Hugging Face's logo

Hemg
/
Wound-Image-classification

Image Classification
Transformers
TensorBoard
Safetensors
vit
Generated from Trainer
Model card Files
xet
Metrics Community
Hemg commited on
Commit
25e4974
·
verified ·
1 Parent(s): 3c59fc3

Model save

Browse files
Files changed (2) hide show
  1. README.md +37 -47
  2. model.safetensors +1 -1
README.md CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.5860
21
- - Accuracy: 0.8851
22
 
23
  ## Model description
24
 
@@ -37,61 +37,51 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - learning_rate: 0.0002
41
- - train_batch_size: 4
42
- - eval_batch_size: 1
43
  - seed: 42
44
  - gradient_accumulation_steps: 4
45
- - total_train_batch_size: 16
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_ratio: 0.1
49
- - num_epochs: 40
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
54
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
55
- | 1.9234 | 0.98 | 21 | 1.7435 | 0.4943 |
56
- | 1.443 | 2.0 | 43 | 1.2346 | 0.7356 |
57
- | 0.9629 | 2.98 | 64 | 0.7876 | 0.8276 |
58
- | 0.5968 | 4.0 | 86 | 0.8439 | 0.6897 |
59
- | 0.4108 | 4.98 | 107 | 0.6555 | 0.7931 |
60
- | 0.3364 | 6.0 | 129 | 0.7266 | 0.7586 |
61
- | 0.3131 | 6.98 | 150 | 0.8055 | 0.7356 |
62
- | 0.2933 | 8.0 | 172 | 0.5411 | 0.8391 |
63
- | 0.3217 | 8.98 | 193 | 0.4768 | 0.8621 |
64
- | 0.2725 | 10.0 | 215 | 0.7087 | 0.7931 |
65
- | 0.293 | 10.98 | 236 | 0.7644 | 0.7356 |
66
- | 0.1382 | 12.0 | 258 | 0.6590 | 0.8046 |
67
- | 0.1442 | 12.98 | 279 | 0.5936 | 0.8161 |
68
- | 0.1446 | 14.0 | 301 | 0.6793 | 0.8046 |
69
- | 0.2143 | 14.98 | 322 | 0.7464 | 0.8046 |
70
- | 0.1917 | 16.0 | 344 | 0.5202 | 0.8391 |
71
- | 0.1257 | 16.98 | 365 | 0.8504 | 0.7701 |
72
- | 0.1281 | 18.0 | 387 | 0.6993 | 0.8161 |
73
- | 0.1363 | 18.98 | 408 | 0.4296 | 0.8736 |
74
- | 0.1394 | 20.0 | 430 | 0.5792 | 0.8161 |
75
- | 0.1044 | 20.98 | 451 | 0.5892 | 0.8391 |
76
- | 0.1179 | 22.0 | 473 | 0.4079 | 0.8621 |
77
- | 0.0645 | 22.98 | 494 | 0.6288 | 0.8161 |
78
- | 0.132 | 24.0 | 516 | 0.5857 | 0.8276 |
79
- | 0.1273 | 24.98 | 537 | 0.5335 | 0.8391 |
80
- | 0.0695 | 26.0 | 559 | 0.5228 | 0.8621 |
81
- | 0.0867 | 26.98 | 580 | 0.5131 | 0.8391 |
82
- | 0.0516 | 28.0 | 602 | 0.4848 | 0.8506 |
83
- | 0.0854 | 28.98 | 623 | 0.3647 | 0.9080 |
84
- | 0.0376 | 30.0 | 645 | 0.5352 | 0.8506 |
85
- | 0.0601 | 30.98 | 666 | 0.6861 | 0.7931 |
86
- | 0.0419 | 32.0 | 688 | 0.5462 | 0.8621 |
87
- | 0.0529 | 32.98 | 709 | 0.3812 | 0.8966 |
88
- | 0.0324 | 34.0 | 731 | 0.4551 | 0.8851 |
89
- | 0.0241 | 34.98 | 752 | 0.4422 | 0.9080 |
90
- | 0.0287 | 36.0 | 774 | 0.4011 | 0.8851 |
91
- | 0.044 | 36.98 | 795 | 0.3596 | 0.9195 |
92
- | 0.0657 | 38.0 | 817 | 0.4139 | 0.8851 |
93
- | 0.0405 | 38.98 | 838 | 0.4047 | 0.9080 |
94
- | 0.0365 | 39.07 | 840 | 0.5860 | 0.8851 |
95
 
