Prgrg/ja-en-JESC-v4.0-wth-lr-decay

This model is a fine-tuned version of Prgrg/ja-en-JESC-v3.0-wth-lr-decay on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 2.0999
• Validation Loss: 2.5673
• Epoch: 7

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': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 0.0002, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0002, 'decay_steps': 199000, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.001}
• training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Epoch
2.4825	2.6526	0
2.4354	2.6193	1
2.3724	2.5918	2
2.3112	2.5749	3
2.2521	2.5652	4
2.1967	2.5622	5
2.1449	2.5608	6
2.0999	2.5673	7

Framework versions

• Transformers 4.26.1
• TensorFlow 2.11.0
• Datasets 2.10.1
• Tokenizers 0.13.2