Labira/LabiraPJOK_3x_50

This model is a fine-tuned version of Labira/LabiraPJOK_2x_50 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0125
• Validation Loss: 1.5431
• Epoch: 49

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': 'Adam', 'weight_decay': None, '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': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 450, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Validation Loss	Epoch
2.7214	1.2242	0
1.5828	1.1158	1
0.9946	1.0677	2
0.7404	1.2115	3
0.5481	1.0920	4
0.3599	1.1031	5
0.2659	1.1035	6
0.2725	1.1251	7
0.2207	1.1364	8
0.1379	1.2039	9
0.1687	1.2331	10
0.1154	1.1677	11
0.1126	1.2093	12
0.0953	1.2532	13
0.0753	1.2455	14
0.0519	1.2544	15
0.0603	1.2511	16
0.0609	1.2736	17
0.0530	1.2692	18
0.0384	1.2869	19
0.0337	1.3048	20
0.0304	1.3314	21
0.0565	1.3378	22
0.0351	1.3842	23
0.0480	1.4148	24
0.0308	1.3959	25
0.0454	1.3768	26
0.0557	1.4469	27
0.0397	1.4431	28
0.0212	1.4441	29
0.0251	1.4262	30
0.0291	1.4412	31
0.0194	1.5155	32
0.0238	1.5136	33
0.0209	1.5002	34
0.0183	1.4976	35
0.0204	1.5533	36
0.0183	1.6057	37
0.0147	1.6047	38
0.0137	1.6029	39
0.0090	1.5879	40
0.0323	1.5802	41
0.0181	1.5748	42
0.0144	1.5629	43
0.0215	1.5534	44
0.0058	1.5442	45
0.0144	1.5485	46
0.0122	1.5449	47
0.0139	1.5428	48
0.0125	1.5431	49

Framework versions

• Transformers 4.44.2
• TensorFlow 2.17.0
• Datasets 3.0.1
• Tokenizers 0.19.1