Arguments#

class Arguments(model, dataset_name, classes, num_epochs, data_dir)#

Bases: object

This class defines the user arguments object, and sets the default values for some parameters

__init__(model, dataset_name, classes, num_epochs, data_dir)#

Initializes the Arguments class, defining user-configurable parameters with their default values.

Parameters:
  • model (str) – The name or path of the model to be used.

  • dataset_name (str) – The name of the dataset for training and evaluation.

  • classes (int) – The number of classes in the dataset.

  • num_epochs (int) – The total number of training epochs.

  • data_dir (str) – The directory path where the dataset is stored.

Methods

__init__(model, dataset_name, classes, ...)

Initializes the Arguments class, defining user-configurable parameters with their default values.

dump_to_file(path)

Dump all the arguments of Arguments class into a file.

dump_to_file(path)#

Dump all the arguments of Arguments class into a file. This method is called when the model checkpoint is saved, and placed next to the checkpoint.