File size: 1,828 Bytes
9a596e3
55cee1e
 
9a596e3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55cee1e
9a596e3
55cee1e
 
 
 
 
9a596e3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55cee1e
 
 
 
9a596e3
 
 
 
 
 
 
 
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
---
license: mit
base_model: indolem/indobert-base-uncased
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: awkokawokawokoaw
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# awkokawokawokoaw

This model is a fine-tuned version of [indolem/indobert-base-uncased](https://huggingface.co/indolem/indobert-base-uncased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5909
- Accuracy: 0.7917
- Precision: 0.7547
- Recall: 0.7583
- F1: 0.7544

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
| 1.0793        | 1.0   | 169  | 1.0677          | 0.4333   | 0.1444    | 0.3333 | 0.2016 |
| 0.9871        | 2.0   | 338  | 0.9369          | 0.625    | 0.4613    | 0.5010 | 0.4515 |
| 0.7801        | 3.0   | 507  | 0.6453          | 0.76     | 0.7061    | 0.6986 | 0.7008 |
| 0.5823        | 4.0   | 676  | 0.5909          | 0.7917   | 0.7547    | 0.7583 | 0.7544 |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1