AykeeSalazar commited on
Commit
b03019b
1 Parent(s): b636a53

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +107 -0
README.md ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - imagefolder
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: vc-bantai-vit-withoutAMBI-adunest
11
+ results:
12
+ - task:
13
+ name: Image Classification
14
+ type: image-classification
15
+ dataset:
16
+ name: imagefolder
17
+ type: imagefolder
18
+ args: Violation-Classification---Raw-6
19
+ metrics:
20
+ - name: Accuracy
21
+ type: accuracy
22
+ value: 0.9388646288209607
23
+ ---
24
+
25
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
26
+ should probably proofread and complete it, then remove this comment. -->
27
+
28
+ # vc-bantai-vit-withoutAMBI-adunest
29
+
30
+ 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 imagefolder dataset.
31
+ It achieves the following results on the evaluation set:
32
+ - Loss: 0.1950
33
+ - Accuracy: 0.9389
34
+
35
+ ## Model description
36
+
37
+ More information needed
38
+
39
+ ## Intended uses & limitations
40
+
41
+ More information needed
42
+
43
+ ## Training and evaluation data
44
+
45
+ More information needed
46
+
47
+ ## Training procedure
48
+
49
+ ### Training hyperparameters
50
+
51
+ The following hyperparameters were used during training:
52
+ - learning_rate: 0.0005
53
+ - train_batch_size: 16
54
+ - eval_batch_size: 16
55
+ - seed: 42
56
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
57
+ - lr_scheduler_type: linear
58
+ - num_epochs: 4
59
+ - mixed_precision_training: Native AMP
60
+
61
+ ### Training results
62
+
63
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
64
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
65
+ | 0.4821 | 0.11 | 100 | 0.7644 | 0.6714 |
66
+ | 0.7032 | 0.23 | 200 | 0.5568 | 0.75 |
67
+ | 0.5262 | 0.34 | 300 | 0.4440 | 0.7806 |
68
+ | 0.4719 | 0.45 | 400 | 0.3893 | 0.8144 |
69
+ | 0.5021 | 0.57 | 500 | 0.5129 | 0.8090 |
70
+ | 0.3123 | 0.68 | 600 | 0.4536 | 0.7980 |
71
+ | 0.3606 | 0.79 | 700 | 0.3679 | 0.8483 |
72
+ | 0.4081 | 0.91 | 800 | 0.3335 | 0.8559 |
73
+ | 0.3624 | 1.02 | 900 | 0.3149 | 0.8592 |
74
+ | 0.1903 | 1.14 | 1000 | 0.3296 | 0.8766 |
75
+ | 0.334 | 1.25 | 1100 | 0.2832 | 0.8897 |
76
+ | 0.2731 | 1.36 | 1200 | 0.2546 | 0.8930 |
77
+ | 0.311 | 1.48 | 1300 | 0.2585 | 0.8908 |
78
+ | 0.3209 | 1.59 | 1400 | 0.2701 | 0.8854 |
79
+ | 0.4005 | 1.7 | 1500 | 0.2643 | 0.8897 |
80
+ | 0.3128 | 1.82 | 1600 | 0.2864 | 0.8843 |
81
+ | 0.3376 | 1.93 | 1700 | 0.2882 | 0.8657 |
82
+ | 0.2698 | 2.04 | 1800 | 0.2876 | 0.9028 |
83
+ | 0.2347 | 2.16 | 1900 | 0.2405 | 0.8974 |
84
+ | 0.2436 | 2.27 | 2000 | 0.2804 | 0.8886 |
85
+ | 0.1764 | 2.38 | 2100 | 0.2852 | 0.8952 |
86
+ | 0.1197 | 2.5 | 2200 | 0.2312 | 0.9127 |
87
+ | 0.1082 | 2.61 | 2300 | 0.2133 | 0.9116 |
88
+ | 0.1245 | 2.72 | 2400 | 0.2677 | 0.8985 |
89
+ | 0.1335 | 2.84 | 2500 | 0.2098 | 0.9181 |
90
+ | 0.2194 | 2.95 | 2600 | 0.1911 | 0.9127 |
91
+ | 0.089 | 3.06 | 2700 | 0.2062 | 0.9181 |
92
+ | 0.0465 | 3.18 | 2800 | 0.2414 | 0.9247 |
93
+ | 0.0985 | 3.29 | 2900 | 0.1869 | 0.9389 |
94
+ | 0.1113 | 3.41 | 3000 | 0.1819 | 0.9323 |
95
+ | 0.1392 | 3.52 | 3100 | 0.2101 | 0.9312 |
96
+ | 0.0621 | 3.63 | 3200 | 0.2201 | 0.9367 |
97
+ | 0.1168 | 3.75 | 3300 | 0.1935 | 0.9389 |
98
+ | 0.059 | 3.86 | 3400 | 0.1946 | 0.9367 |
99
+ | 0.0513 | 3.97 | 3500 | 0.1950 | 0.9389 |
100
+
101
+
102
+ ### Framework versions
103
+
104
+ - Transformers 4.20.1
105
+ - Pytorch 1.12.0+cu113
106
+ - Datasets 2.3.2
107
+ - Tokenizers 0.12.1