File size: 3,642 Bytes
0020927
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
91
92
93
94
95
96
97
98
99
---
tags:
- generated_from_trainer
model-index:
- name: ''
  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. -->

# 

This model was trained from scratch on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3350
- Wer: 0.3441

## 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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 400
- num_epochs: 20.0
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 2.2943        | 0.51  | 400   | 4.4367          | 0.9745 |
| 2.8816        | 1.02  | 800   | 2.8103          | 1.2576 |
| 2.7842        | 1.53  | 1200  | 2.6832          | 1.1330 |
| 2.313         | 2.04  | 1600  | 2.6678          | 1.1348 |
| 2.2897        | 2.55  | 2000  | 2.5744          | 1.2102 |
| 1.3821        | 3.06  | 2400  | 2.9908          | 0.9377 |
| 1.074         | 3.57  | 2800  | 2.6649          | 0.8966 |
| 1.3643        | 4.08  | 3200  | 1.0064          | 0.7236 |
| 0.8286        | 4.59  | 3600  | 0.6339          | 0.5454 |
| 0.3872        | 5.1   | 4000  | 0.5170          | 0.4718 |
| 0.3654        | 5.61  | 4400  | 0.4386          | 0.4420 |
| 0.2672        | 6.12  | 4800  | 0.5186          | 0.4679 |
| 0.2519        | 6.63  | 5200  | 0.4238          | 0.4177 |
| 0.3293        | 7.14  | 5600  | 0.3584          | 0.3970 |
| 0.314         | 7.65  | 6000  | 0.3325          | 0.3911 |
| 0.1698        | 8.16  | 6400  | 0.3411          | 0.3855 |
| 0.1682        | 8.67  | 6800  | 0.3239          | 0.3801 |
| 0.1325        | 9.18  | 7200  | 0.3474          | 0.3832 |
| 0.1577        | 9.69  | 7600  | 0.3289          | 0.3839 |
| 0.2259        | 10.2  | 8000  | 0.3183          | 0.3756 |
| 0.2473        | 10.71 | 8400  | 0.3132          | 0.3654 |
| 0.1136        | 11.22 | 8800  | 0.3242          | 0.3670 |
| 0.108         | 11.73 | 9200  | 0.3201          | 0.3601 |
| 0.0806        | 12.24 | 9600  | 0.3223          | 0.3609 |
| 0.0896        | 12.75 | 10000 | 0.3228          | 0.3584 |
| 0.1642        | 13.27 | 10400 | 0.3140          | 0.3547 |
| 0.1442        | 13.77 | 10800 | 0.3235          | 0.3539 |
| 0.0802        | 14.29 | 11200 | 0.3175          | 0.3553 |
| 0.0747        | 14.8  | 11600 | 0.3126          | 0.3512 |
| 0.0488        | 15.31 | 12000 | 0.3292          | 0.3525 |
| 0.0469        | 15.82 | 12400 | 0.3231          | 0.3504 |
| 0.1021        | 16.33 | 12800 | 0.3230          | 0.3502 |
| 0.0841        | 16.84 | 13200 | 0.3348          | 0.3513 |
| 0.0502        | 17.35 | 13600 | 0.3318          | 0.3486 |
| 0.059         | 17.86 | 14000 | 0.3359          | 0.3462 |
| 0.0417        | 18.37 | 14400 | 0.3310          | 0.3467 |
| 0.0322        | 18.88 | 14800 | 0.3325          | 0.3467 |
| 0.0531        | 19.39 | 15200 | 0.3357          | 0.3449 |
| 0.0623        | 19.9  | 15600 | 0.3350          | 0.3441 |


### Framework versions

- Transformers 4.17.0.dev0
- Pytorch 1.10.0+cu113
- Datasets 1.18.1
- Tokenizers 0.11.0