places205Dataset#

class places205Dataset(classes=201, path='places205_data', **kwards)#

Bases: DatasetWrapper

A wrapper to the places205 dataset, available at http://places.csail.mit.edu/. The current wrapper class supplyes the required transformations, and also implements the required DatasetWrapper methods.

__init__(classes=201, path='places205_data', **kwards)#

Initializes the places205Dataset wrapper, providing an interface for handling the Places205 dataset. This class is registered under the name ‘Places205’ in the DSFactory.

Parameters:
  • classes (int, optional) – The number of classes in the dataset. Defaults to 201.

  • path (str, optional) – The directory where the dataset is stored. Defaults to ‘places205_data’.

  • **kwargs – Unused keyword arguments, included for compatibility with other dataset constructors.

Methods

__init__([classes, path])

Initializes the places205Dataset wrapper, providing an interface for handling the Places205 dataset.

get_approximation_set()

returns the approximation data set, to be used for range approximation

get_class_labels_dict()

Returns the class_name to index mapping

get_samples_per_class(dataset)

Returns the number of samples in each class or None (if the dataset is balanced - None is equivalent to a list with equal numbers) :param dataset: the dataset split :type dataset: Dataset

get_test_data()

Returns the test data

get_train_data()

Returns the training data

get_train_pipe_ffcv(args)

Returns the Dictionary defining for each field the sequence of ffcv Decoders and transforms to apply.

get_val_data()

Returns the validation data

is_imbalanced()

Always returns False - places205 dataset is balanced

get_test_data()#

Returns the test data

get_train_data()#

Returns the training data

get_val_data()#

Returns the validation data

is_imbalanced()#

Always returns False - places205 dataset is balanced