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: 1.3560
• Accuracy: 0.5188

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	5	1.6699	0.4313
1.5821	2.0	10	1.6118	0.4562
1.5821	3.0	15	1.5550	0.475
1.445	4.0	20	1.5128	0.5062
1.445	5.0	25	1.4508	0.5375
1.3202	6.0	30	1.4364	0.5
1.3202	7.0	35	1.3776	0.575
1.2242	8.0	40	1.3966	0.5
1.2242	9.0	45	1.3724	0.525
1.1589	10.0	50	1.3483	0.525
1.1589	11.0	55	1.3186	0.5687
1.0962	12.0	60	1.3295	0.5375
1.0962	13.0	65	1.3058	0.5875
1.0542	14.0	70	1.3296	0.5375
1.0542	15.0	75	1.3185	0.5813

Framework versions

• Transformers 4.52.4
• Pytorch 2.7.1+cu128
• Datasets 3.6.0
• Tokenizers 0.21.1