File size: 4,421 Bytes
f6c4de8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: apache-2.0
base_model: google/vit-large-patch32-224-in21k
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: Adam_ViTL-32_224-2e-4-batch_16_epoch_4_classes_24
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: imagefolder
      type: imagefolder
      config: default
      split: train
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.9482758620689655
---

<!-- 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. -->

# Adam_ViTL-32_224-2e-4-batch_16_epoch_4_classes_24

This model is a fine-tuned version of [google/vit-large-patch32-224-in21k](https://huggingface.co/google/vit-large-patch32-224-in21k) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2083
- Accuracy: 0.9483

## 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: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.7984        | 0.07  | 100  | 0.8039          | 0.8606   |
| 0.4352        | 0.14  | 200  | 0.5735          | 0.8463   |
| 0.3651        | 0.21  | 300  | 0.3951          | 0.8937   |
| 0.3133        | 0.28  | 400  | 0.4525          | 0.8894   |
| 0.2641        | 0.35  | 500  | 0.3618          | 0.9023   |
| 0.2104        | 0.42  | 600  | 0.4240          | 0.8922   |
| 0.1787        | 0.49  | 700  | 0.4070          | 0.8879   |
| 0.1412        | 0.56  | 800  | 0.3259          | 0.9124   |
| 0.2121        | 0.63  | 900  | 0.3575          | 0.8994   |
| 0.1491        | 0.7   | 1000 | 0.2769          | 0.9152   |
| 0.268         | 0.77  | 1100 | 0.3432          | 0.9195   |
| 0.2378        | 0.84  | 1200 | 0.3622          | 0.9109   |
| 0.0812        | 0.91  | 1300 | 0.2857          | 0.9210   |
| 0.127         | 0.97  | 1400 | 0.2787          | 0.9253   |
| 0.0256        | 1.04  | 1500 | 0.3116          | 0.9267   |
| 0.027         | 1.11  | 1600 | 0.2889          | 0.9282   |
| 0.0508        | 1.18  | 1700 | 0.3048          | 0.9310   |
| 0.0932        | 1.25  | 1800 | 0.2732          | 0.9382   |
| 0.0745        | 1.32  | 1900 | 0.3275          | 0.9195   |
| 0.0675        | 1.39  | 2000 | 0.2505          | 0.9440   |
| 0.0347        | 1.46  | 2100 | 0.2686          | 0.9382   |
| 0.0121        | 1.53  | 2200 | 0.2888          | 0.9454   |
| 0.1104        | 1.6   | 2300 | 0.2375          | 0.9440   |
| 0.0778        | 1.67  | 2400 | 0.2345          | 0.9411   |
| 0.0029        | 1.74  | 2500 | 0.2924          | 0.9282   |
| 0.0063        | 1.81  | 2600 | 0.2867          | 0.9353   |
| 0.0394        | 1.88  | 2700 | 0.3384          | 0.9224   |
| 0.0043        | 1.95  | 2800 | 0.2855          | 0.9195   |
| 0.025         | 2.02  | 2900 | 0.3218          | 0.9296   |
| 0.0096        | 2.09  | 3000 | 0.2810          | 0.9368   |
| 0.0018        | 2.16  | 3100 | 0.1971          | 0.9526   |
| 0.0102        | 2.23  | 3200 | 0.2175          | 0.9497   |
| 0.0016        | 2.3   | 3300 | 0.2341          | 0.9454   |
| 0.0024        | 2.37  | 3400 | 0.2607          | 0.9425   |
| 0.0024        | 2.44  | 3500 | 0.2380          | 0.9440   |
| 0.0019        | 2.51  | 3600 | 0.2422          | 0.9382   |
| 0.0062        | 2.58  | 3700 | 0.2191          | 0.9483   |
| 0.0416        | 2.65  | 3800 | 0.2491          | 0.9483   |
| 0.002         | 2.72  | 3900 | 0.2201          | 0.9497   |
| 0.0013        | 2.79  | 4000 | 0.2242          | 0.9468   |
| 0.0012        | 2.86  | 4100 | 0.2182          | 0.9440   |
| 0.0011        | 2.92  | 4200 | 0.2079          | 0.9497   |
| 0.001         | 2.99  | 4300 | 0.2083          | 0.9483   |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2