File size: 3,358 Bytes
61959c0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
110a283
 
 
 
 
 
 
61959c0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
110a283
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61959c0
 
 
 
 
 
 
 
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
---
base_model: AIRI-Institute/gena-lm-bigbird-base-t2t
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: gena-lm-bigbird-base-t2t_ft_BioS45_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. -->

# gena-lm-bigbird-base-t2t_ft_BioS45_1kbpHG19_DHSs_H3K27AC

This model is a fine-tuned version of [AIRI-Institute/gena-lm-bigbird-base-t2t](https://huggingface.co/AIRI-Institute/gena-lm-bigbird-base-t2t) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6943
- F1 Score: 0.8574
- Precision: 0.8076
- Recall: 0.9137
- Accuracy: 0.8414
- Auc: 0.9156
- Prc: 0.9036

## 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    |
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|:------:|:------:|
| 0.5078        | 0.2103 | 500  | 0.4303          | 0.8087   | 0.8369    | 0.7823 | 0.8069   | 0.8886 | 0.8848 |
| 0.4238        | 0.4207 | 1000 | 0.3944          | 0.8478   | 0.8317    | 0.8645 | 0.8380   | 0.9106 | 0.9046 |
| 0.4246        | 0.6310 | 1500 | 0.3930          | 0.8497   | 0.7954    | 0.9121 | 0.8317   | 0.9087 | 0.9014 |
| 0.4045        | 0.8414 | 2000 | 0.3785          | 0.8587   | 0.7948    | 0.9339 | 0.8397   | 0.9126 | 0.9046 |
| 0.3868        | 1.0517 | 2500 | 0.4130          | 0.8458   | 0.8401    | 0.8516 | 0.8380   | 0.9198 | 0.9137 |
| 0.3662        | 1.2621 | 3000 | 0.4857          | 0.8570   | 0.7907    | 0.9355 | 0.8372   | 0.9159 | 0.9076 |
| 0.378         | 1.4724 | 3500 | 0.4466          | 0.8402   | 0.8657    | 0.8161 | 0.8380   | 0.9203 | 0.9092 |
| 0.3633        | 1.6828 | 4000 | 0.4554          | 0.8237   | 0.8821    | 0.7726 | 0.8275   | 0.9226 | 0.9151 |
| 0.3753        | 1.8931 | 4500 | 0.4109          | 0.8610   | 0.8135    | 0.9145 | 0.8460   | 0.9200 | 0.9117 |
| 0.3249        | 2.1035 | 5000 | 0.5210          | 0.8668   | 0.8292    | 0.9081 | 0.8544   | 0.9203 | 0.9094 |
| 0.3429        | 2.3138 | 5500 | 0.5609          | 0.8602   | 0.8363    | 0.8855 | 0.8498   | 0.9185 | 0.9102 |
| 0.3314        | 2.5242 | 6000 | 0.5652          | 0.8598   | 0.7925    | 0.9395 | 0.8401   | 0.9180 | 0.9084 |
| 0.3462        | 2.7345 | 6500 | 0.6094          | 0.8619   | 0.8138    | 0.9161 | 0.8469   | 0.9120 | 0.8976 |
| 0.3412        | 2.9449 | 7000 | 0.5750          | 0.8508   | 0.8021    | 0.9056 | 0.8342   | 0.9108 | 0.8990 |
| 0.2912        | 3.1552 | 7500 | 0.6943          | 0.8574   | 0.8076    | 0.9137 | 0.8414   | 0.9156 | 0.9036 |


### Framework versions

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