Hugging Face's logo

ongkn
/
attraction-classifier

Image Classification
Transformers
Safetensors
vit
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
5
ongkn commited on
Commit
04de4c3
•
1 Parent(s): 493a5c0

End of training

Browse files
Files changed (2) hide show
  1. README.md +17 -19
  2. model.safetensors +1 -1
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.8158995815899581
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  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.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.4121
36
- - Accuracy: 0.8159
37
 
38
  ## Model description
39
 
@@ -56,8 +56,8 @@ The following hyperparameters were used during training:
56
  - train_batch_size: 32
57
  - eval_batch_size: 32
58
  - seed: 69
59
- - gradient_accumulation_steps: 8
60
- - total_train_batch_size: 256
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: cosine
63
  - lr_scheduler_warmup_ratio: 0.05
@@ -68,20 +68,18 @@ The following hyperparameters were used during training:
68
 
69
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
71
- | 0.6839 | 0.89 | 15 | 0.6438 | 0.6757 |
72
- | 0.5555 | 1.78 | 30 | 0.5198 | 0.7364 |
73
- | 0.4995 | 2.67 | 45 | 0.5212 | 0.7469 |
74
- | 0.4177 | 3.56 | 60 | 0.4447 | 0.7866 |
75
- | 0.415 | 4.44 | 75 | 0.4438 | 0.7929 |
76
- | 0.3737 | 5.33 | 90 | 0.4302 | 0.7866 |
77
- | 0.3588 | 6.22 | 105 | 0.4452 | 0.7992 |
78
- | 0.3343 | 7.11 | 120 | 0.4666 | 0.7908 |
79
- | 0.3095 | 8.0 | 135 | 0.4727 | 0.7720 |
80
- | 0.2951 | 8.89 | 150 | 0.4162 | 0.8138 |
81
- | 0.2819 | 9.78 | 165 | 0.4299 | 0.8159 |
82
- | 0.257 | 10.67 | 180 | 0.4497 | 0.8033 |
83
- | 0.2625 | 11.56 | 195 | 0.4642 | 0.7971 |
84
- | 0.2287 | 12.44 | 210 | 0.4121 | 0.8159 |
85
 
86
 
87
  ### Framework versions
 
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.8242677824267782
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  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.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.4274
36
+ - Accuracy: 0.8243
37
 
38
  ## Model description
39
 
 
56
  - train_batch_size: 32
57
  - eval_batch_size: 32
58
  - seed: 69
59
+ - gradient_accumulation_steps: 16
60
+ - total_train_batch_size: 512
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: cosine
63
  - lr_scheduler_warmup_ratio: 0.05
 
68
 
69
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
71
+ | 0.6782 | 1.78 | 15 | 0.5922 | 0.7008 |
72
+ | 0.5096 | 3.56 | 30 | 0.5153 | 0.7552 |
73
+ | 0.4434 | 5.33 | 45 | 0.4520 | 0.7762 |
74
+ | 0.3844 | 7.11 | 60 | 0.4381 | 0.8013 |
75
+ | 0.3642 | 8.89 | 75 | 0.4359 | 0.8054 |
76
+ | 0.322 | 10.67 | 90 | 0.4086 | 0.8138 |
77
+ | 0.2845 | 12.44 | 105 | 0.4111 | 0.8201 |
78
+ | 0.2588 | 14.22 | 120 | 0.4100 | 0.8159 |
79
+ | 0.2516 | 16.0 | 135 | 0.4122 | 0.8389 |
80
+ | 0.2375 | 17.78 | 150 | 0.4085 | 0.8243 |
81
+ | 0.2309 | 19.56 | 165 | 0.4149 | 0.8117 |
82
+ | 0.2175 | 21.33 | 180 | 0.4274 | 0.8243 |
 
 
83
 
84
 
85
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ea348e133a88767021c155a43c8543ef529a817ed82e304781a98ae3482094da
3
  size 343223968
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4af3317e723148f0af2f8102318ced1bc7e05a9367f11f93c23486795ad0d91c
3
  size 343223968