nj1867/roof_classification_new_dataset_4_march

This model is a fine-tuned version of google/vit-base-patch32-384 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.1523
• Validation Loss: 0.7698
• Train Accuracy: 0.8375
• Epoch: 19

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': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 7440, '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': 0.0001}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Accuracy	Epoch
4.2343	4.0917	0.0437	0
3.7984	3.8481	0.1313	1
3.3245	3.5615	0.1062	2
2.8352	3.1190	0.3	3
2.3077	2.7024	0.4	4
1.8445	2.4423	0.4625	5
1.4533	2.0483	0.5813	6
1.1535	1.8203	0.65	7
0.9074	1.4581	0.725	8
0.7190	1.3655	0.7812	9
0.5880	1.2685	0.7875	10
0.4729	1.2005	0.7875	11
0.4004	0.9699	0.825	12
0.3349	0.9435	0.8313	13
0.2789	0.8993	0.8438	14
0.2179	0.8071	0.8625	15
0.2061	0.7852	0.8812	16
0.2124	0.7664	0.8625	17
0.1623	0.8276	0.8562	18
0.1523	0.7698	0.8375	19

Framework versions

• Transformers 4.38.2
• TensorFlow 2.13.1
• Datasets 2.18.0
• Tokenizers 0.15.2