sentencepiece

Fast and customizable text tokenization library with BPE and SentencePiece support

🌿 An easy-to-use Japanese Text Processing tool, which makes it possible to switch tokenizers with small changes of code.

Minimal example of using a traced huggingface transformers model with libtorch

R package for Byte Pair Encoding / Unigram modelling based on Sentencepiece

Learning BPE embeddings by first learning a segmentation model and then training word2vec

A Robustly Optimized BERT Pretraining Approach for Vietnamese

Extremely simple and understandable GPT2 implementation with minor tweaks

BERT implementation of PyTorch

To investigate various DNN text classifiers including MLP, CNN, RNN, BERT approaches.

Rust binding for the sentencepiece library

Bengali language Tokenizer (SentencePiece)

Sentencepiece Dart is a wrapper for Google's Sentencepiece C++ library modified

NMT with RNN Models: (1) in Vanilla style, (2) with Sentencepiece, (3) using Pre-trained models from FairSeq

Automated generation of new `The Office` scripts using deep neural networks

This repository contains codes related to the experiments in "An Experimental Evaluation of Japanese Tokenizers for Sentiment-Based Text Classification" presented at https://www.anlp.jp/nlp2021/. Authors: Andre Rusli and Makoto Shishido (Tokyo Denki University).

Bengali SentencePiece Model created with wiki dump data.

Escape unknown symbols in SentecePiece vocabularies

Workshops of natural language processing