Labira/indobert-qa-asean

This model is a fine-tuned version of indolem/indobert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 3.3671
• Validation Loss: 3.8957
• Epoch: 11

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': 2e-05, 'decay_steps': 96, '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
5.6394	4.8426	0
4.8298	4.5540	1
4.4180	4.3460	2
4.1194	4.1989	3
3.8009	4.1006	4
3.6069	4.0052	5
3.4114	3.9418	6
3.3122	3.8957	7
3.3178	3.8957	8
3.2873	3.8957	9
3.2908	3.8957	10
3.3671	3.8957	11

Framework versions

• Transformers 4.40.1
• TensorFlow 2.15.0
• Datasets 2.19.1
• Tokenizers 0.19.1