emotion_classification
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2002
- Accuracy: 0.6438
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 12
- num_epochs: 15
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Accuracy |
| No log |
1.0 |
40 |
1.9756 |
0.2313 |
| No log |
2.0 |
80 |
1.6788 |
0.3937 |
| No log |
3.0 |
120 |
1.5219 |
0.5375 |
| No log |
4.0 |
160 |
1.4542 |
0.45 |
| No log |
5.0 |
200 |
1.3923 |
0.5 |
| No log |
6.0 |
240 |
1.3595 |
0.4437 |
| No log |
7.0 |
280 |
1.3111 |
0.5125 |
| No log |
8.0 |
320 |
1.2050 |
0.5625 |
| No log |
9.0 |
360 |
1.2387 |
0.5437 |
| No log |
10.0 |
400 |
1.2847 |
0.5437 |
| No log |
11.0 |
440 |
1.2048 |
0.5625 |
| No log |
12.0 |
480 |
1.2270 |
0.5563 |
| 1.0855 |
13.0 |
520 |
1.2058 |
0.5875 |
| 1.0855 |
14.0 |
560 |
1.1999 |
0.5625 |
| 1.0855 |
15.0 |
600 |
1.2032 |
0.5687 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1
