saribasmetehan
commited on
Commit
•
eaa6e6d
1
Parent(s):
20a2369
Update README.md
Browse files
README.md
CHANGED
@@ -23,6 +23,8 @@ model-index:
|
|
23 |
- name: F1
|
24 |
type: f1
|
25 |
value: 0.7821495486288537
|
|
|
|
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -37,16 +39,58 @@ It achieves the following results on the evaluation set:
|
|
37 |
|
38 |
## Model description
|
39 |
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
50 |
## Training procedure
|
51 |
|
52 |
### Training hyperparameters
|
@@ -75,4 +119,4 @@ The following hyperparameters were used during training:
|
|
75 |
- Transformers 4.41.2
|
76 |
- Pytorch 2.3.0+cu121
|
77 |
- Datasets 2.19.2
|
78 |
-
- Tokenizers 0.19.1
|
|
|
23 |
- name: F1
|
24 |
type: f1
|
25 |
value: 0.7821495486288537
|
26 |
+
language:
|
27 |
+
- tr
|
28 |
---
|
29 |
|
30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
39 |
|
40 |
## Model description
|
41 |
|
42 |
+
This model is a fine-tuned version of dbmdz/bert-base-turkish-uncased on the turkish-wiki_ner dataset. The training dataset consists of 18,967 samples, and the validation dataset consists of 1,000 samples, both derived from Wikipedia data.
|
43 |
+
|
44 |
+
|
45 |
+
For more detailed information, please visit this link: https://huggingface.co/datasets/turkish-nlp-suite/turkish-wikiNER
|
46 |
+
-
|
47 |
+
Labels:
|
48 |
+
|
49 |
+
<ul>
|
50 |
+
<li>CARDINAL</li>
|
51 |
+
<li>DATE</li>
|
52 |
+
<li>EVENT</li>
|
53 |
+
<li>FAC</li>
|
54 |
+
<li>GPE</li>
|
55 |
+
<li>LANGUAGE</li>
|
56 |
+
<li>LAW</li>
|
57 |
+
<li>LOC</li>
|
58 |
+
<li>MONEY</li>
|
59 |
+
<li>NORP</li>
|
60 |
+
<li>ORDINAL</li>
|
61 |
+
<li>ORG</li>
|
62 |
+
<li>PERCENT</li>
|
63 |
+
<li>PERSON</li>
|
64 |
+
<li>PRODUCT</li>
|
65 |
+
<li>QUANTITY</li>
|
66 |
+
<li>TIME</li>
|
67 |
+
<li>TITLE</li>
|
68 |
+
<li>WORK_OF_ART</li>
|
69 |
+
</ul>
|
70 |
+
|
71 |
+
Fine-Tuning Process : https://github.com/saribasmetehan/bert-base-turkish-uncased-ner
|
72 |
+
-
|
73 |
+
## Example
|
74 |
+
```markdown
|
75 |
+
from transformers import pipeline
|
76 |
+
text = "Bu toplam sıfır ise, Newton'ın birinci yasası cismin hareket durumunun değişmeyeceğini söyler."
|
77 |
+
model_id = saribasmetehan/bert-base-turkish-uncased-ner"
|
78 |
+
ner = pipeline("ner",model = model_id)
|
79 |
+
preds= ner(text, aggregation_strategy = "simple")
|
80 |
+
|
81 |
+
pd.DataFrame(preds)
|
82 |
+
|
83 |
+
```
|
84 |
+
|
85 |
+
## Load model directly
|
86 |
+
```markdown
|
87 |
+
from transformers import AutoModelForTokenClassification, AutoTokenizer
|
88 |
+
|
89 |
+
model_name = "saribasmetehan/bert-base-turkish-uncased-ner"
|
90 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
91 |
+
model = AutoModelForTokenClassification.from_pretrained(model_name)
|
92 |
+
|
93 |
+
```
|
94 |
## Training procedure
|
95 |
|
96 |
### Training hyperparameters
|
|
|
119 |
- Transformers 4.41.2
|
120 |
- Pytorch 2.3.0+cu121
|
121 |
- Datasets 2.19.2
|
122 |
+
- Tokenizers 0.19.1
|