This class is a light wrapper around an allocated buffer which adds shape information. 1. it can be resized, reshaped, copied, and assigned; 2. it ca

Class for 1.31 unsigned floating-point computation

@brief exception indicating access out of the defined range @sa https://json.nlohmann.me/api/basic_json/out_of_range/

Model variable as a Numpy array.

Tag type for binary construction.

Implements a standard indexed vocabulary.

! Input adapter for a (caching) istream. Ignores a UFT Byte Order Mark at beginning of input. Does not support changing the underlying std::streambuf

@brief exception indicating executing a member function with a wrong type @sa https://json.nlohmann.me/api/basic_json/type_error/

Converts models generated by OpenNMT-py.

An example is a collection of sequences or streams (e.g. source and target).

Converts models trained with Fairseq.

Configuration for language models.

Converts models trained with Marian.

Implements multinomial sampling with replacement.

Converts GPT-2 models from https://github.com/openai/gpt-2.

Converts models generated by OpenNMT-py.

Converts models trained with OPUS-MT.

Model variable as a PyTorch tensor.

Configuration for sequence-to-sequence models.

A pool of threads.

Configuration for Transformer models.

Configuration for Transformer decoder models.

Configuration for Transformer encoder models.