File size: 1,717 Bytes
11bcbb5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6e7b887
 
 
 
 
11bcbb5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6e7b887
 
 
11bcbb5
 
 
 
 
 
 
 
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
base_model: google/mt5-small
tags:
- summarization
- generated_from_trainer
metrics:
- rouge
model-index:
- name: mt5-small-finetuned-cnn_dailymail
  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. -->

# mt5-small-finetuned-cnn_dailymail

This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.3254
- Rouge1: 27.84
- Rouge2: 12.76
- Rougel: 24.39
- Rougelsum: 25.91

## 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: 5.6e-05
- 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
- num_epochs: 3

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:---------:|
| 3.5881        | 1.0   | 3510  | 2.4243          | 26.44  | 11.33  | 22.82  | 24.39     |
| 2.7944        | 2.0   | 7020  | 2.3306          | 27.77  | 12.64  | 24.29  | 25.84     |
| 2.6698        | 3.0   | 10530 | 2.3254          | 27.84  | 12.76  | 24.39  | 25.91     |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1