Hugging Face's logo

NabeelShar
/
emotions_classifier

Image Classification
Transformers
TensorFlow
vit
generated_from_keras_callback
Inference Endpoints
Model card Files Community
NabeelShar commited on
Commit
0d4fe32
•
1 Parent(s): de55721

End of training

Browse files
Files changed (2) hide show
  1. README.md +10 -55
  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: 1.1146
19
- - Validation Loss: 1.6637
20
- - Train Accuracy: 0.3625
21
- - Epoch: 49
22
 
23
  ## Model description
24
 
@@ -37,63 +37,18 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0003, 'decay_steps': 32000, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
41
  - training_precision: float32
42
 
43
  ### Training results
44
 
45
  | Train Loss | Validation Loss | Train Accuracy | Epoch |
46
  |:----------:|:---------------:|:--------------:|:-----:|
47
- | 2.0910 | 2.0947 | 0.1062 | 0 |
48
- | 2.0049 | 1.9103 | 0.2062 | 1 |
49
- | 2.0473 | 1.9654 | 0.175 | 2 |
50
- | 1.9824 | 2.1773 | 0.125 | 3 |
51
- | 2.0538 | 2.0144 | 0.1875 | 4 |
52
- | 1.9921 | 2.0826 | 0.1437 | 5 |
53
- | 2.0904 | 2.0995 | 0.1812 | 6 |
54
- | 2.0866 | 2.0908 | 0.1313 | 7 |
55
- | 2.0718 | 2.0800 | 0.125 | 8 |
56
- | 2.0511 | 2.0358 | 0.1938 | 9 |
57
- | 1.9794 | 1.9049 | 0.2313 | 10 |
58
- | 1.9289 | 1.8717 | 0.1875 | 11 |
59
- | 1.8696 | 1.8451 | 0.2062 | 12 |
60
- | 1.8361 | 1.8010 | 0.2062 | 13 |
61
- | 1.8122 | 1.7457 | 0.225 | 14 |
62
- | 1.7571 | 1.7331 | 0.25 | 15 |
63
- | 1.6846 | 1.8783 | 0.25 | 16 |
64
- | 1.6954 | 1.8015 | 0.25 | 17 |
65
- | 1.7414 | 1.7329 | 0.1625 | 18 |
66
- | 1.6662 | 1.6900 | 0.2625 | 19 |
67
- | 1.6322 | 1.7607 | 0.25 | 20 |
68
- | 1.5822 | 1.6670 | 0.3063 | 21 |
69
- | 1.6279 | 1.6800 | 0.3 | 22 |
70
- | 1.5737 | 1.7843 | 0.25 | 23 |
71
- | 1.5851 | 1.6927 | 0.2875 | 24 |
72
- | 1.4926 | 1.6640 | 0.2687 | 25 |
73
- | 1.4879 | 1.7408 | 0.2812 | 26 |
74
- | 1.5564 | 1.6668 | 0.275 | 27 |
75
- | 1.5093 | 1.6259 | 0.3187 | 28 |
76
- | 1.4428 | 1.6973 | 0.2437 | 29 |
77
- | 1.4328 | 1.6412 | 0.2812 | 30 |
78
- | 1.3778 | 1.6470 | 0.3187 | 31 |
79
- | 1.4635 | 1.6411 | 0.325 | 32 |
80
- | 1.4044 | 1.6643 | 0.2938 | 33 |
81
- | 1.2991 | 1.6864 | 0.2875 | 34 |
82
- | 1.3467 | 1.6124 | 0.2687 | 35 |
83
- | 1.3422 | 1.6517 | 0.2687 | 36 |
84
- | 1.3998 | 1.5634 | 0.325 | 37 |
85
- | 1.2963 | 1.7403 | 0.3 | 38 |
86
- | 1.3050 | 1.7550 | 0.3187 | 39 |
87
- | 1.2988 | 1.6917 | 0.3438 | 40 |
88
- | 1.2601 | 1.6739 | 0.3125 | 41 |
89
- | 1.1943 | 1.7200 | 0.35 | 42 |
90
- | 1.2663 | 1.6505 | 0.3312 | 43 |
91
- | 1.2228 | 1.7337 | 0.3312 | 44 |
92
- | 1.1413 | 1.7777 | 0.2812 | 45 |
93
- | 1.1429 | 1.7138 | 0.3375 | 46 |
94
- | 1.0760 | 1.7160 | 0.3187 | 47 |
95
- | 1.1625 | 1.8049 | 0.3063 | 48 |
96
- | 1.1146 | 1.6637 | 0.3625 | 49 |
97
 
98
 
99
  ### 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: 0.1867
19
+ - Validation Loss: 2.0372
20
+ - Train Accuracy: 0.4437
21
+ - Epoch: 4
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0003, 'decay_steps': 64000, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.02}
41
  - training_precision: float32
42
 
43
  ### Training results
44
 
45
  | Train Loss | Validation Loss | Train Accuracy | Epoch |
46
  |:----------:|:---------------:|:--------------:|:-----:|
47
+ | 0.2056 | 2.1272 | 0.425 | 0 |
48
+ | 0.2375 | 1.7537 | 0.5375 | 1 |
49
+ | 0.1793 | 2.0755 | 0.4562 | 2 |
50
+ | 0.1725 | 2.2938 | 0.4062 | 3 |
51
+ | 0.1867 | 2.0372 | 0.4437 | 4 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52
 
53
 
54
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:677efa1f1cbc418deff950dc354cfa82594a383fd176d49b66f500717ab1607b
3
  size 343504568
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29dee88dec820d7360044896c89eb9a38f688522c097f5f3aaa506d0bcc0b0d2
3
  size 343504568