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: 0.6218
Accuracy: 0.8672

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: 0.0002
train_batch_size: 16
eval_batch_size: 16
seed: 42
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
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6535	1.0	32	0.7740	0.7812
0.296	2.0	64	0.5949	0.8047
0.1542	3.0	96	0.6261	0.8047
0.0838	4.0	128	0.6315	0.8047
0.0189	5.0	160	0.8360	0.7656
0.0223	6.0	192	0.7264	0.8047
0.0114	7.0	224	0.8634	0.8047
0.0408	8.0	256	0.6393	0.8438
0.0075	9.0	288	0.7023	0.8438
0.006	10.0	320	0.6481	0.8516
0.0053	11.0	352	0.6247	0.8516
0.0047	12.0	384	0.6220	0.8594
0.0043	13.0	416	0.6186	0.8594
0.0041	14.0	448	0.6193	0.8594
0.0038	15.0	480	0.6205	0.8594
0.0036	16.0	512	0.6205	0.8594
0.0035	17.0	544	0.6218	0.8672
0.0034	18.0	576	0.6224	0.8672
0.0033	19.0	608	0.6229	0.8672
0.0033	20.0	640	0.6232	0.8672

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.1

panjiallatief/vit-emotion-classifier

vit-emotion-classifier

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

panjiallatief/
vit-emotion-classifier