MeMo_BERT-WSD-danish-bert-botxo

This model is a fine-tuned version of Maltehb/danish-bert-botxo on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.4175
• F1-score: 0.1942

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	61	1.4175	0.1942
No log	2.0	122	1.4359	0.1229
No log	3.0	183	1.4326	0.1477
No log	4.0	244	1.3837	0.1229
No log	5.0	305	1.3969	0.1229
No log	6.0	366	1.4012	0.1229
No log	7.0	427	1.3349	0.1229
No log	8.0	488	1.3937	0.1229
1.301	9.0	549	1.3602	0.1229
1.301	10.0	610	1.4346	0.1229
1.301	11.0	671	1.3953	0.1229
1.301	12.0	732	1.4039	0.1229
1.301	13.0	793	1.3816	0.1229
1.301	14.0	854	1.3988	0.1229
1.301	15.0	915	1.4007	0.1229
1.301	16.0	976	1.3924	0.1229
1.278	17.0	1037	1.4031	0.1229
1.278	18.0	1098	1.3957	0.1229
1.278	19.0	1159	1.3956	0.1229
1.278	20.0	1220	1.3970	0.1229

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2