr.logger module
    • TorchLogger
      • TorchLogger.getLogger()
      • TorchLogger.logger
      • TorchLogger.obtainCredentials()
  • fontr.fontr.transforms module
    • AddGaussianNoise
      • AddGaussianNoise.__init__()
    • Patch
      • Patch.__init__()
    • ResizePatches
      • ResizePatches.__init__()
• fontr.pipelines package
  • Subpackages
    • fontr.pipelines.bcf_preprocessing package
      • Submodules
      • fontr.pipelines.bcf_preprocessing.nodes module
      • fontr.pipelines.bcf_preprocessing.pipeline module
    • fontr.pipelines.data_processing package
      • Submodules
      • fontr.pipelines.data_processing.nodes module
      • fontr.pipelines.data_processing.pipeline module
    • fontr.pipelines.data_science package
      • Submodules
      • fontr.pipelines.data_science.nodes module
      • fontr.pipelines.data_science.pipeline module
  • Submodules
  • fontr.pipelines.nodes module
    • set_random_state()

Submodules

fontr.datasets module

class FileWithDirAsLabel(filepath)

Bases: AbstractDataSet

__init__(filepath)

class GoogleDriveDataset(filepath, file_name)

Bases: AbstractVersionedDataSet

__init__(filepath, file_name)

Creates a new instance of AbstractVersionedDataSet.

Parameters:

• filepath (str) – Filepath in POSIX format to a file.

• version – If specified, should be an instance of kedro.io.core.Version. If its load attribute is None, the latest version will be loaded. If its save attribute is None, save version will be autogenerated.

• exists_function – Function that is used for determining whether a path exists in a filesystem.

• glob_function – Function that is used for finding all paths in a filesystem, which match a given pattern.

exists()

Checks whether a data set’s output already exists by calling the provided _exists() method.

Return type:

bool

Returns:

Flag indicating whether the output already exists.

Raises:

DataSetError – when underlying exists method raises error.

class KedroPytorchImageDataset(filepath, path_column=0, label_column=1, fs_args=None, credentials=None, transform=None, target_transform=None, return_labels=True)

Bases: Dataset, AbstractDataSet

__init__(filepath, path_column=0, label_column=1, fs_args=None, credentials=None, transform=None, target_transform=None, return_labels=True)

torch.utils.data.Dataset mixed with kedro.io.AbstractDataSet. filepath should be a CSV listing paths of images relative to the directory of filepath in the first column. The optional label_column column can contain labels for the images. The images should be located in the directory mentioned before. TODO fix this docstring.

target_transform(label)

Return type:

Tensor

transform(img)

Return type:

Tensor

with_transforms(transform=None, target_transform=None)

Return type:

KedroPytorchImageDataset

class TorchPickleModelDataset(filepath, map_location='cpu', fs_args=None, credentials=None)

Bases: AbstractDataSet

Kedro DataSet for a model to be (de-)serialized with torch.{load,save}

__init__(filepath, map_location='cpu', fs_args=None, credentials=None)

class TorchScriptModelDataset(filepath, map_location='cpu', fs_args=None, credentials=None)

Bases: AbstractDataSet

Kedro DataSet for a model to be (de-)serialized with torch.jit.{load,save}

__init__(filepath, map_location='cpu', fs_args=None, credentials=None)

fontr.pipeline_registry module

Project pipelines.

register_pipelines()

Register the project’s pipelines.

Return type:

Dict[str, Pipeline]

Returns:

A mapping from pipeline names to Pipeline objects.

fontr.settings module

File needed by kedro