File size: 2,854 Bytes
569c136
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: cc-by-nc-sa-4.0
base_model: InstaDeepAI/nucleotide-transformer-500m-1000g
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: nucleotide-transformer-500m-1000g_ft_BioS74_1kbpHG19_DHSs_H3K27AC
  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. -->

# nucleotide-transformer-500m-1000g_ft_BioS74_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of [InstaDeepAI/nucleotide-transformer-500m-1000g](https://huggingface.co/InstaDeepAI/nucleotide-transformer-500m-1000g) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4446
- F1 Score: 0.8119
- Precision: 0.8121
- Recall: 0.8117
- Accuracy: 0.8031
- Auc: 0.8875
- Prc: 0.8848

## 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | F1 Score | Precision | Recall | Accuracy | Auc    | Prc    |
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|:------:|:------:|
| 0.504         | 0.1314 | 500  | 0.4497          | 0.8107   | 0.7871    | 0.8358 | 0.7957   | 0.8712 | 0.8603 |
| 0.4867        | 0.2629 | 1000 | 0.5440          | 0.7578   | 0.8477    | 0.6851 | 0.7707   | 0.8763 | 0.8672 |
| 0.4688        | 0.3943 | 1500 | 0.4343          | 0.8176   | 0.7839    | 0.8543 | 0.8004   | 0.8833 | 0.8815 |
| 0.4524        | 0.5258 | 2000 | 0.4655          | 0.8244   | 0.7696    | 0.8875 | 0.8020   | 0.8830 | 0.8774 |
| 0.4413        | 0.6572 | 2500 | 0.4481          | 0.8281   | 0.7660    | 0.9011 | 0.8041   | 0.8832 | 0.8773 |
| 0.4604        | 0.7886 | 3000 | 0.5288          | 0.7311   | 0.8553    | 0.6384 | 0.7541   | 0.8779 | 0.8693 |
| 0.4502        | 0.9201 | 3500 | 0.4725          | 0.8135   | 0.8088    | 0.8182 | 0.8036   | 0.8863 | 0.8821 |
| 0.4247        | 1.0515 | 4000 | 0.5722          | 0.8087   | 0.8332    | 0.7855 | 0.8054   | 0.8890 | 0.8837 |
| 0.3702        | 1.1830 | 4500 | 0.4329          | 0.8254   | 0.8064    | 0.8453 | 0.8128   | 0.8834 | 0.8701 |
| 0.3697        | 1.3144 | 5000 | 0.4446          | 0.8119   | 0.8121    | 0.8117 | 0.8031   | 0.8875 | 0.8848 |


### Framework versions

- Transformers 4.42.3
- Pytorch 2.3.0+cu121
- Datasets 2.18.0
- Tokenizers 0.19.0