6 projects
clddp
A package for training and doing inference with contrastive learning with multiple GPUs (Pytorch-DDP).
crash-ipdb
Trigger ipdb whenever Python crashes
gpl
GPL is an unsupervised domain adaptation method for training dense retrievers. It is based on query generation and pseudo labeling with powerful cross-encoders. To train a domain-adapted model, it needs only the unlabeled target corpus and can achieve significant improvement over zero-shot models.
easy-elasticsearch
An easy-to-use Elasticsearch BM25 interface
faiss-instant
This package contains toolkit for faiss-instant. It mainly helps to encode texts via Transformers and build Faiss indexes in an automatic way.
useb
Heterogenous, Task- and Domain-Specific Benchmark for Unsupervised Sentence Embeddings used in the TSDAE paper.