BatchEncoding hold the output of the encode and batch_encode methods (tokens, attention_masks, etc). This class is derived from a python Dict

r""" A reader for H5 files containing pre-extracted image features. A typical image features file should have at least two H5 datasets, na

Defines a general datatype. Every dataset consists of one or more types of data. For instance, a machine translation dataset contains paired

Bleu scorer.

Character span in the original string Args: start: index of the first character in the original string end: index of

Main Class to compute the CIDEr metric

CIDEr scorer.

Multi-head attention layer with Dropout and Layer Normalization.

Position-wise feed forward layer

Class for computing ROUGE-L score for a set of candidate sentences for the MS COCO test set

Scaled dot-product attention

Token span in an encoded string (list of tokens) Args: start: index of the first token in the span end: index of the

r""" A reader for annotation files containing training captions. These are JSON files in COCO format. Parameters ---------- capti

r""" A PyTorch :class:`~torch.utils.data.Dataset` providing image features for inference. When wrapped with a :class:`~torch.utils.data.DataLo

r""" A PyTorch :class:`~torch.utils.data.Dataset` providing image features for inference, along with constraints for :class:`~updown.modules.c

Defines a single training or test example. Stores each column of the example as an attribute.

Enum with more explicit error message for missing values.

GPT-2 BPE tokenizer. Peculiarities: - Byte-level Byte-Pair-Encoding - Requires a space to start the input string => the encoding methods

Constructs a "Fast" GPT-2 BPE tokenizer (backed by HuggingFace's `tokenizers` library). Peculiarities: - Byte-level Byte-Pair-Encoding

Python wrapper of Stanford PTBTokenizer

Base class for all slow tokenizers. Handle all the shared methods for tokenization and special tokens as well as methods downloading/caching

Base class for slow and fast tokenizers. Handle shared (mostly boiler plate) methods for slow and fast tokenizers.

Base class for all fast tokenizers (wrapping HuggingFace tokenizers library). Inherit from PreTrainedTokenizer. Handle all the shared metho

Scaled dot-product attention with memory

SpecialTokensMixin is derived by ``PreTrainedTokenizer`` and ``PreTrainedTokenizerFast`` and handles specific behaviors related to special to

r""" A PyTorch `:class:`~torch.utils.data.Dataset` providing access to COCO train2017 captions data for training :class:`~updown.models.updown

Defines a vocabulary object that will be used to numericalize a field. Attributes: freqs: A collections.Counter object holding the freque

Defines a vocabulary object that will be used to numericalize a field. Attributes: freqs: A collections.Counter object holding the freque

Descriptor that mimics @property but caches output in member variable. From tensorflow_datasets Built-in in functools from Python 3.8.