BERT Was 2.5x
📰 Medium · Deep Learning
A technical comparison of BERT-base and DistilBERT-base fine-tuned on the same NER and sentiment tasks — with the architecture internals… Continue reading on Medium »
A technical comparison of BERT-base and DistilBERT-base fine-tuned on the same NER and sentiment tasks — with the architecture internals… Continue reading on Medium »