File size: 2,681 Bytes
9389aab
 
 
 
c2b9437
9389aab
 
 
 
 
 
 
 
 
 
 
 
c2b9437
9389aab
 
 
 
 
 
 
c2b9437
9389aab
 
 
 
 
 
 
c2b9437
9389aab
c2b9437
 
9389aab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: apache-2.0
base_model: dima806/deepfake_vs_real_image_detection
tags:
- image-classification
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: realFake-img
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: ai_real_images
      type: imagefolder
      config: default
      split: train
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.8518181818181818
---

<!-- 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. -->

# realFake-img

This model is a fine-tuned version of [dima806/deepfake_vs_real_image_detection](https://huggingface.co/dima806/deepfake_vs_real_image_detection) on the ai_real_images dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3329
- Accuracy: 0.8518

## 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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
|:-------------:|:------:|:----:|:---------------:|:--------:|
| 0.4892        | 0.2564 | 100  | 0.5756          | 0.7227   |
| 0.683         | 0.5128 | 200  | 0.6742          | 0.6373   |
| 0.3737        | 0.7692 | 300  | 0.5462          | 0.7555   |
| 0.3554        | 1.0256 | 400  | 0.4354          | 0.8009   |
| 0.2368        | 1.2821 | 500  | 0.4046          | 0.8309   |
| 0.3696        | 1.5385 | 600  | 0.5547          | 0.7809   |
| 0.2824        | 1.7949 | 700  | 0.3329          | 0.8518   |
| 0.2366        | 2.0513 | 800  | 0.4582          | 0.8255   |
| 0.2212        | 2.3077 | 900  | 0.4885          | 0.8255   |
| 0.2031        | 2.5641 | 1000 | 0.4282          | 0.8564   |
| 0.1717        | 2.8205 | 1100 | 0.4373          | 0.85     |
| 0.1303        | 3.0769 | 1200 | 0.3659          | 0.8718   |
| 0.0889        | 3.3333 | 1300 | 0.3663          | 0.8736   |
| 0.1157        | 3.5897 | 1400 | 0.4588          | 0.8436   |
| 0.1215        | 3.8462 | 1500 | 0.4350          | 0.8655   |


### Framework versions

- Transformers 4.42.4
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1