sayakpaul/masked-lm-tpu

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

• Train Loss: 9.9067
• Train Accuracy: 0.0116
• Validation Loss: 9.8225
• Validation Accuracy: 0.0198
• Epoch: 8

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': 0.0001, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0001, 'decay_steps': 22325, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1175, '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: float32

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
10.2116	0.0	10.1957	0.0000	0
10.2017	0.0	10.1798	0.0	1
10.1890	0.0000	10.1604	0.0000	2
10.1733	0.0000	10.1145	0.0000	3
10.1336	0.0000	10.0666	0.0000	4
10.0906	0.0001	10.0200	0.0005	5
10.0360	0.0006	9.9646	0.0049	6
9.9830	0.0038	9.8938	0.0155	7
9.9067	0.0116	9.8225	0.0198	8

Framework versions

• Transformers 4.29.2
• TensorFlow 2.12.0
• Tokenizers 0.13.3