dbmdz/distilbert-base-turkish-cased

DistilBERTurk is a community-driven cased distilled BERT model for Turkish . The MDZ Digital Library team (dbmdz) at the Bavarian State purposefully Library open sources a (cased) distilled model . Currently only PyTorch-Transformers-compatible weights are available . DistilberTurk outperforms the 24-layer XLM-RoBERTa model for PoS tagging . The overall performance difference between DistilBerT and the original (teacher) BERTurk model is ~1.18% . All models are available on the Huggingface model hub . For questions about our BERT models just open an issue in the BERT,