File size: 4,558 Bytes
4bc9036
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
100
101
102
103
104
105
106
107
108
109
110
111
112
113
---
license: cc-by-4.0
base_model: carlosdanielhernandezmena/wav2vec2-large-xlsr-53-icelandic-ep10-1000h
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Icelandic_model_aug_checkpoints
  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. -->

# Icelandic_model_aug_checkpoints

This model is a fine-tuned version of [carlosdanielhernandezmena/wav2vec2-large-xlsr-53-icelandic-ep10-1000h](https://huggingface.co/carlosdanielhernandezmena/wav2vec2-large-xlsr-53-icelandic-ep10-1000h) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2766
- Wer: 0.2136

## 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: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| No log        | 0.2   | 10   | 0.6818          | 0.3244 |
| No log        | 0.4   | 20   | 0.4570          | 0.2841 |
| No log        | 0.6   | 30   | 0.4281          | 0.2562 |
| No log        | 0.8   | 40   | 0.4284          | 0.2282 |
| 0.8009        | 1.0   | 50   | 0.4335          | 0.2260 |
| 0.8009        | 1.2   | 60   | 0.4302          | 0.2159 |
| 0.8009        | 1.4   | 70   | 0.4113          | 0.2114 |
| 0.8009        | 1.6   | 80   | 0.3847          | 0.2159 |
| 0.8009        | 1.8   | 90   | 0.3795          | 0.2148 |
| 0.4567        | 2.0   | 100  | 0.3716          | 0.2058 |
| 0.4567        | 2.2   | 110  | 0.3848          | 0.2081 |
| 0.4567        | 2.4   | 120  | 0.3797          | 0.2036 |
| 0.4567        | 2.6   | 130  | 0.3695          | 0.2036 |
| 0.4567        | 2.8   | 140  | 0.3545          | 0.1969 |
| 0.3723        | 3.0   | 150  | 0.3433          | 0.1957 |
| 0.3723        | 3.2   | 160  | 0.3213          | 0.1957 |
| 0.3723        | 3.4   | 170  | 0.3068          | 0.1913 |
| 0.3723        | 3.6   | 180  | 0.3049          | 0.1902 |
| 0.3723        | 3.8   | 190  | 0.3039          | 0.1991 |
| 0.3259        | 4.0   | 200  | 0.2986          | 0.1913 |
| 0.3259        | 4.2   | 210  | 0.3181          | 0.1969 |
| 0.3259        | 4.4   | 220  | 0.3216          | 0.1857 |
| 0.3259        | 4.6   | 230  | 0.3336          | 0.1913 |
| 0.3259        | 4.8   | 240  | 0.3317          | 0.1980 |
| 0.2809        | 5.0   | 250  | 0.2948          | 0.2025 |
| 0.2809        | 5.2   | 260  | 0.3165          | 0.1879 |
| 0.2809        | 5.4   | 270  | 0.3289          | 0.2081 |
| 0.2809        | 5.6   | 280  | 0.3071          | 0.1913 |
| 0.2809        | 5.8   | 290  | 0.3116          | 0.1957 |
| 0.2711        | 6.0   | 300  | 0.2441          | 0.2136 |
| 0.2711        | 6.2   | 310  | 0.2961          | 0.2013 |
| 0.2711        | 6.4   | 320  | 0.3473          | 0.2069 |
| 0.2711        | 6.6   | 330  | 0.2201          | 0.2058 |
| 0.2711        | 6.8   | 340  | 0.2004          | 0.1991 |
| 0.2578        | 7.0   | 350  | 0.2213          | 0.2081 |
| 0.2578        | 7.2   | 360  | 0.3698          | 0.2081 |
| 0.2578        | 7.4   | 370  | 0.3921          | 0.2114 |
| 0.2578        | 7.6   | 380  | 0.3861          | 0.2204 |
| 0.2578        | 7.8   | 390  | 0.3792          | 0.2271 |
| 0.2503        | 8.0   | 400  | 0.3084          | 0.2092 |
| 0.2503        | 8.2   | 410  | 0.3273          | 0.2058 |
| 0.2503        | 8.4   | 420  | 0.3965          | 0.2069 |
| 0.2503        | 8.6   | 430  | 0.2838          | 0.2036 |
| 0.2503        | 8.8   | 440  | 0.3243          | 0.2148 |
| 0.2485        | 9.0   | 450  | 0.3139          | 0.2136 |
| 0.2485        | 9.2   | 460  | 0.3155          | 0.2036 |
| 0.2485        | 9.4   | 470  | 0.3140          | 0.2103 |
| 0.2485        | 9.6   | 480  | 0.2902          | 0.2349 |
| 0.2485        | 9.8   | 490  | 0.2389          | 0.2047 |
| 0.2567        | 10.0  | 500  | 0.2766          | 0.2136 |


### Framework versions

- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0