DanskBERT_FGN

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

• Loss: 1.4535
• F1-score: 0.8102

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:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1-score
No log	1.0	120	0.6078	0.7542
No log	2.0	240	0.5319	0.7838
No log	3.0	360	0.7563	0.7942
No log	4.0	480	0.9098	0.7940
0.4909	5.0	600	1.2334	0.7728
0.4909	6.0	720	1.1453	0.7935
0.4909	7.0	840	1.2240	0.8017
0.4909	8.0	960	1.3105	0.7907
0.0877	9.0	1080	1.4305	0.8002
0.0877	10.0	1200	1.4420	0.7764
0.0877	11.0	1320	1.4535	0.8102
0.0877	12.0	1440	1.4482	0.7734
0.0326	13.0	1560	1.5240	0.8046
0.0326	14.0	1680	1.5290	0.8031
0.0326	15.0	1800	1.5302	0.7964
0.0326	16.0	1920	1.5503	0.7904
0.0159	17.0	2040	1.5510	0.7904
0.0159	18.0	2160	1.5714	0.7957
0.0159	19.0	2280	1.5932	0.7963
0.0159	20.0	2400	1.5998	0.7955

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1