File size: 4,879 Bytes
f9353a8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
---
license: apache-2.0
base_model: microsoft/beit-large-patch16-224
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: hushem_40x_beit_large_adamax_001_fold3
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      config: default
      split: test
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.7674418604651163
---

<!-- 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. -->

# hushem_40x_beit_large_adamax_001_fold3

This model is a fine-tuned version of [microsoft/beit-large-patch16-224](https://huggingface.co/microsoft/beit-large-patch16-224) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 3.1862
- Accuracy: 0.7674

## 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: 0.001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 0.3341        | 1.0   | 217   | 1.0412          | 0.6512   |
| 0.1567        | 2.0   | 434   | 0.8679          | 0.7907   |
| 0.1261        | 3.0   | 651   | 0.9519          | 0.7907   |
| 0.0451        | 4.0   | 868   | 1.0258          | 0.7674   |
| 0.0494        | 5.0   | 1085  | 1.2986          | 0.8140   |
| 0.0356        | 6.0   | 1302  | 1.5169          | 0.7442   |
| 0.0142        | 7.0   | 1519  | 1.4785          | 0.7907   |
| 0.048         | 8.0   | 1736  | 1.4460          | 0.8140   |
| 0.0363        | 9.0   | 1953  | 1.0943          | 0.7674   |
| 0.0409        | 10.0  | 2170  | 1.6345          | 0.7907   |
| 0.0021        | 11.0  | 2387  | 1.2558          | 0.8140   |
| 0.0072        | 12.0  | 2604  | 1.1994          | 0.8372   |
| 0.0193        | 13.0  | 2821  | 1.2732          | 0.8372   |
| 0.0006        | 14.0  | 3038  | 1.5708          | 0.7674   |
| 0.0013        | 15.0  | 3255  | 1.1380          | 0.8837   |
| 0.0001        | 16.0  | 3472  | 1.3578          | 0.8837   |
| 0.0           | 17.0  | 3689  | 1.3940          | 0.8837   |
| 0.0           | 18.0  | 3906  | 1.4630          | 0.8605   |
| 0.0           | 19.0  | 4123  | 1.4804          | 0.8140   |
| 0.0           | 20.0  | 4340  | 1.5039          | 0.8372   |
| 0.0           | 21.0  | 4557  | 1.5153          | 0.8605   |
| 0.0           | 22.0  | 4774  | 1.6110          | 0.8372   |
| 0.0           | 23.0  | 4991  | 1.6351          | 0.8372   |
| 0.0           | 24.0  | 5208  | 1.6586          | 0.8372   |
| 0.0           | 25.0  | 5425  | 1.6837          | 0.8605   |
| 0.0           | 26.0  | 5642  | 2.1644          | 0.8140   |
| 0.0           | 27.0  | 5859  | 1.8231          | 0.8372   |
| 0.0           | 28.0  | 6076  | 1.8592          | 0.8837   |
| 0.0           | 29.0  | 6293  | 2.3766          | 0.7907   |
| 0.0004        | 30.0  | 6510  | 2.2802          | 0.7674   |
| 0.0           | 31.0  | 6727  | 2.0919          | 0.7907   |
| 0.0           | 32.0  | 6944  | 2.0989          | 0.7907   |
| 0.0           | 33.0  | 7161  | 2.1214          | 0.7907   |
| 0.0           | 34.0  | 7378  | 2.1583          | 0.7907   |
| 0.0           | 35.0  | 7595  | 2.1876          | 0.7907   |
| 0.0           | 36.0  | 7812  | 2.1795          | 0.7907   |
| 0.0007        | 37.0  | 8029  | 3.1536          | 0.7674   |
| 0.0           | 38.0  | 8246  | 3.0845          | 0.7674   |
| 0.0           | 39.0  | 8463  | 2.9748          | 0.7907   |
| 0.0           | 40.0  | 8680  | 2.9984          | 0.7907   |
| 0.0           | 41.0  | 8897  | 3.0029          | 0.7907   |
| 0.0           | 42.0  | 9114  | 3.0143          | 0.7907   |
| 0.0           | 43.0  | 9331  | 3.0354          | 0.7907   |
| 0.0           | 44.0  | 9548  | 3.0480          | 0.7907   |
| 0.0           | 45.0  | 9765  | 3.0564          | 0.7907   |
| 0.0           | 46.0  | 9982  | 3.1685          | 0.7674   |
| 0.0           | 47.0  | 10199 | 3.1763          | 0.7674   |
| 0.0           | 48.0  | 10416 | 3.1810          | 0.7674   |
| 0.0           | 49.0  | 10633 | 3.1846          | 0.7674   |
| 0.0           | 50.0  | 10850 | 3.1862          | 0.7674   |


### Framework versions

- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2