File size: 3,676 Bytes
a4a72e6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model: google/pegasus-large
tags:
- generated_from_trainer
metrics:
- rouge
- bleu
model-index:
- name: Health_MainSections_PegasusLargeModel
  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. -->

# Health_MainSections_PegasusLargeModel

This model is a fine-tuned version of [google/pegasus-large](https://huggingface.co/google/pegasus-large) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.2414
- Rouge1: 50.1795
- Rouge2: 17.4664
- Rougel: 33.5547
- Rougelsum: 45.6086
- Bertscore Precision: 79.0557
- Bertscore Recall: 82.3708
- Bertscore F1: 80.6737
- Bleu: 0.1302
- Gen Len: 230.3297

## 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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- 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: 1

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Bertscore Precision | Bertscore Recall | Bertscore F1 | Bleu   | Gen Len  |
|:-------------:|:------:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------------------:|:----------------:|:------------:|:------:|:--------:|
| 6.6491        | 0.0835 | 100  | 6.2799          | 39.9618 | 11.2774 | 25.6909 | 36.1285   | 76.1843             | 80.1799          | 78.1239      | 0.0837 | 230.3297 |
| 6.2174        | 0.1671 | 200  | 5.9405          | 42.5055 | 13.4081 | 28.4622 | 38.6545   | 76.9801             | 80.932           | 78.9         | 0.1013 | 230.3297 |
| 6.0623        | 0.2506 | 300  | 5.7883          | 44.9098 | 14.7089 | 29.9801 | 40.7648   | 77.6093             | 81.3921          | 79.4496      | 0.1118 | 230.3297 |
| 5.8987        | 0.3342 | 400  | 5.6549          | 46.317  | 15.8065 | 30.9576 | 41.9584   | 77.7634             | 81.688           | 79.6712      | 0.1182 | 230.3297 |
| 5.7395        | 0.4177 | 500  | 5.5221          | 47.7104 | 16.4157 | 31.8171 | 43.3162   | 78.0166             | 81.8334          | 79.8733      | 0.1221 | 230.3297 |
| 5.7026        | 0.5013 | 600  | 5.4323          | 47.7147 | 16.492  | 32.171  | 43.3503   | 78.266              | 81.9538          | 80.0615      | 0.1225 | 230.3297 |
| 5.6432        | 0.5848 | 700  | 5.3587          | 49.1073 | 17.0141 | 32.8254 | 44.5003   | 78.7001             | 82.1587          | 80.3864      | 0.1264 | 230.3297 |
| 5.6131        | 0.6684 | 800  | 5.3146          | 49.7061 | 17.3206 | 33.0712 | 45.1323   | 78.8811             | 82.2757          | 80.5371      | 0.1290 | 230.3297 |
| 5.5168        | 0.7519 | 900  | 5.2820          | 50.0791 | 17.439  | 33.333  | 45.4496   | 79.031              | 82.3405          | 80.6464      | 0.1300 | 230.3297 |
| 5.5748        | 0.8355 | 1000 | 5.2569          | 50.176  | 17.4666 | 33.4584 | 45.398    | 79.0421             | 82.3676          | 80.6652      | 0.1301 | 230.3297 |
| 5.4357        | 0.9190 | 1100 | 5.2414          | 50.1795 | 17.4664 | 33.5547 | 45.6086   | 79.0557             | 82.3708          | 80.6737      | 0.1302 | 230.3297 |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.2.1
- Tokenizers 0.19.1