Xanadu00/galaxy_classifier_mobilevit_3

This model is a fine-tuned version of apple/mobilevit-small on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.1914
• Train Accuracy: 0.9341
• Validation Loss: 0.5148
• Validation Accuracy: 0.8512
• Epoch: 16

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': 'AdamW', 'weight_decay': 0.01, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'ExponentialDecay', 'config': {'initial_learning_rate': 0.002, 'decay_steps': 10000, 'decay_rate': 0.01, 'staircase': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
1.1049	0.6128	0.7422	0.7517	0
0.7149	0.7564	0.6376	0.7821	1
0.6080	0.7945	0.6947	0.7745	2
0.5376	0.8160	0.5589	0.8134	3
0.4977	0.8279	0.5458	0.8162	4
0.4564	0.8407	0.4799	0.8441	5
0.4271	0.8557	0.4765	0.8413	6
0.3957	0.8619	0.4790	0.8453	7
0.3701	0.8741	0.5376	0.8329	8
0.3425	0.8829	0.4359	0.8619	9
0.3192	0.8892	0.4475	0.8585	10
0.2972	0.8967	0.4143	0.8712	11
0.2691	0.9080	0.4819	0.8498	12
0.2445	0.9144	0.4543	0.8563	13
0.2261	0.9220	0.4221	0.8689	14
0.2127	0.9251	0.5076	0.8540	15
0.1914	0.9341	0.5148	0.8512	16

Framework versions

• Transformers 4.30.2
• TensorFlow 2.12.0
• Datasets 2.13.1
• Tokenizers 0.13.3