Hugging Face's logo

pradanaadn
/
vit-emotional-classifier

Image Classification
Transformers
TensorBoard
Safetensors
vit
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
pradanaadn commited on
Commit
89c9856
•
1 Parent(s): a40e175

End of training

Browse files
Files changed (3) hide show
  1. README.md +48 -89
  2. model.safetensors +1 -1
  3. runs/Jun04_03-23-15_2a967390fece/events.out.tfevents.1717472301.2a967390fece.1464.16 +3 -0
README.md CHANGED
@@ -4,7 +4,7 @@ base_model: google/vit-base-patch16-224-in21k
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
- - FastJobs/Visual_Emotional_Analysis
8
  metrics:
9
  - accuracy
10
  model-index:
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.375
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
30
 
31
  # vit-emotional-classifier
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 [FastJobs/Visual_Emotional_Analysis](https://huggingface.co/datasets/FastJobs/Visual_Emotional_Analysis) the dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.4780
36
- - Accuracy: 0.375
37
 
38
  ## Model description
39
 
@@ -57,94 +57,53 @@ The following hyperparameters were used during training:
57
  - eval_batch_size: 16
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
- - lr_scheduler_type: reduce_lr_on_plateau
61
- - num_epochs: 100
62
- - mixed_precision_training: Native AMP
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
67
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
68
- | 1.7296 | 1.25 | 50 | 1.8725 | 0.275 |
69
- | 1.7073 | 2.5 | 100 | 1.8573 | 0.275 |
70
- | 1.7104 | 3.75 | 150 | 1.8416 | 0.275 |
71
- | 1.7008 | 5.0 | 200 | 1.8270 | 0.2625 |
72
- | 1.6479 | 6.25 | 250 | 1.8165 | 0.25 |
73
- | 1.5797 | 7.5 | 300 | 1.8003 | 0.2687 |
74
- | 1.5891 | 8.75 | 350 | 1.7862 | 0.2938 |
75
- | 1.5674 | 10.0 | 400 | 1.7679 | 0.2812 |
76
- | 1.5281 | 11.25 | 450 | 1.7500 | 0.2938 |
77
- | 1.5248 | 12.5 | 500 | 1.7465 | 0.2938 |
78
- | 1.5182 | 13.75 | 550 | 1.7399 | 0.2938 |
79
- | 1.4921 | 15.0 | 600 | 1.7138 | 0.2938 |
80
- | 1.4743 | 16.25 | 650 | 1.7074 | 0.3187 |
81
- | 1.4422 | 17.5 | 700 | 1.6937 | 0.3312 |
82
- | 1.4113 | 18.75 | 750 | 1.6860 | 0.325 |
83
- | 1.4055 | 20.0 | 800 | 1.6777 | 0.3312 |
84
- | 1.422 | 21.25 | 850 | 1.6581 | 0.3375 |
85
- | 1.3458 | 22.5 | 900 | 1.6507 | 0.35 |
86
- | 1.3649 | 23.75 | 950 | 1.6448 | 0.3375 |
87
- | 1.3444 | 25.0 | 1000 | 1.6350 | 0.3812 |
88
- | 1.2815 | 26.25 | 1050 | 1.6277 | 0.3625 |
89
- | 1.3048 | 27.5 | 1100 | 1.6071 | 0.3625 |
90
- | 1.3027 | 28.75 | 1150 | 1.6028 | 0.3563 |
91
- | 1.207 | 30.0 | 1200 | 1.6021 | 0.3688 |
92
- | 1.2375 | 31.25 | 1250 | 1.5948 | 0.3875 |
93
- | 1.2621 | 32.5 | 1300 | 1.5910 | 0.3875 |
94
- | 1.2179 | 33.75 | 1350 | 1.5882 | 0.3688 |
95
- | 1.1803 | 35.0 | 1400 | 1.5709 | 0.3625 |
96
- | 1.1388 | 36.25 | 1450 | 1.5733 | 0.3875 |
97
- | 1.1098 | 37.5 | 1500 | 1.5591 | 0.3688 |
98
- | 1.1207 | 38.75 | 1550 | 1.5535 | 0.3875 |
99