96
 
97
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.4094
21
+ - Accuracy: 0.8621
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - learning_rate: 0.0003
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 8
43
  - seed: 42
44
  - gradient_accumulation_steps: 4
45
+ - total_train_batch_size: 32
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_ratio: 0.1
49
+ - num_epochs: 32
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
54
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
55
+ | 2.0067 | 0.93 | 10 | 1.7096 | 0.4713 |
56
+ | 1.3813 | 1.95 | 21 | 1.1621 | 0.7126 |
57
+ | 0.8497 | 2.98 | 32 | 0.7038 | 0.8161 |
58
+ | 0.548 | 4.0 | 43 | 0.6777 | 0.7931 |
59
+ | 0.4304 | 4.93 | 53 | 0.6226 | 0.7931 |
60
+ | 0.3439 | 5.95 | 64 | 0.7137 | 0.7816 |
61
+ | 0.3283 | 6.98 | 75 | 0.6002 | 0.8161 |
62
+ | 0.2221 | 8.0 | 86 | 0.5427 | 0.8391 |
63
+ | 0.3835 | 8.93 | 96 | 0.5764 | 0.7931 |
64
+ | 0.2604 | 9.95 | 107 | 0.7187 | 0.7126 |
65
+ | 0.2842 | 10.98 | 118 | 0.6400 | 0.8046 |
66
+ | 0.1638 | 12.0 | 129 | 0.4891 | 0.8391 |
67
+ | 0.2011 | 12.93 | 139 | 0.6157 | 0.8161 |
68
+ | 0.1667 | 13.95 | 150 | 0.5799 | 0.8046 |
69
+ | 0.1711 | 14.98 | 161 | 0.7119 | 0.7701 |
70
+ | 0.179 | 16.0 | 172 | 0.7303 | 0.7701 |
71
+ | 0.1221 | 16.93 | 182 | 0.6209 | 0.8161 |
72
+ | 0.1055 | 17.95 | 193 | 0.6667 | 0.8046 |
73
+ | 0.1023 | 18.98 | 204 | 0.3990 | 0.8736 |
74
+ | 0.1546 | 20.0 | 215 | 0.4699 | 0.8506 |
75
+ | 0.0895 | 20.93 | 225 | 0.4024 | 0.8966 |
76
+ | 0.0815 | 21.95 | 236 | 0.4166 | 0.8736 |
77
+ | 0.0691 | 22.98 | 247 | 0.5295 | 0.8621 |
78
+ | 0.0821 | 24.0 | 258 | 0.5359 | 0.8391 |
79
+ | 0.1286 | 24.93 | 268 | 0.4657 | 0.8851 |
80
+ | 0.0523 | 25.95 | 279 | 0.4145 | 0.8621 |
81
+ | 0.0711 | 26.98 | 290 | 0.3788 | 0.8966 |
82
+ | 0.0546 | 28.0 | 301 | 0.3526 | 0.9080 |
83
+ | 0.0846 | 28.93 | 311 | 0.2510 | 0.9425 |
84
+ | 0.0503 | 29.77 | 320 | 0.4094 | 0.8621 |
 
 
 
 
 
 
 
 
 
 
85
 
86
 
87
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b709fe12d8713a38a9f6caa568cd3f6ad6f23ce00caea6ee6e63f9e487bb0f9e
3
  size 343239356
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:429cc67ebc75215add62fd24e06a6cab72bb624e0c3845cc3f17e5d5f25e2055
3
  size 343239356