Gorengoz commited on
Commit
4ddb3dd
1 Parent(s): 0c74da2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +56 -3
README.md CHANGED
@@ -1,3 +1,56 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - tr
5
+ metrics:
6
+ - accuracy
7
+ - f1
8
+ pipeline_tag: token-classification
9
+ tags:
10
+ - ner
11
+ - turkish-ner
12
+ - turkish
13
+ - nlp
14
+ ---
15
+
16
+ Bu model "https://github.com/stefan-it/turkish-bert" base alınarak geliştirilmiş bir NER(Varlık ismi tanıma) modelidir.
17
+
18
+
19
+ ## Eğitim ve validasyon verisi
20
+
21
+ Fine-tune işlemi için TDD-NER-202112-CC-001 veri seti kullanılmıştır.
22
+ @article{tur-etal-2003-milliyet,
23
+ title={A statistical information extraction system for Turkish},
24
+ volume={9},
25
+ DOI={10.1017/S135132490200284X},
26
+ number={2},
27
+ journal={Natural Language Engineering},
28
+ publisher={Cambridge University Press},
29
+ author={Tür, Gökhan and Hakkani-Tür, Dilek and Oflazer, Kemal},
30
+ year={2003},
31
+ pages={181–210}
32
+ }
33
+
34
+
35
+ ### Hiperparametreler
36
+
37
+ custom_labels = ["O","B-LOC","I-LOC","B-ORG","I-ORG","B-PER","I-PER"]
38
+ model_args = {
39
+ "train_batch_size": 32,
40
+ "eval_batch_size": 32,
41
+ "num_train_epochs": 3,
42
+ "seed":1,
43
+ "save_steps": 625,
44
+ "overwrite_output_dir": True,
45
+ "output_dir": "/content/Model"
46
+ }
47
+
48
+ ### Nasıl Kullanılacağı
49
+
50
+ ```
51
+ from transformers import pipeline
52
+
53
+ pipe = pipeline("token-classification", model="Gorengoz/bert-based-Turkish-NER")
54
+ pipe("Entity X'in müşteri hizmetleri hızlı ve etkili, Entity Y'nin ürün kalitesi çok kötü.",aggregation_strategy = "simple"")
55
+
56
+ ```