Commit
•
acdf52f
1
Parent(s):
5bfcc1b
Update README.md
Browse files
README.md
CHANGED
@@ -1,57 +1,19 @@
|
|
1 |
---
|
2 |
license: mit
|
3 |
-
tags:
|
4 |
-
- generated_from_keras_callback
|
5 |
model-index:
|
6 |
- name: indobert-abusive-language-classifier
|
7 |
results: []
|
|
|
|
|
|
|
|
|
8 |
---
|
9 |
|
10 |
-
<!-- This model card has been generated automatically according to the information Keras had access to. You should
|
11 |
-
probably proofread and complete it, then remove this comment. -->
|
12 |
-
|
13 |
# indobert-abusive-language-classifier
|
14 |
|
15 |
-
This model is a fine-tuned version of [indolem/indobert-base-uncased](https://huggingface.co/indolem/indobert-base-uncased) on an unknown dataset.
|
16 |
-
It achieves the following results on the evaluation set:
|
17 |
-
- Train Loss: 0.1777
|
18 |
-
- Train Sparse Categorical Accuracy: 0.9318
|
19 |
-
- Validation Loss: 0.2693
|
20 |
-
- Validation Sparse Categorical Accuracy: 0.9017
|
21 |
-
- Epoch: 2
|
22 |
|
23 |
## Model description
|
24 |
|
25 |
-
|
26 |
-
|
27 |
-
## Intended uses & limitations
|
28 |
-
|
29 |
-
More information needed
|
30 |
-
|
31 |
-
## Training and evaluation data
|
32 |
-
|
33 |
-
More information needed
|
34 |
-
|
35 |
-
## Training procedure
|
36 |
-
|
37 |
-
### Training hyperparameters
|
38 |
-
|
39 |
-
The following hyperparameters were used during training:
|
40 |
-
- optimizer: {'name': 'Adam', 'learning_rate': 5e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
|
41 |
-
- training_precision: float32
|
42 |
-
|
43 |
-
### Training results
|
44 |
-
|
45 |
-
| Train Loss | Train Sparse Categorical Accuracy | Validation Loss | Validation Sparse Categorical Accuracy | Epoch |
|
46 |
-
|:----------:|:---------------------------------:|:---------------:|:--------------------------------------:|:-----:|
|
47 |
-
| 0.4750 | 0.7658 | 0.3018 | 0.8709 | 0 |
|
48 |
-
| 0.2758 | 0.8871 | 0.2713 | 0.8804 | 1 |
|
49 |
-
| 0.1777 | 0.9318 | 0.2693 | 0.9017 | 2 |
|
50 |
-
|
51 |
-
|
52 |
-
### Framework versions
|
53 |
|
54 |
-
- Transformers 4.26.1
|
55 |
-
- TensorFlow 2.11.0
|
56 |
-
- Datasets 2.10.1
|
57 |
-
- Tokenizers 0.13.2
|
|
|
1 |
---
|
2 |
license: mit
|
|
|
|
|
3 |
model-index:
|
4 |
- name: indobert-abusive-language-classifier
|
5 |
results: []
|
6 |
+
tags:
|
7 |
+
- text-classification
|
8 |
+
- bert
|
9 |
+
- indobert
|
10 |
---
|
11 |
|
|
|
|
|
|
|
12 |
# indobert-abusive-language-classifier
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
## Model description
|
16 |
|
17 |
+
This model is an abusive language classifier for Bahasa Indonesia based on [indolem/indobert-base-uncased](https://huggingface.co/indolem/indobert-base-uncased).
|
18 |
+
It was trained on a modified [Multi-label Hate Speech and Abusive Language Detection](https://github.com/okkyibrohim/id-multi-label-hate-speech-and-abusive-language-detection) dataset and achieves an evaluation accuracy of 90.17%.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
|
|
|
|
|
|
|
|