Metadata-Version: 2.1
Name: strideformer
Version: 0.1.2
Summary: Package to use a two-stage transformer for long-document classification
Home-page: https://github.com/nbroad1881/strideformer
Download-URL: https://github.com/nbroad1881/strideformer
Author: Nicholas Broad
Author-email: nicholas@huggingface.co
License: Apache 2.0
Keywords: nlp,machine learning,transformers
License-File: LICENSE
Requires-Dist: transformers (>=4.21.0)
Provides-Extra: dev
Requires-Dist: black ; extra == 'dev'
Requires-Dist: tabulate ; extra == 'dev'
Requires-Dist: isort ; extra == 'dev'
Requires-Dist: hydra-core (>=1.2.0) ; extra == 'dev'
Requires-Dist: flake8 ; extra == 'dev'
Provides-Extra: quality
Requires-Dist: black ; extra == 'quality'
Requires-Dist: flake8 ; extra == 'quality'
Requires-Dist: isort ; extra == 'quality'
Requires-Dist: tabulate ; extra == 'quality'
Provides-Extra: train
Requires-Dist: hydra-core (>=1.2.0) ; extra == 'train'


A package that includes an implementation of a two-transformer
setup to classify long documents. The first stage is a 
sentence-transformer which embeds chunks of the text. The second 
stage is a generic transformer used to classify the documents
based on a sequence of embeddings from the first model. 
Training scripts included.
