Prahas10/roof-test

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: 0.0637
• Validation Loss: 0.1264
• Train Accuracy: 0.9474
• Epoch: 28

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': 3990, '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
2.6939	2.4863	0.2807	0
2.1820	2.2454	0.4912	1
1.8026	1.8798	0.4912	2
1.4641	1.6673	0.5439	3
1.1288	1.3594	0.6842	4
0.9426	1.0517	0.8070	5
0.6577	0.8531	0.8421	6
0.5025	0.6971	0.8772	7
0.3976	0.5785	0.8596	8
0.3052	0.5568	0.9123	9
0.2562	0.5137	0.8947	10
0.3250	0.4415	0.9298	11
0.2773	0.8003	0.7368	12
0.2694	0.4544	0.8421	13
0.2180	0.5179	0.8947	14
0.1515	0.3450	0.9825	15
0.1386	0.2818	0.9825	16
0.1058	0.1962	0.9649	17
0.0724	0.2456	0.9825	18
0.0604	0.2432	0.9649	19
0.0718	0.2548	1.0	20
0.0507	0.2760	0.9474	21
0.0453	0.1565	0.9825	22
0.0274	0.1377	0.9825	23
0.0396	0.1906	0.9649	24
0.0360	0.1217	0.9825	25
0.0307	0.2234	0.9474	26
0.0427	0.2861	0.9298	27
0.0637	0.1264	0.9474	28

Framework versions

• Transformers 4.38.2
• TensorFlow 2.15.0
• Datasets 2.16.1
• Tokenizers 0.15.2