nguyennghia0902/electra-small-discriminator_0.0005_32_15e

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: 0.5725
• Train End Logits Accuracy: 0.8401
• Train Start Logits Accuracy: 0.8151
• Validation Loss: 0.2404
• Validation End Logits Accuracy: 0.9316
• Validation Start Logits Accuracy: 0.9222
• Epoch: 14

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': 0.0005, 'decay_steps': 23445, '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
3.3489	0.2750	0.2432	2.6409	0.3858	0.3668	0
2.7567	0.3772	0.3444	2.3037	0.4607	0.4455	1
2.5118	0.4254	0.3927	2.0684	0.5046	0.4834	2
2.3234	0.4624	0.4283	1.8489	0.5461	0.5257	3
2.1433	0.4977	0.4608	1.6848	0.5907	0.5742	4
1.9832	0.5289	0.4980	1.4704	0.6378	0.6177	5
1.8204	0.5619	0.5290	1.2837	0.6769	0.6665	6
1.6387	0.5991	0.5696	1.0838	0.7217	0.7115	7
1.4657	0.6379	0.6048	0.9057	0.7589	0.7562	8
1.2902	0.6729	0.6458	0.7410	0.8034	0.7975	9
1.1103	0.7149	0.6867	0.5707	0.8407	0.8374	10
0.9500	0.7493	0.7214	0.4523	0.8761	0.8660	11
0.7931	0.7855	0.7606	0.3483	0.9018	0.8924	12
0.6702	0.8166	0.7889	0.2710	0.9236	0.9152	13
0.5725	0.8401	0.8151	0.2404	0.9316	0.9222	14

Framework versions

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