robbert0510_lrate10b32

This model is a fine-tuned version of pdelobelle/robbert-v2-dutch-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5022
• Precisions: 0.8217
• Recall: 0.7879
• F-measure: 0.8011
• Accuracy: 0.9168

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Precisions	Recall	F-measure	Accuracy
0.625	1.0	118	0.4136	0.8611	0.6751	0.6866	0.8713
0.3164	2.0	236	0.3284	0.8110	0.7374	0.7474	0.8959
0.187	3.0	354	0.3490	0.7578	0.7662	0.7524	0.9060
0.1083	4.0	472	0.3738	0.8164	0.7748	0.7855	0.9120
0.0778	5.0	590	0.4234	0.7579	0.7677	0.7583	0.9042
0.0526	6.0	708	0.4802	0.8348	0.7556	0.7688	0.9049
0.0358	7.0	826	0.4723	0.8322	0.7671	0.7909	0.9107
0.023	8.0	944	0.4758	0.8024	0.7975	0.7973	0.9107
0.016	9.0	1062	0.5112	0.7991	0.7889	0.7917	0.9084
0.0117	10.0	1180	0.5022	0.8217	0.7879	0.8011	0.9168
0.0072	11.0	1298	0.5286	0.8190	0.7875	0.8009	0.9165
0.0052	12.0	1416	0.5207	0.8056	0.7929	0.7987	0.9135

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.0