Prahas10/roof-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: 0.1015
• Validation Loss: 0.3231
• Train Accuracy: 0.9083
• 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': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 4e-05, 'decay_steps': 138270, '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
3.8367	2.9703	0.4403	0
1.3092	1.6169	0.7093	1
0.4529	1.4414	0.7112	2
0.2229	0.8445	0.8368	3
0.1451	0.7074	0.8556	4
0.1053	0.8585	0.7992	5
0.1175	1.0721	0.7389	6
0.1388	0.5802	0.8542	7
0.0647	0.3764	0.9083	8
0.1049	1.0484	0.7366	9
0.0740	0.6191	0.8321	10
0.0816	0.6273	0.8283	11
0.0981	0.2901	0.9172	12
0.0614	0.5081	0.8523	13
0.0548	0.4983	0.8612	14
0.0652	0.8008	0.7850	15
0.0857	0.5845	0.8415	16
0.0847	0.6887	0.8184	17
0.0645	0.6104	0.8405	18
0.0891	0.4770	0.8532	19
0.0532	0.5074	0.8500	20
0.0483	0.8208	0.7850	21
0.0498	0.2679	0.9083	22
0.0406	0.3261	0.9036	23
0.0578	0.6373	0.8340	24
0.1010	0.5037	0.8481	25
0.0583	0.2993	0.8984	26
0.0398	0.1538	0.9492	27
0.0492	0.4397	0.8641	28
0.1015	0.3231	0.9083	29

Framework versions

• Transformers 4.41.1
• TensorFlow 2.15.0
• Datasets 2.19.1
• Tokenizers 0.19.1