gustavokpc/IC_10

This model is a fine-tuned version of bert-large-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0756
• Train Accuracy: 0.9729
• Train F1 M: 0.5577
• Train Precision M: 0.4039
• Train Recall M: 0.9673
• Validation Loss: 0.2783
• Validation Accuracy: 0.9208
• Validation F1 M: 0.5605
• Validation Precision M: 0.4027
• Validation Recall M: 0.9687
• Epoch: 6

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': False, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 1e-05, 'decay_steps': 5306, '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 Accuracy	Train F1 M	Train Precision M	Train Recall M	Validation Loss	Validation Accuracy	Validation F1 M	Validation Precision M	Validation Recall M	Epoch
0.4941	0.7625	0.4930	0.3937	0.7597	0.3258	0.8608	0.5445	0.3975	0.9062	0
0.2861	0.8881	0.5436	0.3975	0.9163	0.2484	0.8925	0.5735	0.4141	0.9729	1
0.2029	0.9272	0.5530	0.4014	0.9460	0.2034	0.9169	0.5533	0.4020	0.9314	2
0.1480	0.9475	0.5520	0.4012	0.9488	0.2172	0.9222	0.5650	0.4062	0.9732	3
0.1161	0.9598	0.5549	0.4023	0.9600	0.2588	0.9109	0.5664	0.4076	0.9706	4
0.0916	0.9665	0.5575	0.4033	0.9647	0.2673	0.9156	0.5573	0.4013	0.9574	5
0.0756	0.9729	0.5577	0.4039	0.9673	0.2783	0.9208	0.5605	0.4027	0.9687	6

Framework versions

• Transformers 4.34.1
• TensorFlow 2.14.0
• Datasets 2.14.5
• Tokenizers 0.14.1