Danish NER model v8 - 91.02% F1

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,80 @@
+---
+language: da
+license: mit
+tags:
+- token-classification
+- ner
+- named-entity-recognition
+- danish
+- xlm-roberta
+- scandinavian
+datasets:
+- alexandrainst/dane
+- wikiann
+- tollefj/nordic-ner
+metrics:
+- f1
+- precision
+- recall
+pipeline_tag: token-classification
+model-index:
+- name: danish-ner-xlmr-base
+  results:
+  - task:
+      type: token-classification
+      name: Named Entity Recognition
+    dataset:
+      name: DaNE
+      type: alexandrainst/dane
+      split: validation
+    metrics:
+    - name: F1
+      type: f1
+      value: 0.9102
+---
+
+# Danish NER XLM-RoBERTa (v8)
+
+State-of-the-art Named Entity Recognition model for Danish, fine-tuned from XLM-RoBERTa.
+
+**Updated 2026-02-03**: Now v8 with 91.02% F1 (previously 84.6%)
+
+## Performance
+
+| Benchmark | F1 Score |
+|-----------|----------|
+| **DaNE (validation)** | **91.02%** |
+| Previous version | 84.6% |
+| nbailab baseline | 87.09% |
+
+## Quick Start
+
+```python
+from transformers import pipeline
+
+ner = pipeline("ner", model="thomasbeste/danish-ner-xlmr-base", aggregation_strategy="simple")
+result = ner("Anders Jensen arbejder hos Novo Nordisk i København.")
+
+for entity in result:
+    print(f"{entity['word']}: {entity['entity_group']} ({entity['score']:.2f})")
+```
+
+## Entity Types
+
+| Label | Description | Example |
+|-------|-------------|---------|
+| `PER` | Person names | Anders Jensen |
+| `ORG` | Organizations | Novo Nordisk A/S |
+| `LOC` | Locations | København |
+| `MISC` | Miscellaneous | Dansk |
+
+## Training Data
+
+- DaNE (4.4k samples)
+- WikiANN Danish (20k samples)
+- NorNE Norwegian (30k samples)
+- High-quality synthetic data (60k samples)
+
+## License
+
+MIT

config.json

@@ -0,0 +1,52 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "O",
+    "1": "B-PER",
+    "2": "I-PER",
+    "3": "B-ORG",
+    "4": "I-ORG",
+    "5": "B-LOC",
+    "6": "I-LOC",
+    "7": "B-MISC",
+    "8": "I-MISC"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "B-LOC": 5,
+    "B-MISC": 7,
+    "B-ORG": 3,
+    "B-PER": 1,
+    "I-LOC": 6,
+    "I-MISC": 8,
+    "I-ORG": 4,
+    "I-PER": 2,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.0.0",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 250002
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5925ae3610b89bea5a2485d3d87abfcaf6bbf1a5248e7ebe4633aa112c1d0201
+size 1109863932

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:733035ab05a88e9a792538ad44ddd0ad270dffe3539aa83767434d7a95ab814c
+size 16781751

tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": true,
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f89c71a33f3aa3917d4fb5c8a9e6c3784a951dc7132a62adb7bf5fdab848fd08
+size 4792