xlmroberta-ner-prostata-bs32

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

• Loss: 0.0326
• F1: 0.9542
• Precision: 0.9484
• Recall: 0.9600

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 32
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Recall
No log	1.0	98	0.4072	0.2942	0.2728	0.3191
No log	2.0	196	0.0944	0.8335	0.8155	0.8524
No log	3.0	294	0.0405	0.9191	0.9120	0.9262
No log	4.0	392	0.0306	0.9354	0.9204	0.9508
No log	5.0	490	0.0415	0.9294	0.9060	0.9540
10.6497	6.0	588	0.0257	0.9577	0.9555	0.9599
10.6497	7.0	686	0.0273	0.9624	0.9637	0.9612
10.6497	8.0	784	0.0232	0.9683	0.9689	0.9676
10.6497	9.0	882	0.0249	0.9693	0.9677	0.9709
10.6497	10.0	980	0.0237	0.9706	0.9684	0.9728

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 5.0.0
• Tokenizers 0.22.2