File size: 2,850 Bytes
f733c19
6ce9822
f733c19
2adfb5a
6ce9822
f733c19
 
2adfb5a
f733c19
 
 
 
 
2adfb5a
6ce9822
2adfb5a
6ce9822
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f733c19
 
 
 
 
 
 
49d9e51
f733c19
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6ce9822
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
84
85
86
87
88
89
90
91
92
---
license: apache-2.0
base_model: google/flan-t5-small
datasets:
- grammarly/coedit
tags:
- generated_from_trainer
- text-generation-inference
metrics:
- rouge
model-index:
- name: coedit-small
  results: []
language:
- en
widget:
- text: >-
    Fix the grammar: When I grow up, I start to understand what he said is quite
    right.
  example_title: Fluency
- text: >-
    Make this text coherent: Their flight is weak. They run quickly through the
    tree canopy.
  example_title: Coherence
- text: >-
    Rewrite to make this easier to understand: A storm surge is what forecasters
    consider a hurricane's most treacherous aspect.
  example_title: Simplification
- text: 'Paraphrase this: Do you know where I was born?'
  example_title: Paraphrase
- text: 'Write this more formally: omg i love that song im listening to it right now'
  example_title: Formalize
- text: 'Write in a more neutral way: The authors'' exposé on nutrition studies.'
  example_title: Neutralize
---

<!-- 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. -->

# coedit-small

This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on the [CoEdIT dataset](https://huggingface.co/datasets/grammarly/coedit).
It achieves the following results on the evaluation set:
- Loss: 0.8242
- Rouge1: 58.7504
- Rouge2: 45.1374
- Rougel: 55.4161
- Rougelsum: 55.4599
- Gen Len: 16.5245

## 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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
| 0.9482        | 1.0   | 4317  | 0.8878          | 58.4501 | 44.2623 | 54.4468 | 54.51     | 16.5088 |
| 0.9155        | 2.0   | 8634  | 0.8485          | 58.6609 | 44.7759 | 54.9844 | 55.0503   | 16.5339 |
| 0.8964        | 3.0   | 12951 | 0.8402          | 58.712  | 44.9838 | 55.2171 | 55.2697   | 16.5251 |
| 0.9049        | 4.0   | 17268 | 0.8305          | 58.7767 | 45.1325 | 55.3955 | 55.4522   | 16.5181 |
| 0.8948        | 5.0   | 21585 | 0.8242          | 58.7504 | 45.1374 | 55.4161 | 55.4599   | 16.5245 |


### Framework versions

- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.14.7
- Tokenizers 0.15.0