Text Classification
Transformers
PyTorch
Turkish
bert
Eval Results (legacy)
text-embeddings-inference
Instructions to use Ertugrul/deprem_bert_128k_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Ertugrul/deprem_bert_128k_v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Ertugrul/deprem_bert_128k_v2")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("Ertugrul/deprem_bert_128k_v2") model = AutoModelForSequenceClassification.from_pretrained("Ertugrul/deprem_bert_128k_v2") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,80 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- tr
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
- recall
|
| 8 |
+
- f1
|
| 9 |
+
|
| 10 |
+
library_name: transformers
|
| 11 |
+
pipeline_tag: text-classification
|
| 12 |
+
model-index:
|
| 13 |
+
- name: deprem_v1_3
|
| 14 |
+
results:
|
| 15 |
+
- task:
|
| 16 |
+
type: text-classification
|
| 17 |
+
dataset:
|
| 18 |
+
type: deprem_private_dataset_v1_3
|
| 19 |
+
name: deprem_private_dataset_v1_3
|
| 20 |
+
metrics:
|
| 21 |
+
- type: recall
|
| 22 |
+
value: 0.85
|
| 23 |
+
verified: false
|
| 24 |
+
- type: f1
|
| 25 |
+
value: 0.84
|
| 26 |
+
verified: false
|
| 27 |
+
widget:
|
| 28 |
+
- text: >-
|
| 29 |
+
acil acil acil antakyadan istanbula gitmek için antakya expoya ulaşmaya çalışan 19 kişilik bir aile için şehir içi ulaşım desteği istiyoruz. dışardalar üşüyorlar.iletebileceğiniz numaraları bekliyorum
|
| 30 |
+
example_title: Örnek
|
| 31 |
---
|
| 32 |
+
|
| 33 |
+
## Eval Results
|
| 34 |
+
```
|
| 35 |
+
precision recall f1-score support
|
| 36 |
+
|
| 37 |
+
Lojistik 0.81 0.79 0.80 38
|
| 38 |
+
Elektrik Kaynagi 0.73 0.91 0.81 56
|
| 39 |
+
Arama Ekipmani 0.79 0.74 0.76 128
|
| 40 |
+
Cenaze 1.00 0.50 0.67 2
|
| 41 |
+
Giysi 0.82 0.96 0.89 138
|
| 42 |
+
Enkaz Kaldirma 0.94 0.94 0.94 919
|
| 43 |
+
Isinma 0.84 0.89 0.86 185
|
| 44 |
+
Barınma 0.96 0.96 0.96 483
|
| 45 |
+
Tuvalet 0.67 0.80 0.73 10
|
| 46 |
+
Su 0.83 0.87 0.85 67
|
| 47 |
+
Yemek 0.89 0.96 0.92 202
|
| 48 |
+
Saglik 0.80 0.88 0.83 104
|
| 49 |
+
Alakasiz 0.90 0.82 0.86 377
|
| 50 |
+
|
| 51 |
+
micro avg 0.89 0.91 0.90 2709
|
| 52 |
+
macro avg 0.84 0.85 0.84 2709
|
| 53 |
+
weighted avg 0.90 0.91 0.90 2709
|
| 54 |
+
samples avg 0.91 0.92 0.91 2709
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
```
|
| 58 |
+
|
| 59 |
+
## Threshold:
|
| 60 |
+
- **Best Threshold:** 0.53
|
| 61 |
+
|
| 62 |
+
## Class Loss Weights
|
| 63 |
+
|
| 64 |
+
```python
|
| 65 |
+
[3.017203135650159,
|
| 66 |
+
2.4823691788825464,
|
| 67 |
+
1.941736822154725,
|
| 68 |
+
6.172646581418988,
|
| 69 |
+
1.8759436445637834,
|
| 70 |
+
1.0,
|
| 71 |
+
1.75011143349181,
|
| 72 |
+
1.2730236191357969,
|
| 73 |
+
4.849237178079731,
|
| 74 |
+
2.4857419672410703,
|
| 75 |
+
1.6324480531290084,
|
| 76 |
+
2.0033774839735035,
|
| 77 |
+
1.3688883733394182]
|
| 78 |
+
|
| 79 |
+
```
|
| 80 |
+
|