File size: 3,510 Bytes
3147d3d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
---
license: apache-2.0
base_model: projecte-aina/roberta-base-ca-v2-cased-te
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: SYN_300524_epoch_1
  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. -->

# SYN_300524_epoch_1

This model is a fine-tuned version of [projecte-aina/roberta-base-ca-v2-cased-te](https://huggingface.co/projecte-aina/roberta-base-ca-v2-cased-te) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3748
- Accuracy: 0.961
- Precision: 0.9615
- Recall: 0.961
- F1: 0.9610
- Ratio: 0.483

## 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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 47
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- lr_scheduler_warmup_steps: 4
- num_epochs: 1
- label_smoothing_factor: 0.1

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Precision | Recall | F1     | Ratio |
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:-----:|
| 2.9367        | 0.0533 | 10   | 1.4668          | 0.65     | 0.7117    | 0.65   | 0.6225 | 0.77  |
| 1.1009        | 0.1067 | 20   | 0.6674          | 0.856    | 0.8560    | 0.856  | 0.8560 | 0.502 |
| 0.6993        | 0.16   | 30   | 0.5583          | 0.908    | 0.9095    | 0.9080 | 0.9079 | 0.53  |
| 0.6377        | 0.2133 | 40   | 0.4923          | 0.934    | 0.9343    | 0.9340 | 0.9340 | 0.486 |
| 0.5192        | 0.2667 | 50   | 0.4930          | 0.926    | 0.9282    | 0.9260 | 0.9259 | 0.464 |
| 0.5189        | 0.32   | 60   | 0.4687          | 0.937    | 0.9383    | 0.937  | 0.9370 | 0.527 |
| 0.5083        | 0.3733 | 70   | 0.4321          | 0.944    | 0.9445    | 0.944  | 0.9440 | 0.484 |
| 0.4645        | 0.4267 | 80   | 0.4026          | 0.949    | 0.9490    | 0.949  | 0.9490 | 0.495 |
| 0.4268        | 0.48   | 90   | 0.3990          | 0.949    | 0.9498    | 0.9490 | 0.9490 | 0.479 |
| 0.4327        | 0.5333 | 100  | 0.3949          | 0.952    | 0.9524    | 0.952  | 0.9520 | 0.486 |
| 0.4283        | 0.5867 | 110  | 0.3894          | 0.954    | 0.9540    | 0.954  | 0.9540 | 0.496 |
| 0.4263        | 0.64   | 120  | 0.3829          | 0.957    | 0.9572    | 0.957  | 0.9570 | 0.489 |
| 0.4205        | 0.6933 | 130  | 0.3800          | 0.962    | 0.9620    | 0.962  | 0.9620 | 0.496 |
| 0.4291        | 0.7467 | 140  | 0.3760          | 0.962    | 0.9620    | 0.962  | 0.9620 | 0.502 |
| 0.4124        | 0.8    | 150  | 0.3723          | 0.964    | 0.9641    | 0.964  | 0.9640 | 0.494 |
| 0.4142        | 0.8533 | 160  | 0.3720          | 0.964    | 0.9641    | 0.964  | 0.9640 | 0.492 |
| 0.4209        | 0.9067 | 170  | 0.3767          | 0.96     | 0.9605    | 0.96   | 0.9600 | 0.484 |
| 0.3908        | 0.96   | 180  | 0.3748          | 0.96     | 0.9605    | 0.96   | 0.9600 | 0.484 |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1