File size: 1,768 Bytes
5b99b55
 
 
 
 
38c9614
 
5b99b55
 
 
 
 
 
 
 
 
 
 
38c9614
dbce8f3
 
4786672
 
dbce8f3
 
4786672
dbce8f3
 
38c9614
5b99b55
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dcf8d00
5b99b55
 
 
dbce8f3
 
 
 
5b99b55
 
 
 
 
 
 
 
54a274c
 
 
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
---
tags:
- generated_from_trainer
datasets:
- fleurs
metrics:
- wer
model-index:
- name: experiment
  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. -->

# experiment

This model is a fine-tuned version of [](https://huggingface.co/) on the fleurs dataset.
It achieves the following results on the evaluation set:
- Loss: 7.9280
- Cer: 1.0
- Wer: 1.0
- Mer: 1.0
- Wil: 1
- Wip: 0
- Hits: 0
- Substitutions: 0
- Deletions: 34
- Insertions: 0

## 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: 0.0001
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 10

### Training results

| Training Loss | Epoch | Step | Validation Loss | Cer | Wer | Mer | Wil | Wip | Hits | Substitutions | Deletions | Insertions |
|:-------------:|:-----:|:----:|:---------------:|:---:|:---:|:---:|:---:|:---:|:----:|:-------------:|:---------:|:----------:|
| 8.0922        | 0.0   | 4    | 7.9783          | 1.0 | 1.0 | 1.0 | 1   | 0   | 0    | 0             | 34        | 0          |
| 7.9733        | 0.01  | 8    | 7.9280          | 1.0 | 1.0 | 1.0 | 1   | 0   | 0    | 0             | 34        | 0          |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.0
- Datasets 2.18.0
- Tokenizers 0.15.2

### Wandb run
https://wandb.ai/alexander-polok/huggingface/runs/imoe89wk