File size: 3,658 Bytes
0863110
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
83
---
license: mit
base_model: facebook/bart-large-cnn
tags:
- generated_from_trainer
metrics:
- rouge
- bleu
model-index:
- name: LifeScienceBART
  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. -->

# LifeScienceBART

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: 4.4310
- Rouge1: 52.3694
- Rouge2: 17.5874
- Rougel: 36.4217
- Rougelsum: 48.765
- Bertscore Precision: 82.295
- Bertscore Recall: 83.951
- Bertscore F1: 83.1121
- Bleu: 0.1308
- Gen Len: 227.8869

## 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  |
|:-------------:|:------:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------------------:|:----------------:|:------------:|:------:|:--------:|
| 6.1988        | 0.0881 | 100  | 6.0815          | 43.5317 | 12.8172 | 29.5886 | 40.6668   | 78.4798             | 81.4664          | 79.9395      | 0.0939 | 227.8869 |
| 5.7388        | 0.1762 | 200  | 5.6510          | 41.3899 | 12.8237 | 29.1108 | 38.0304   | 77.5037             | 81.7443          | 79.5601      | 0.0978 | 227.8869 |
| 5.3718        | 0.2643 | 300  | 5.2822          | 46.279  | 14.1045 | 31.7158 | 43.1347   | 79.8268             | 82.2875          | 81.0344      | 0.1041 | 227.8869 |
| 5.1682        | 0.3524 | 400  | 5.1072          | 48.1957 | 15.1732 | 32.7384 | 44.0672   | 80.3745             | 82.94            | 81.6328      | 0.1137 | 227.8869 |
| 5.1315        | 0.4405 | 500  | 4.9408          | 48.9502 | 15.6058 | 33.6297 | 45.5085   | 81.0706             | 83.1289          | 82.0835      | 0.1158 | 227.8869 |
| 4.9456        | 0.5286 | 600  | 4.7786          | 48.4843 | 15.8565 | 34.014  | 45.2987   | 80.9541             | 83.0806          | 81.9998      | 0.1151 | 227.8869 |
| 4.8396        | 0.6167 | 700  | 4.6607          | 51.3313 | 16.5503 | 35.0136 | 47.9755   | 82.0251             | 83.4743          | 82.7408      | 0.1210 | 227.8869 |
| 4.7481        | 0.7048 | 800  | 4.5922          | 51.9257 | 16.9939 | 35.583  | 48.1998   | 82.2219             | 83.8107          | 83.0061      | 0.1262 | 227.8869 |
| 4.6688        | 0.7929 | 900  | 4.5112          | 51.3896 | 17.1313 | 35.8696 | 47.7303   | 81.926              | 83.7943          | 82.8465      | 0.1277 | 227.8869 |
| 4.4321        | 0.8810 | 1000 | 4.4624          | 52.6168 | 17.6855 | 36.2987 | 49.0759   | 82.3644             | 83.8994          | 83.1222      | 0.1305 | 227.8869 |
| 4.5732        | 0.9691 | 1100 | 4.4310          | 52.3694 | 17.5874 | 36.4217 | 48.765    | 82.295              | 83.951           | 83.1121      | 0.1308 | 227.8869 |


### Framework versions

- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1