Hugging Face's logo

arieg
/
bw_spec_cls_4_01_noise_200_confirm

Image Classification
Transformers
TensorFlow
vit
generated_from_keras_callback
Inference Endpoints
Model card Files Community
arieg commited on
Commit
97c5f6f
·
1 Parent(s): f09a937

End of training

Browse files
Files changed (2) hide show
  1. README.md +26 -26
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -15,10 +15,10 @@ probably proofread and complete it, then remove this comment. -->
15
 
16
  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.
17
  It achieves the following results on the evaluation set:
18
- - Train Loss: 0.0149
19
- - Train Sparse Categorical Accuracy: 1.0
20
- - Validation Loss: 0.0144
21
- - Validation Sparse Categorical Accuracy: 1.0
22
  - Epoch: 19
23
 
24
  ## Model description
@@ -43,28 +43,28 @@ The following hyperparameters were used during training:
43
 
44
  ### Training results
45
 
46
- | Train Loss | Train Sparse Categorical Accuracy | Validation Loss | Validation Sparse Categorical Accuracy | Epoch |
47
- |:----------:|:---------------------------------:|:---------------:|:--------------------------------------:|:-----:|
48
- | 0.6427 | 0.9556 | 0.2343 | 1.0 | 0 |
49
- | 0.1648 | 1.0 | 0.1189 | 1.0 | 1 |
50
- | 0.1027 | 1.0 | 0.0884 | 1.0 | 2 |
51
- | 0.0804 | 1.0 | 0.0728 | 1.0 | 3 |
52
- | 0.0674 | 1.0 | 0.0622 | 1.0 | 4 |
53
- | 0.0583 | 1.0 | 0.0545 | 1.0 | 5 |
54
- | 0.0512 | 1.0 | 0.0480 | 1.0 | 6 |
55
- | 0.0453 | 1.0 | 0.0426 | 1.0 | 7 |
56
- | 0.0403 | 1.0 | 0.0381 | 1.0 | 8 |
57
- | 0.0361 | 1.0 | 0.0343 | 1.0 | 9 |
58
- | 0.0326 | 1.0 | 0.0309 | 1.0 | 10 |
59
- | 0.0295 | 1.0 | 0.0281 | 1.0 | 11 |
60
- | 0.0267 | 1.0 | 0.0255 | 1.0 | 12 |
61
- | 0.0244 | 1.0 | 0.0233 | 1.0 | 13 |
62
- | 0.0223 | 1.0 | 0.0214 | 1.0 | 14 |
63
- | 0.0205 | 1.0 | 0.0196 | 1.0 | 15 |
64
- | 0.0188 | 1.0 | 0.0181 | 1.0 | 16 |
65
- | 0.0174 | 1.0 | 0.0167 | 1.0 | 17 |
66
- | 0.0161 | 1.0 | 0.0155 | 1.0 | 18 |
67
- | 0.0149 | 1.0 | 0.0144 | 1.0 | 19 |
68
 
69
 
70
  ### Framework versions
 
15
 
16
  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.
17
  It achieves the following results on the evaluation set:
18
+ - Train Loss: 0.0141
19
+ - Train Categorical Accuracy: 0.2486
20
+ - Validation Loss: 0.0136
21
+ - Validation Categorical Accuracy: 0.2625
22
  - Epoch: 19
23
 
24
  ## Model description
 
43
 
44
  ### Training results
45
 
46
+ | Train Loss | Train Categorical Accuracy | Validation Loss | Validation Categorical Accuracy | Epoch |
47
+ |:----------:|:--------------------------:|:---------------:|:-------------------------------:|:-----:|
48
+ | 0.5757 | 0.2444 | 0.2065 | 0.2625 | 0 |
49
+ | 0.1471 | 0.2486 | 0.1088 | 0.2625 | 1 |
50
+ | 0.0944 | 0.2486 | 0.0818 | 0.2625 | 2 |
51
+ | 0.0748 | 0.2486 | 0.0681 | 0.2625 | 3 |
52
+ | 0.0630 | 0.2486 | 0.0582 | 0.2625 | 4 |
53
+ | 0.0545 | 0.2486 | 0.0508 | 0.2625 | 5 |
54
+ | 0.0478 | 0.2486 | 0.0449 | 0.2625 | 6 |
55
+ | 0.0423 | 0.2486 | 0.0399 | 0.2625 | 7 |
56
+ | 0.0377 | 0.2486 | 0.0356 | 0.2625 | 8 |
57
+ | 0.0339 | 0.2486 | 0.0321 | 0.2625 | 9 |
58
+ | 0.0305 | 0.2486 | 0.0290 | 0.2625 | 10 |
59
+ | 0.0276 | 0.2486 | 0.0263 | 0.2625 | 11 |
60
+ | 0.0251 | 0.2486 | 0.0239 | 0.2625 | 12 |
61
+ | 0.0229 | 0.2486 | 0.0219 | 0.2625 | 13 |
62
+ | 0.0209 | 0.2486 | 0.0200 | 0.2625 | 14 |
63
+ | 0.0192 | 0.2486 | 0.0184 | 0.2625 | 15 |
64
+ | 0.0177 | 0.2486 | 0.0170 | 0.2625 | 16 |
65
+ | 0.0164 | 0.2486 | 0.0157 | 0.2625 | 17 |
66
+ | 0.0151 | 0.2486 | 0.0146 | 0.2625 | 18 |
67
+ | 0.0141 | 0.2486 | 0.0136 | 0.2625 | 19 |
68
 
69
 
70
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f1f05a431386251543df9d3423348c81ceb397567c3717c03c3acfa96fa22733
3
  size 343492280
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ed3cd3e9ce33c40a91bb4d1f963545e333710a9c888603458869572eb51bf4a
3
  size 343492280