Adapter Based BERT code which uses bottleneck adapters trained on relations from ConceptNet which acts as a source of external commonsense knowledge to improve downstream NLP task of multichoice re…

Network Course project

💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

PyTorch original implementation of Cross-lingual Language Model Pretraining.

An Open-Source Library for Training Binarized Neural Networks

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

DeepRED: Deep Image Prior Powered by RED