Prahas10/shingles

This model is a fine-tuned version of google/vit-base-patch16-384 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0993
• Validation Loss: 0.6967
• Train Accuracy: 0.8166
• Epoch: 29

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': 'AdamWeightDecay', 'learning_rate': {'module': 'transformers.optimization_tf', 'class_name': 'WarmUp', 'config': {'initial_learning_rate': 4e-05, 'decay_schedule_fn': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 4e-05, 'decay_steps': 127899.75, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'warmup_steps': 10370.25, 'power': 1.0, 'name': None}, 'registered_name': 'WarmUp'}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.0001}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Accuracy	Epoch
5.2368	5.2154	0.0047	0
5.1655	5.1337	0.0113	1
5.0415	4.9860	0.0278	2
4.8179	4.7812	0.0781	3
4.4541	4.4703	0.1844	4
3.9330	4.0779	0.2841	5
3.3155	3.6691	0.3650	6
2.6546	3.3371	0.4313	7
2.0435	3.0037	0.4727	8
1.5258	2.7059	0.5193	9
1.1079	2.4174	0.5588	10
0.7989	2.3590	0.5532	11
0.5857	1.9721	0.6298	12
0.4337	1.7442	0.6896	13
0.3352	1.7334	0.6580	14
0.2641	1.6197	0.6670	15
0.2042	1.7021	0.6289	16
0.1642	1.3843	0.7070	17
0.1500	1.4422	0.6787	18
0.1251	1.2797	0.7098	19
0.1093	0.9233	0.8020	20
0.1215	0.9209	0.7977	21
0.1007	0.9143	0.7803	22
0.0811	0.7952	0.8090	23
0.0953	0.7678	0.8260	24
0.1033	0.8928	0.7705	25
0.0636	0.3480	0.9271	26
0.0880	0.5916	0.8669	27
0.0861	0.8892	0.7789	28
0.0993	0.6967	0.8166	29

Framework versions

• Transformers 4.41.0
• TensorFlow 2.15.0
• Datasets 2.19.1
• Tokenizers 0.19.1