Svenskmodell

This model is a fine-tuned version of KB/bert-base-swedish-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.3411
• Train End Logits Accuracy: 0.8882
• Train Start Logits Accuracy: 0.8838
• Validation Loss: 1.4988
• Validation End Logits Accuracy: 0.6713
• Validation Start Logits Accuracy: 0.6669
• Epoch: 3

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', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 28792, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, '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
1.2392	0.6241	0.6197	1.0970	0.6588	0.6543	0
0.7549	0.7615	0.7569	1.1185	0.6673	0.6665	1
0.5005	0.8376	0.8323	1.3299	0.6571	0.6503	2
0.3411	0.8882	0.8838	1.4988	0.6713	0.6669	3

Framework versions

• Transformers 4.20.1
• TensorFlow 2.6.4
• Datasets 2.1.0
• Tokenizers 0.12.1