File size: 3,595 Bytes
497e812
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
93
94
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: Finetuned_FLAN-T5_VALUE_adapterfusion_lr5e-4_bs32
  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. -->

# Finetuned_FLAN-T5_VALUE_adapterfusion_lr5e-4_bs32

This model is a fine-tuned version of [liuyanchen1015/FLAN-T5_GLUE_finetuning_lr3e-4](https://huggingface.co/liuyanchen1015/FLAN-T5_GLUE_finetuning_lr3e-4) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0870
- Accuracy: 0.8692

## 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: 0.0005
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 0.0862        | 0.08  | 2500  | 0.1059          | 0.8526   |
| 0.092         | 0.17  | 5000  | 0.1025          | 0.856    |
| 0.0943        | 0.25  | 7500  | 0.1126          | 0.8516   |
| 0.0899        | 0.34  | 10000 | 0.0955          | 0.8578   |
| 0.0896        | 0.42  | 12500 | 0.1046          | 0.8564   |
| 0.0952        | 0.51  | 15000 | 0.0978          | 0.851    |
| 0.0901        | 0.59  | 17500 | 0.0958          | 0.8498   |
| 0.095         | 0.68  | 20000 | 0.0974          | 0.8532   |
| 0.0955        | 0.76  | 22500 | 0.0982          | 0.853    |
| 0.0912        | 0.85  | 25000 | 0.0980          | 0.853    |
| 0.0913        | 0.93  | 27500 | 0.0944          | 0.8528   |
| 0.0889        | 1.02  | 30000 | 0.0907          | 0.8592   |
| 0.0871        | 1.1   | 32500 | 0.0933          | 0.855    |
| 0.0872        | 1.18  | 35000 | 0.0904          | 0.861    |
| 0.0859        | 1.27  | 37500 | 0.0879          | 0.8594   |
| 0.0847        | 1.35  | 40000 | 0.0950          | 0.8584   |
| 0.0827        | 1.44  | 42500 | 0.0909          | 0.8622   |
| 0.0836        | 1.52  | 45000 | 0.0933          | 0.8552   |
| 0.0805        | 1.61  | 47500 | 0.0928          | 0.8646   |
| 0.0799        | 1.69  | 50000 | 0.0905          | 0.8648   |
| 0.0789        | 1.78  | 52500 | 0.0863          | 0.87     |
| 0.0786        | 1.86  | 55000 | 0.0907          | 0.8612   |
| 0.0772        | 1.95  | 57500 | 0.0883          | 0.8672   |
| 0.075         | 2.03  | 60000 | 0.0886          | 0.8664   |
| 0.0727        | 2.12  | 62500 | 0.0878          | 0.8688   |
| 0.0724        | 2.2   | 65000 | 0.0881          | 0.8708   |
| 0.0729        | 2.28  | 67500 | 0.0879          | 0.8664   |
| 0.0714        | 2.37  | 70000 | 0.0883          | 0.8694   |
| 0.0694        | 2.45  | 72500 | 0.0876          | 0.8724   |
| 0.0698        | 2.54  | 75000 | 0.0869          | 0.8698   |
| 0.0706        | 2.62  | 77500 | 0.0872          | 0.8712   |
| 0.0685        | 2.71  | 80000 | 0.0874          | 0.8692   |
| 0.068         | 2.79  | 82500 | 0.0873          | 0.869    |
| 0.0685        | 2.88  | 85000 | 0.0863          | 0.8688   |
| 0.068         | 2.96  | 87500 | 0.0870          | 0.8692   |


### Framework versions

- Transformers 4.26.1
- Pytorch 1.13.0+cu117
- Datasets 2.10.1
- Tokenizers 0.12.1