vit-Facial-Expression-Recognition

This model is a fine-tuned version of motheecreator/vit-Facial-Expression-Recognition on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.5713
• Accuracy: 0.8308

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• 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: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3894	1.2198	100	0.3854	0.8667
0.3556	2.4397	200	0.3952	0.8628
0.3545	3.6595	300	0.4057	0.8604
0.3243	4.8794	400	0.4035	0.8565
0.3718	6.0977	500	0.4029	0.8586
0.342	7.3176	600	0.4279	0.8493
0.3021	8.5374	700	0.4375	0.8449
0.2795	9.7573	800	0.4542	0.8468
0.2642	10.9771	900	0.4717	0.8434
0.2223	12.1954	1000	0.4991	0.8367
0.1875	13.4153	1100	0.5185	0.8346
0.1481	14.6351	1200	0.5357	0.8351
0.1291	15.8550	1300	0.5409	0.8350
0.1021	17.0733	1400	0.5684	0.8292
0.0851	18.2931	1500	0.5752	0.8296
0.076	19.5130	1600	0.5712	0.8308

Framework versions

• Transformers 4.48.2
• Pytorch 2.6.0+cu126
• Datasets 3.2.0
• Tokenizers 0.21.0