task2_flausch_classification_gbert-large_span_classifier_with_nonspan

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: 0.2307
• Accuracy: 0.9479
• F1: 0.9467

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	Accuracy	F1
0.3749	0.2697	1000	0.3099	0.9149	0.9174
0.2804	0.5394	2000	0.2330	0.9371	0.9305
0.2672	0.8091	3000	0.2369	0.9391	0.9339
0.2248	1.0787	4000	0.2295	0.9427	0.9406
0.1942	1.3484	5000	0.2244	0.9452	0.9422
0.1873	1.6181	6000	0.2310	0.9423	0.9393
0.1768	1.8878	7000	0.2155	0.9469	0.9450
0.1443	2.1575	8000	0.2295	0.9454	0.9449
0.1198	2.4272	9000	0.2295	0.9484	0.9474
0.1238	2.6969	10000	0.2278	0.9479	0.9469
0.1166	2.9666	11000	0.2307	0.9479	0.9467

Framework versions

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