File size: 2,323 Bytes
ac39db5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
base_model: AIRI-Institute/gena-lm-bert-base-lastln-t2t
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: gut_1024-finetuned-lora-bert-base-lastln-t2t
  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. -->

# gut_1024-finetuned-lora-bert-base-lastln-t2t

This model is a fine-tuned version of [AIRI-Institute/gena-lm-bert-base-lastln-t2t](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-lastln-t2t) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4049
- F1: 0.8657
- Mcc Score: 0.6421
- Accuracy: 0.8290

## 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: 8
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 1000

### Training results

| Training Loss | Epoch | Step | Validation Loss | F1     | Mcc Score | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:--------:|
| 0.6139        | 0.02  | 100  | 0.5419          | 0.7847 | 0.4843    | 0.7487   |
| 0.5575        | 0.04  | 200  | 0.5407          | 0.7875 | 0.4941    | 0.7530   |
| 0.5339        | 0.05  | 300  | 0.5027          | 0.8347 | 0.5669    | 0.7943   |
| 0.4807        | 0.07  | 400  | 0.4584          | 0.8551 | 0.6032    | 0.8053   |
| 0.5071        | 0.09  | 500  | 0.4408          | 0.8583 | 0.6192    | 0.8180   |
| 0.4011        | 0.11  | 600  | 0.4346          | 0.8576 | 0.6127    | 0.8129   |
| 0.4133        | 0.12  | 700  | 0.4694          | 0.8461 | 0.6007    | 0.8100   |
| 0.424         | 0.14  | 800  | 0.4073          | 0.8651 | 0.6357    | 0.8239   |
| 0.4084        | 0.16  | 900  | 0.3974          | 0.8677 | 0.6429    | 0.8264   |
| 0.4427        | 0.18  | 1000 | 0.4049          | 0.8657 | 0.6421    | 0.8290   |


### Framework versions

- Transformers 4.38.1
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2