efontes-ner-person

This model is a fine-tuned version of efontes/efontes-ner on a dataset of the Lvivs documents from Center of Digital Humanities . It achieves the following results on the evaluation set:

• F1: 0.8488
• Precision: 0.8163
• Recall: 0.8840

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	7	0.0873	0.9745	0.9747	0.9754	0.9745
No log	2.0	14	0.0603	0.9813	0.9811	0.9811	0.9813
No log	3.0	21	0.0554	0.9819	0.9826	0.9841	0.9819
No log	4.0	28	0.0483	0.9853	0.9853	0.9853	0.9853
No log	5.0	35	0.0482	0.9850	0.9850	0.9851	0.9850

Framework versions

• Transformers 5.18.0
• Pytorch 2.11.0+cu130
• Datasets 4.8.5
• Tokenizers 0.23.2