BERTje_finetuned_1994

This model is a fine-tuned version of GroNLP/bert-base-dutch-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1184
• F1: 0.1799
• Precision: 0.1350
• Recall: 0.2698

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: 2e-05
• train_batch_size: 3
• eval_batch_size: 3
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 6
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Recall
No log	1.0	110	0.2231	0.0522	0.0268	0.9688
No log	2.0	221	0.1446	0.1050	0.0576	0.5899
No log	3.0	331	0.1264	0.1726	0.1151	0.3452
No log	4.0	442	0.1201	0.1812	0.1350	0.2759
0.1954	4.98	550	0.1184	0.1799	0.1350	0.2698

Framework versions

• Transformers 4.35.0
• Pytorch 1.11.0+cu102
• Datasets 2.14.6
• Tokenizers 0.14.1