File size: 1,601 Bytes
7e03c10
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
language:
- hi
license: apache-2.0
base_model: openai/whisper-large
tags:
- generated_from_trainer
model-index:
- name: Whisper_e8eae673-8dea-4ce6-b9ac-7541bbcff1c8
  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. -->

# Whisper_e8eae673-8dea-4ce6-b9ac-7541bbcff1c8

This model is a fine-tuned version of [openai/whisper-large](https://huggingface.co/openai/whisper-large) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0833

## 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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 20
- training_steps: 50

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 3.3764        | 10.0  | 10   | 2.3759          |
| 0.7583        | 20.0  | 20   | 1.4695          |
| 8.3335        | 30.0  | 30   | 6.5259          |
| 6.1162        | 40.0  | 40   | 3.0373          |
| 0.468         | 50.0  | 50   | 2.0833          |


### Framework versions

- Transformers 4.34.1
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1