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: 0.2606
• Accuracy: 0.9148

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 512
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6309	0.3328	100	0.2618	0.9145
0.6165	0.6656	200	0.2600	0.9150
0.6283	0.9983	300	0.2659	0.9135
0.6171	1.3311	400	0.2561	0.9174
0.6112	1.6639	500	0.2606	0.9148
0.6081	1.9967	600	0.2624	0.9137
0.5885	2.3295	700	0.2671	0.9113
0.5975	2.6622	800	0.2572	0.9156
0.6067	2.9950	900	0.2683	0.9116

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1