File size: 3,656 Bytes
99581cd
 
 
 
 
 
 
 
b137eb8
 
99581cd
 
 
 
 
 
2b01398
99581cd
 
 
 
 
 
 
 
 
 
4b459fa
99581cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2b01398
99581cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2b01398
99581cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
---
tags:
- flair
- token-classification
- sequence-tagger-model
language: de
datasets:
- conll2003
widget:
- text: "George Washington ging nach Washington"
---

## German NER in Flair (large model)

This is the large 4-class NER model for German that ships with [Flair](https://github.com/flairNLP/flair/).

F1-Score: **92,31** (CoNLL-03 German revised)

Predicts 4 tags:

| **tag**                        | **meaning** |
|---------------------------------|-----------|
| PER         | person name | 
| LOC         | location name | 
| ORG         | organization name | 
| MISC         | other name | 

Based on document-level XLM-R embeddings and [FLERT](https://arxiv.org/pdf/2011.06993v1.pdf).

---

### Demo: How to use in Flair

Requires: **[Flair](https://github.com/flairNLP/flair/)** (`pip install flair`)

```python
from flair.data import Sentence
from flair.models import SequenceTagger

# load tagger
tagger = SequenceTagger.load("flair/ner-german-large")

# make example sentence
sentence = Sentence("George Washington ging nach Washington")

# predict NER tags
tagger.predict(sentence)

# print sentence
print(sentence)

# print predicted NER spans
print('The following NER tags are found:')
# iterate over entities and print
for entity in sentence.get_spans('ner'):
    print(entity)

```

This yields the following output:
```
Span [1,2]: "George Washington"   [− Labels: PER (1.0)]
Span [5]: "Washington"   [− Labels: LOC (1.0)]
```

So, the entities "*George Washington*" (labeled as a **person**) and "*Washington*" (labeled as a **location**) are found in the sentence "*George Washington ging nach Washington*". 


---

### Training: Script to train this model

The following Flair script was used to train this model: 

```python
import torch

# 1. get the corpus
from flair.datasets import CONLL_03_GERMAN

corpus = CONLL_03_GERMAN()

# 2. what tag do we want to predict?
tag_type = 'ner'

# 3. make the tag dictionary from the corpus
tag_dictionary = corpus.make_tag_dictionary(tag_type=tag_type)

# 4. initialize fine-tuneable transformer embeddings WITH document context
from flair.embeddings import TransformerWordEmbeddings

embeddings = TransformerWordEmbeddings(
    model='xlm-roberta-large',
    layers="-1",
    subtoken_pooling="first",
    fine_tune=True,
    use_context=True,
)

# 5. initialize bare-bones sequence tagger (no CRF, no RNN, no reprojection)
from flair.models import SequenceTagger

tagger = SequenceTagger(
    hidden_size=256,
    embeddings=embeddings,
    tag_dictionary=tag_dictionary,
    tag_type='ner',
    use_crf=False,
    use_rnn=False,
    reproject_embeddings=False,
)

# 6. initialize trainer with AdamW optimizer
from flair.trainers import ModelTrainer

trainer = ModelTrainer(tagger, corpus, optimizer=torch.optim.AdamW)

# 7. run training with XLM parameters (20 epochs, small LR)
from torch.optim.lr_scheduler import OneCycleLR

trainer.train('resources/taggers/ner-german-large',
              learning_rate=5.0e-6,
              mini_batch_size=4,
              mini_batch_chunk_size=1,
              max_epochs=20,
              scheduler=OneCycleLR,
              embeddings_storage_mode='none',
              weight_decay=0.,
              )

)
```



---

### Cite

Please cite the following paper when using this model.

```
@misc{schweter2020flert,
    title={FLERT: Document-Level Features for Named Entity Recognition},
    author={Stefan Schweter and Alan Akbik},
    year={2020},
    eprint={2011.06993},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}
```

---

### Issues?

The Flair issue tracker is available [here](https://github.com/flairNLP/flair/issues/).