File size: 3,679 Bytes
211a1ae
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
95
96
97
98
99
100
101
102
103
104
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: ec-biogpt-noised-pubmed-v2
  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. -->

# ec-biogpt-noised-pubmed-v2

This model is a fine-tuned version of [microsoft/biogpt](https://huggingface.co/microsoft/biogpt) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2703

## 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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 5
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 1.1503        | 0.11  | 500   | 1.3369          |
| 1.3766        | 0.21  | 1000  | 1.2721          |
| 1.3523        | 0.32  | 1500  | 1.2516          |
| 1.3123        | 0.43  | 2000  | 1.2394          |
| 1.1954        | 0.54  | 2500  | 1.2265          |
| 1.226         | 0.64  | 3000  | 1.2182          |
| 1.1269        | 0.75  | 3500  | 1.2118          |
| 1.212         | 0.86  | 4000  | 1.2053          |
| 1.3253        | 0.96  | 4500  | 1.1984          |
| 1.0722        | 1.07  | 5000  | 1.2016          |
| 1.1208        | 1.18  | 5500  | 1.2009          |
| 1.132         | 1.28  | 6000  | 1.1992          |
| 1.1228        | 1.39  | 6500  | 1.1967          |
| 1.1529        | 1.5   | 7000  | 1.1918          |
| 1.0342        | 1.61  | 7500  | 1.1916          |
| 1.0881        | 1.71  | 8000  | 1.1889          |
| 1.084         | 1.82  | 8500  | 1.1852          |
| 1.1409        | 1.93  | 9000  | 1.1807          |
| 0.9794        | 2.03  | 9500  | 1.2098          |
| 0.9821        | 2.14  | 10000 | 1.2146          |
| 0.9695        | 2.25  | 10500 | 1.2096          |
| 0.9866        | 2.35  | 11000 | 1.2088          |
| 1.0305        | 2.46  | 11500 | 1.2059          |
| 0.9532        | 2.57  | 12000 | 1.2060          |
| 0.9978        | 2.68  | 12500 | 1.2041          |
| 1.0013        | 2.78  | 13000 | 1.2006          |
| 1.0401        | 2.89  | 13500 | 1.2023          |
| 1.0899        | 3.0   | 14000 | 1.1988          |
| 0.8229        | 3.1   | 14500 | 1.2410          |
| 0.8598        | 3.21  | 15000 | 1.2420          |
| 0.9295        | 3.32  | 15500 | 1.2414          |
| 0.8477        | 3.43  | 16000 | 1.2386          |
| 0.9302        | 3.53  | 16500 | 1.2382          |
| 0.8284        | 3.64  | 17000 | 1.2374          |
| 0.8242        | 3.75  | 17500 | 1.2410          |
| 0.8422        | 3.85  | 18000 | 1.2346          |
| 0.8742        | 3.96  | 18500 | 1.2362          |
| 0.798         | 4.07  | 19000 | 1.2667          |
| 0.7821        | 4.17  | 19500 | 1.2701          |
| 0.7788        | 4.28  | 20000 | 1.2714          |
| 0.7701        | 4.39  | 20500 | 1.2702          |
| 0.7348        | 4.5   | 21000 | 1.2722          |
| 0.762         | 4.6   | 21500 | 1.2705          |
| 0.7385        | 4.71  | 22000 | 1.2705          |
| 0.7837        | 4.82  | 22500 | 1.2695          |
| 0.8371        | 4.92  | 23000 | 1.2703          |


### Framework versions

- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3