flausch_span_gbert-large_non_labeled_spans

This model is a fine-tuned version of deepset/gbert-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1736
• Model Preparation Time: 0.0411
• Precision: 0.6750
• Recall: 0.7428
• F1: 0.7073

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Precision	Recall	F1
0.2595	0.4233	750	0.1754	0.0411	0.5105	0.6319	0.5648
0.1821	0.8465	1500	0.1683	0.0411	0.5954	0.6371	0.6156
0.1394	1.2698	2250	0.1649	0.0411	0.5722	0.6831	0.6228
0.1099	1.6930	3000	0.1547	0.0411	0.6677	0.7224	0.6940
0.0989	2.1163	3750	0.1507	0.0411	0.6429	0.7309	0.6841
0.0605	2.5395	4500	0.1712	0.0411	0.6724	0.7399	0.7046
0.0603	2.9628	5250	0.1736	0.0411	0.6750	0.7428	0.7073

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1