NInjaQuarrior/vit-base-patch16-224-in21k-disaster2

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: 0.1002
• Train Caccuracy: 0.9936
• Train Fn: 14.0
• Train Fp: 17.0
• Train Tn: 4647.0
• Train Tp: 2318.0
• Validation Loss: 0.1452
• Validation Caccuracy: 0.9757
• Validation Fn: 9.0
• Validation Fp: 12.0
• Validation Tn: 812.0
• Validation Tp: 403.0
• Epoch: 2

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: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 219, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
• training_precision: mixed_float16

Training results

Train Loss	Train Caccuracy	Train Fn	Train Fp	Train Tn	Train Tp	Validation Loss	Validation Caccuracy	Validation Fn	Validation Fp	Validation Tn	Validation Tp	Epoch
0.5143	0.9112	729.0	38.0	4626.0	1603.0	0.2132	0.9684	14.0	14.0	810.0	398.0	0
0.1474	0.9850	36.0	41.0	4623.0	2296.0	0.1596	0.9709	12.0	14.0	810.0	400.0	1
0.1002	0.9936	14.0	17.0	4647.0	2318.0	0.1452	0.9757	9.0	12.0	812.0	403.0	2

Framework versions

• Transformers 4.35.2
• TensorFlow 2.10.0
• Datasets 2.15.0
• Tokenizers 0.15.0