File size: 2,994 Bytes
4fd75e3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5afc521
 
 
 
 
 
 
 
4fd75e3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12e0bb0
 
5afc521
 
 
 
 
 
 
 
 
 
4fd75e3
 
 
 
 
 
 
 
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
---
base_model: csebuetnlp/banglabert
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: banglabert-MLTC-BB1
  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. -->

# banglabert-MLTC-BB1

This model is a fine-tuned version of [csebuetnlp/banglabert](https://huggingface.co/csebuetnlp/banglabert) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3599
- F1: 0.8582
- F1 Weighted: 0.8565
- Roc Auc: 0.8547
- Accuracy: 0.5835
- Hamming Loss: 0.1452
- Jaccard Score: 0.7516
- Zero One Loss: 0.4165

## 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: 24
- eval_batch_size: 24
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | F1     | F1 Weighted | Roc Auc | Accuracy | Hamming Loss | Jaccard Score | Zero One Loss |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-----------:|:-------:|:--------:|:------------:|:-------------:|:-------------:|
| 0.5692        | 1.0   | 49   | 0.5109          | 0.7781 | 0.7194      | 0.7685  | 0.4216   | 0.2314       | 0.6367        | 0.5784        |
| 0.4149        | 2.0   | 98   | 0.4230          | 0.8469 | 0.8467      | 0.8405  | 0.5604   | 0.1594       | 0.7345        | 0.4396        |
| 0.3732        | 3.0   | 147  | 0.3856          | 0.8479 | 0.8474      | 0.8425  | 0.5527   | 0.1575       | 0.7360        | 0.4473        |
| 0.3321        | 4.0   | 196  | 0.3750          | 0.8542 | 0.8522      | 0.8476  | 0.5578   | 0.1523       | 0.7454        | 0.4422        |
| 0.2817        | 5.0   | 245  | 0.3721          | 0.8545 | 0.8514      | 0.8482  | 0.5630   | 0.1517       | 0.7460        | 0.4370        |
| 0.2781        | 6.0   | 294  | 0.3553          | 0.8561 | 0.8547      | 0.8528  | 0.5656   | 0.1472       | 0.7484        | 0.4344        |
| 0.2264        | 7.0   | 343  | 0.3576          | 0.8566 | 0.8550      | 0.8534  | 0.5733   | 0.1465       | 0.7492        | 0.4267        |
| 0.2441        | 8.0   | 392  | 0.3595          | 0.8575 | 0.8560      | 0.8534  | 0.5733   | 0.1465       | 0.7505        | 0.4267        |
| 0.2547        | 9.0   | 441  | 0.3608          | 0.8561 | 0.8548      | 0.8528  | 0.5784   | 0.1472       | 0.7484        | 0.4216        |
| 0.2211        | 10.0  | 490  | 0.3599          | 0.8582 | 0.8565      | 0.8547  | 0.5835   | 0.1452       | 0.7516        | 0.4165        |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1