flausch_span_gbert-large_all

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: 0.2622
• Model Preparation Time: 0.0328
• Precision: 0.4348
• Recall: 0.5821
• F1: 0.4978

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
0.9007	0.2822	500	0.7977	0.0328	0.0	0.0	0.0
0.7223	0.5643	1000	0.6214	0.0328	0.2047	0.1514	0.1741
0.5327	0.8465	1500	0.4041	0.0328	0.2262	0.3583	0.2773
0.3691	1.1287	2000	0.3349	0.0328	0.2819	0.4330	0.3415
0.3021	1.4108	2500	0.3013	0.0328	0.3222	0.4524	0.3764
0.2548	1.6930	3000	0.2655	0.0328	0.3821	0.5263	0.4428
0.2744	1.9752	3500	0.2666	0.0328	0.3072	0.4037	0.3489
0.1874	2.2573	4000	0.2803	0.0328	0.4124	0.5461	0.4700
0.1767	2.5395	4500	0.2625	0.0328	0.4421	0.5802	0.5018
0.1708	2.8217	5000	0.2622	0.0328	0.4348	0.5821	0.4978

Framework versions

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