File size: 2,877 Bytes
46d882e
 
da03fe3
46d882e
 
 
 
ebf8d3a
 
46d882e
 
ebf8d3a
 
 
 
 
91dc10c
ebf8d3a
 
 
 
 
 
 
2bf220f
46d882e
 
 
 
 
 
 
91dc10c
ebf8d3a
2bf220f
 
46d882e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2fe4aa0
412612a
46d882e
 
 
da03fe3
 
2bf220f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
46d882e
 
 
 
412612a
 
 
 
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
---
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- generated_from_trainer
datasets:
- renovation
metrics:
- accuracy
model-index:
- name: vit-base-renovation
  results:
  - task:
      name: Image Classification
      type: image-classification
    dataset:
      name: renovation
      type: renovation
      config: default
      split: validation
      args: default
    metrics:
    - name: Accuracy
      type: accuracy
      value: 0.6666666666666666
---

<!-- 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. -->

# vit-base-renovation

This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co./google/vit-base-patch16-224-in21k) on the renovation dataset.
It achieves the following results on the evaluation set:
- Loss: 1.4621
- Accuracy: 0.6667

## 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 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.382         | 0.2   | 25   | 1.1103          | 0.6073   |
| 0.5741        | 0.4   | 50   | 1.0628          | 0.6210   |
| 0.5589        | 0.6   | 75   | 1.0025          | 0.6667   |
| 0.4074        | 0.81  | 100  | 1.1324          | 0.6073   |
| 0.3581        | 1.01  | 125  | 1.1935          | 0.6438   |
| 0.2618        | 1.21  | 150  | 1.8300          | 0.5023   |
| 0.1299        | 1.41  | 175  | 1.2577          | 0.6301   |
| 0.2562        | 1.61  | 200  | 1.0924          | 0.6895   |
| 0.2573        | 1.81  | 225  | 1.1285          | 0.6849   |
| 0.2471        | 2.02  | 250  | 1.3387          | 0.6256   |
| 0.0618        | 2.22  | 275  | 1.2246          | 0.6667   |
| 0.0658        | 2.42  | 300  | 1.4132          | 0.6347   |
| 0.0592        | 2.62  | 325  | 1.4326          | 0.6530   |
| 0.0464        | 2.82  | 350  | 1.2484          | 0.6849   |
| 0.0567        | 3.02  | 375  | 1.5350          | 0.6347   |
| 0.0269        | 3.23  | 400  | 1.4797          | 0.6667   |
| 0.0239        | 3.43  | 425  | 1.4444          | 0.6530   |
| 0.0184        | 3.63  | 450  | 1.4474          | 0.6575   |
| 0.0286        | 3.83  | 475  | 1.4621          | 0.6667   |


### Framework versions

- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2