Initial commit

README.md

@@ -0,0 +1,60 @@
+---
+language: da
+tags:
+- danish
+- bert
+- masked-lm
+- botxo
+license: cc-by-4.0
+datasets:
+- common_crawl
+- wikipedia
+- dindebat.dk
+- hestenettet.dk
+- danish OpenSubtitles
+pipeline_tag: token-classification
+widget:
+- text: "Jens er en konge og er født i Danmark."
+---
+
+# Danish BERT (version 2, uncased) by [BotXO.ai](https://www.botxo.ai/) finetuned for Named Entity Recognition on the [DaNE dataset](https://danlp.alexandra.dk/304bd159d5de/datasets/ddt.zip) (Hvingelby et al., 2020) by Malte Højmark-Bertelsen
+
+All credit goes to [BotXO.ai](https://www.botxo.ai/) who developed Danish BERT. For data and training details see their [GitHub repository](https://github.com/botxo/nordic_bert) or [this article](https://www.botxo.ai/en/blog/danish-bert-model/).  
+
+It is both available in TensorFlow and Pytorch format. 
+
+The original TensorFlow version can be downloaded using [this link](https://www.dropbox.com/s/19cjaoqvv2jicq9/danish_bert_uncased_v2.zip?dl=1).
+
+
+Here is an example on how to load Danish BERT in PyTorch using the [🤗Transformers](https://github.com/huggingface/transformers) library:
+
+
+
+```python
+from transformers import AutoTokenizer, AutoModelForTokenClassification
+
+tokenizer = AutoTokenizer.from_pretrained("Maltehb/danish-bert-botxo-ner-dane")
+model = AutoModelForTokenClassification.from_pretrained("Maltehb/danish-bert-botxo-ner-dane")
+
+```
+
+### References
+Danish BERT. (2020). BotXO. https://github.com/botxo/nordic_bert (Original work published 2019)
+
+Hvingelby, R., Pauli, A. B., Barrett, M., Rosted, C., Lidegaard, L. M., & Søgaard, A. (2020). DaNE: A Named Entity Resource for Danish. Proceedings of the 12th Language Resources and Evaluation Conference, 4597–4604. https://www.aclweb.org/anthology/2020.lrec-1.565
+
+#### Contact
+
+For help or further information feel free to connect with the author Malte Højmark-Bertelsen on [hjb@kmd.dk](mailto:hjb@kmd.dk?subject=[GitHub]%20DanishBERTUncasedNER) or any of the following platforms:
+
+[<img align="left" alt="MalteHB | Twitter" width="22px" src="https://cdn.jsdelivr.net/npm/simple-icons@v3/icons/twitter.svg" />][twitter]
+[<img align="left" alt="MalteHB | LinkedIn" width="22px" src="https://cdn.jsdelivr.net/npm/simple-icons@v3/icons/linkedin.svg" />][linkedin]
+[<img align="left" alt="MalteHB | Instagram" width="22px" src="https://cdn.jsdelivr.net/npm/simple-icons@v3/icons/instagram.svg" />][instagram]
+
+<br />
+
+</details>
+
+[twitter]: https://twitter.com/malteH_B
+[instagram]: https://www.instagram.com/maltemusen/
+[linkedin]: https://www.linkedin.com/in/malte-h%C3%B8jmark-bertelsen-9a618017b/

config.json

@@ -0,0 +1,50 @@
+{
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "B-PER",
+    "1": "I-PER",
+    "2": "B-LOC",
+    "3": "I-LOC",
+    "4": "B-ORG",
+    "5": "I-ORG",
+    "6": "O",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "type_vocab_size": 2,
+  "vocab_size": 32000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc35edd66a00d96a8195e805fda4b35008c5a3ae9947bff8b9e9d8cb1fa3767e
+size 440224266

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "strip_accents": false}