File size: 1,733 Bytes
e12b197
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d573623
 
e12b197
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d573623
e12b197
 
 
 
 
 
d573623
 
 
 
 
 
e12b197
 
 
 
 
 
 
 
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
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: edos-2023-baseline-albert-base-v2-label_vector
  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. -->

# edos-2023-baseline-albert-base-v2-label_vector

This model is a fine-tuned version of [albert-base-v2](https://huggingface.co/albert-base-v2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.8762
- F1: 0.1946

## 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: 1e-05
- 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
- lr_scheduler_warmup_steps: 5
- num_epochs: 12
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | F1     |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 2.1002        | 1.18  | 100  | 1.9982          | 0.1023 |
| 1.7832        | 2.35  | 200  | 1.8435          | 0.1310 |
| 1.57          | 3.53  | 300  | 1.8097          | 0.1552 |
| 1.3719        | 4.71  | 400  | 1.8216          | 0.1631 |
| 1.2072        | 5.88  | 500  | 1.8138          | 0.1811 |
| 1.0186        | 7.06  | 600  | 1.8762          | 0.1946 |


### Framework versions

- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2