VIT_fourclass_classifier

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0945
• Validation Loss: 1.7241
• Train Accuracy: 0.6974
• Epoch: 14

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:

• optimizer: {'name': 'SGD', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': np.float32(0.01), 'momentum': 0.0, 'nesterov': False}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Accuracy	Epoch
0.7946	1.1484	0.6272	0
0.3246	1.1792	0.6769	1
0.2266	1.2812	0.6842	2
0.1841	1.5085	0.6754	3
0.1589	1.4224	0.6944	4
0.1244	1.4229	0.6901	5
0.1174	1.4858	0.6784	6
0.1133	1.4221	0.6974	7
0.1026	1.4273	0.7003	8
0.1083	1.5406	0.7003	9
0.1038	1.6223	0.6974	10
0.0876	1.5613	0.6959	11
0.1018	1.4540	0.7149	12
0.0808	1.4853	0.7193	13
0.0945	1.7241	0.6974	14

Framework versions

• Transformers 4.52.4
• TensorFlow 2.18.0
• Datasets 3.6.0
• Tokenizers 0.21.1