- | 1.0707 | 40.0 | 1600 | 1.5564 | 0.3812 |
100
- | 1.0868 | 41.25 | 1650 | 1.5487 | 0.375 |
101
- | 1.0571 | 42.5 | 1700 | 1.5392 | 0.3688 |
102
- | 1.0839 | 43.75 | 1750 | 1.5378 | 0.4 |
103
- | 1.0452 | 45.0 | 1800 | 1.5313 | 0.3937 |
104
- | 1.0277 | 46.25 | 1850 | 1.5277 | 0.375 |
105
- | 0.9694 | 47.5 | 1900 | 1.5224 | 0.375 |
106
- | 1.0246 | 48.75 | 1950 | 1.5361 | 0.375 |
107
- | 0.99 | 50.0 | 2000 | 1.5217 | 0.3625 |
108
- | 0.9201 | 51.25 | 2050 | 1.5177 | 0.375 |
109
- | 0.9722 | 52.5 | 2100 | 1.5150 | 0.3688 |
110
- | 0.9347 | 53.75 | 2150 | 1.5103 | 0.4 |
111
- | 0.9169 | 55.0 | 2200 | 1.5067 | 0.3937 |
112
- | 0.8822 | 56.25 | 2250 | 1.4976 | 0.3625 |
113
- | 0.9071 | 57.5 | 2300 | 1.4955 | 0.3937 |
114
- | 0.8527 | 58.75 | 2350 | 1.4948 | 0.3875 |
115
- | 0.8607 | 60.0 | 2400 | 1.4901 | 0.3875 |
116
- | 0.8248 | 61.25 | 2450 | 1.5012 | 0.375 |
117
- | 0.8277 | 62.5 | 2500 | 1.4976 | 0.375 |
118
- | 0.7854 | 63.75 | 2550 | 1.4926 | 0.375 |
119
- | 0.8214 | 65.0 | 2600 | 1.4890 | 0.4 |
120
- | 0.7536 | 66.25 | 2650 | 1.4903 | 0.3812 |
121
- | 0.7268 | 67.5 | 2700 | 1.4937 | 0.3812 |
122
- | 0.7435 | 68.75 | 2750 | 1.4999 | 0.3563 |
123
- | 0.6917 | 70.0 | 2800 | 1.4961 | 0.35 |
124
- | 0.7504 | 71.25 | 2850 | 1.4931 | 0.3625 |
125
- | 0.6844 | 72.5 | 2900 | 1.4905 | 0.3812 |
126
- | 0.716 | 73.75 | 2950 | 1.4839 | 0.3812 |
127
- | 0.7411 | 75.0 | 3000 | 1.4834 | 0.3812 |
128
- | 0.7297 | 76.25 | 3050 | 1.4780 | 0.375 |
129
- | 0.6271 | 77.5 | 3100 | 1.4899 | 0.3563 |
130
- | 0.6132 | 78.75 | 3150 | 1.4973 | 0.3563 |
131
- | 0.6074 | 80.0 | 3200 | 1.4859 | 0.3688 |
132
- | 0.6556 | 81.25 | 3250 | 1.4871 | 0.3688 |
133
- | 0.5756 | 82.5 | 3300 | 1.4897 | 0.3563 |
134
- | 0.5904 | 83.75 | 3350 | 1.4871 | 0.3563 |
135
- | 0.5581 | 85.0 | 3400 | 1.4955 | 0.35 |
136
- | 0.5234 | 86.25 | 3450 | 1.4804 | 0.375 |
137
- | 0.5386 | 87.5 | 3500 | 1.4948 | 0.3563 |
138
- | 0.5279 | 88.75 | 3550 | 1.4928 | 0.35 |
139
- | 0.5393 | 90.0 | 3600 | 1.5030 | 0.35 |
140
- | 0.514 | 91.25 | 3650 | 1.5010 | 0.3438 |
141
- | 0.4654 | 92.5 | 3700 | 1.5036 | 0.35 |
142
- | 0.5209 | 93.75 | 3750 | 1.4993 | 0.3438 |
143
- | 0.4703 | 95.0 | 3800 | 1.4993 | 0.3438 |
144
- | 0.5013 | 96.25 | 3850 | 1.5010 | 0.3438 |
145
- | 0.5065 | 97.5 | 3900 | 1.5011 | 0.3438 |
146
- | 0.5142 | 98.75 | 3950 | 1.5013 | 0.35 |
147
- | 0.4589 | 100.0 | 4000 | 1.4997 | 0.3438 |
148
 
149
 
150
  ### Framework versions
@@ -152,4 +111,4 @@ The following hyperparameters were used during training:
152
  - Transformers 4.41.1
153
  - Pytorch 2.3.0+cu121
154
  - Datasets 2.19.2
155
- - Tokenizers 0.19.1
 
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
+ - imagefolder
8
  metrics:
9
  - accuracy
10
  model-index:
 
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.56875
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
30
 
31
  # vit-emotional-classifier
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: 1.2330
36
+ - Accuracy: 0.5687
37
 
