File size: 1,777 Bytes
af314f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2d9e711
 
 
 
 
 
af314f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: mit
base_model: facebook/bart-large-cnn
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: bart-large-combined
  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. -->

# bart-large-combined

This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.1837
- Rouge1: 40.9079
- Rouge2: 15.6807
- Rougel: 27.7883
- Rougelsum: 37.5415
- Gen Len: 95.0854

## 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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- 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 | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
| 2.2593        | 1.0   | 5435  | 2.2562          | 39.8535 | 14.987  | 26.8435 | 36.4216   | 96.1483 |
| 1.8173        | 2.0   | 10870 | 2.1837          | 40.9079 | 15.6807 | 27.7883 | 37.5415   | 95.0854 |
| 1.4421        | 3.0   | 16305 | 2.2059          | 41.3435 | 15.9165 | 28.0347 | 37.9954   | 96.1373 |


### Framework versions

- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2