File size: 1,734 Bytes
e7e940c
 
 
 
 
 
 
 
 
 
 
e0eea3a
 
 
 
 
 
e7e940c
 
 
 
 
 
 
f9ed31a
 
 
 
 
 
 
e7e940c
 
 
 
 
 
 
 
 
 
 
e0eea3a
 
e7e940c
 
 
e0eea3a
 
 
 
 
 
 
 
 
 
 
e7e940c
 
 
 
 
 
 
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
---
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: summarise_v6
  results: []
---

# summarise_v6

This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0497
- Rouge2 Precision: 0.3109
- Rouge2 Recall: 0.406
- Rouge2 Fmeasure: 0.3375

## Model description

More information needed

## Intended uses & limitations

max_input_length = 3072

max_output_length = 1000

led.config.max_length = 1000

led.config.min_length = 100

## 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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
|:-------------:|:-----:|:----:|:---------------:|:----------------:|:-------------:|:---------------:|
| 1.7163        | 0.22  | 10   | 1.2307          | 0.1428           | 0.5118        | 0.2089          |
| 1.632         | 0.44  | 20   | 1.1337          | 0.36             | 0.3393        | 0.3181          |
| 1.0916        | 0.67  | 30   | 1.0738          | 0.2693           | 0.3487        | 0.2731          |
| 1.573         | 0.89  | 40   | 1.0497          | 0.3109           | 0.406         | 0.3375          |

### Framework versions

- Transformers 4.21.3
- Pytorch 1.12.1+cu113
- Datasets 1.2.1
- Tokenizers 0.12.1