38
  ## Model description
39
 
 
57
  - eval_batch_size: 16
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
+ - lr_scheduler_type: cosine_with_restarts
61
+ - num_epochs: 20
 
62
 
63
  ### Training results
64
 
65
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
66
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
67
+ | 2.0354 | 0.5 | 20 | 1.9924 | 0.275 |
68
+ | 1.9057 | 1.0 | 40 | 1.8266 | 0.3937 |
69
+ | 1.67 | 1.5 | 60 | 1.6951 | 0.3438 |
70
+ | 1.6237 | 2.0 | 80 | 1.5888 | 0.4437 |
71
+ | 1.5413 | 2.5 | 100 | 1.5295 | 0.45 |
72
+ | 1.4604 | 3.0 | 120 | 1.5311 | 0.4437 |
73
+ | 1.4092 | 3.5 | 140 | 1.5212 | 0.4562 |
74
+ | 1.4291 | 4.0 | 160 | 1.5151 | 0.4188 |
75
+ | 1.4044 | 4.5 | 180 | 1.4869 | 0.4313 |
76
+ | 1.3961 | 5.0 | 200 | 1.4961 | 0.4062 |
77
+ | 1.2358 | 5.5 | 220 | 1.3494 | 0.55 |
78
+ | 1.1714 | 6.0 | 240 | 1.3767 | 0.4875 |
79
+ | 1.0897 | 6.5 | 260 | 1.3382 | 0.5125 |
80
+ | 1.1153 | 7.0 | 280 | 1.2991 | 0.525 |
81
+ | 1.0265 | 7.5 | 300 | 1.3381 | 0.5188 |
82
+ | 1.0261 | 8.0 | 320 | 1.3143 | 0.5188 |
83
+ | 1.0227 | 8.5 | 340 | 1.3201 | 0.5125 |
84
+ | 0.9978 | 9.0 | 360 | 1.3939 | 0.4375 |
85
+ | 0.9222 | 9.5 | 380 | 1.3469 | 0.4875 |
86
+ | 0.9489 | 10.0 | 400 | 1.2344 | 0.525 |
87
+ | 0.8079 | 10.5 | 420 | 1.1800 | 0.5938 |
88
+ | 0.7336 | 11.0 | 440 | 1.1935 | 0.5687 |
89
+ | 0.7158 | 11.5 | 460 | 1.2030 | 0.5813 |
90
+ | 0.7004 | 12.0 | 480 | 1.2705 | 0.525 |
91
+ | 0.7589 | 12.5 | 500 | 1.2186 | 0.5687 |
92
+ | 0.6966 | 13.0 | 520 | 1.2049 | 0.6125 |
93
+ | 0.5767 | 13.5 | 540 | 1.2057 | 0.55 |
94
+ | 0.6569 | 14.0 | 560 | 1.2047 | 0.5312 |
95
+ | 0.5291 | 14.5 | 580 | 1.2649 | 0.55 |
96
+ | 0.4644 | 15.0 | 600 | 1.2103 | 0.5938 |
97
+ | 0.4895 | 15.5 | 620 | 1.2741 | 0.5687 |
98
+ | 0.4769 | 16.0 | 640 | 1.2278 | 0.5875 |
99
+ | 0.4804 | 16.5 | 660 | 1.3078 | 0.5188 |
100
+ | 0.5169 | 17.0 | 680 | 1.2963 | 0.5 |
101
+ | 0.4137 | 17.5 | 700 | 1.3284 | 0.55 |
102
+ | 0.3856 | 18.0 | 720 | 1.2393 | 0.5563 |
103
+ | 0.4545 | 18.5 | 740 | 1.2368 | 0.5687 |
104
+ | 0.4186 | 19.0 | 760 | 1.2490 | 0.5938 |
105
+ | 0.3814 | 19.5 | 780 | 1.2246 | 0.55 |
106
+ | 0.3813 | 20.0 | 800 | 1.3488 | 0.5375 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
107
 
108
 
109
  ### Framework versions
 
111
  - Transformers 4.41.1
112
  - Pytorch 2.3.0+cu121
113
  - Datasets 2.19.2
114
+ - Tokenizers 0.19.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:73ee0716b8979dc47d56eb94e140e0113696b4d1fee3ff870b068e9d0a7afb90
3
  size 343242432
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a38e6e28a72e69f611c351a248033c277f5df03385d87dc691310868df716a2
3
  size 343242432
runs/Jun04_03-23-15_2a967390fece/events.out.tfevents.1717472301.2a967390fece.1464.16 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c69bca6190c1193ef45f67cb0262f9bccf9ee82e2b7de89536850e1023dc078b
3
+ size 411