arieg/bw_spec_cls_4_01_noise_200

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: 0.0370
• Train Categorical Accuracy: 0.2486
• Validation Loss: 0.0349
• Validation Categorical Accuracy: 0.2625
• Epoch: 9

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', 'clipnorm': 1.0, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 7200, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Train Categorical Accuracy	Validation Loss	Validation Categorical Accuracy	Epoch
0.6021	0.2458	0.2372	0.2625	0
0.1654	0.2486	0.1210	0.2625	1
0.1042	0.2486	0.0902	0.2625	2
0.0819	0.2486	0.0741	0.2625	3
0.0688	0.2486	0.0634	0.2625	4
0.0595	0.2486	0.0553	0.2625	5
0.0522	0.2486	0.0488	0.2625	6
0.0462	0.2486	0.0434	0.2625	7
0.0412	0.2486	0.0388	0.2625	8
0.0370	0.2486	0.0349	0.2625	9

Framework versions

• Transformers 4.35.0
• TensorFlow 2.14.0
• Datasets 2.14.6
• Tokenizers 0.14.1