inet.package

Subpackages

• inet.data package
  • Submodules
  • inet.data.augmentation module
  • inet.data.constants module
  • inet.data.datasets module
  • inet.data.load_dataset module
  • inet.data.visualization module
• inet.losses package
  • Submodules
  • inet.losses.giou_loss module
• inet.models package
  • Subpackages
    • inet.models.architectures package
      • Submodules
      • inet.models.architectures.base_model module
      • inet.models.architectures.bounding_boxes module
      • inet.models.architectures.classifier module
    • inet.models.solvers package
      • Submodules
      • inet.models.solvers.common module
      • inet.models.solvers.independent module
      • inet.models.solvers.tf_lite module
      • inet.models.solvers.two_in_one module
      • inet.models.solvers.two_stage module
    • inet.models.tf_lite package
      • Submodules
      • inet.models.tf_lite.convert_to_tflite module
      • inet.models.tf_lite.tflite_methods module
  • Submodules
  • inet.models.data_structures module
  • inet.models.hyper_parameter_optimization module

Submodules

inet.helpers module

Common helper methods that are required in most sub-directories of the current directory

copy_model(input_model, layer_range: Tuple[int, int] = (0, - 1))

Method to copy layer_range from a model into a new model instance.

Parameters

• input_model – the model to copy

• layer_range – the range of layers to copy from the input model

Returns

a new model having layer_range based on the input_model

extract_labels_and_features(dataset)

Helper to extract labels and features from a given data set

Parameters

dataset – dataset to extract data from

Returns

tuple of features and labels

get_train_logdir(name: Optional[str] = None)

Helper function to populate the current time into a given name

Example:

>>> get_train_logdir('foo')
'../train_logs/foo-run_2022_06_09-13-21-50'

Parameters

name – name prefix for the directory

Returns

generated directory name