nguyennghia0902/electra-small-discriminator_1e-05_16

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

• Train Loss: 2.2953
• Train End Logits Accuracy: 0.4633
• Train Start Logits Accuracy: 0.4286
• Validation Loss: 2.1111
• Validation End Logits Accuracy: 0.4964
• Validation Start Logits Accuracy: 0.4762
• Epoch: 9
• {'name': 'project02_google/electra-small-discriminator_1e-05_16', 'lnr': 1e-05, 'epoch': 10, 'batch_size': 16, 'time': 15051.128346920013, 'accuracy': 0, 'f1_score': 0}

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': 1e-05, 'decay_steps': 31270, '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	Train End Logits Accuracy	Train Start Logits Accuracy	Validation Loss	Validation End Logits Accuracy	Validation Start Logits Accuracy	Epoch
3.6922	0.2260	0.1855	2.9570	0.3296	0.2955	0
2.9538	0.3373	0.2952	2.6908	0.3760	0.3455	1
2.7599	0.3690	0.3326	2.5323	0.4072	0.3820	2
2.6351	0.3920	0.3568	2.4256	0.4286	0.4008	3
2.5472	0.4089	0.3742	2.3283	0.4498	0.4264	4
2.4725	0.4221	0.3912	2.2602	0.4605	0.4399	5
2.4119	0.4369	0.4017	2.1953	0.4765	0.4559	6
2.3562	0.4505	0.4144	2.1406	0.4888	0.4689	7
2.3220	0.4566	0.4216	2.1207	0.4947	0.4749	8
2.2953	0.4633	0.4286	2.1111	0.4964	0.4762	9

Framework versions

• Transformers 4.39.3
• TensorFlow 2.15.0
• Datasets 2.18.0
• Tokenizers 0.15.2