Update README.md

Extend model card.

README.md

@@ -7,11 +7,31 @@ language: de
 datasets:
 - conll2003
 - germeval_14
-- europeananewspapers2016
 license: apache-2.0
 ---
+# About `sbb_ner`
+
 This is a BERT model for named entity recognition in historical German. 
+It can predict the classes `PER`, `LOC` and `ORG`.
+
+The model is based on the 🤗 [Transformers](https://github.com/huggingface/transformers) 
+`BERT Base multi-lingual cased` model. 
+We applied unsupervised pre-training on 2,333,647 pages of 
+unlabeled historical German text from the Berlin State Library 
+digital collections, and supervised pre-training on two datasets 
+with contemporary German text, [conll2003](https://huggingface.co/models?dataset=dataset:conll2003) 
+and [germeval_14](https://huggingface.co/models?dataset=dataset:germeval_14).
+
+# Results
+
+In a 5-fold cross validation with different historical German NER corpora, 
+the model obtained an F1-Score of **84.3**±1.1%. 
 
-It predicts the classes `PER`, `LOC` and `ORG`.
+For details, see our [paper](https://corpora.linguistik.uni-erlangen.de/data/konvens/proceedings/papers/KONVENS2019_paper_4.pdf) 
+or have a look at [sbb_ner](https://github.com/qurator-spk/sbb_ner) on GitHub.
 
-For details, see [https://github.com/qurator-spk/sbb_ner](sbb_ner).
+# Weights
+We provide model weights for PyTorch. 
+| Model | Downloads
+| ------------------------| ------------------------
+| `bert-sbb-de-finetuned` | [`config.json`](https://huggingface.co/SBB/sbb_ner/blob/main/config.json) • [`pytorch_model_ep7.bin`](https://huggingface.co/SBB/sbb_ner/blob/main/pytorch_model_ep7.bin) • [`vocab.txt`](https://huggingface.co/SBB/sbb_ner/blob/main/vocab.txt)