xlmr_norwegian_immigration2

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

• Loss: 0.2207
• Accuracy: 0.9231
• 1-f1: 0.8810
• 1-recall: 0.8605
• 1-precision: 0.9024
• Balanced Acc: 0.9072

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: 128
• eval_batch_size: 128
• 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: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	1-f1	1-recall	1-precision	Balanced Acc
0.6746	1.0	5	0.6397	0.6692	0.0	0.0	0.0	0.5
0.5485	2.0	10	0.6313	0.6692	0.0	0.0	0.0	0.5
0.6165	3.0	15	0.6220	0.6692	0.0	0.0	0.0	0.5
0.7306	4.0	20	0.6108	0.6692	0.0	0.0	0.0	0.5
0.604	5.0	25	0.5968	0.6692	0.0	0.0	0.0	0.5
0.5031	6.0	30	0.5714	0.6692	0.0	0.0	0.0	0.5
0.5496	7.0	35	0.5302	0.6692	0.0	0.0	0.0	0.5
0.5351	8.0	40	0.4655	0.7769	0.4912	0.3256	1.0	0.6628
0.4308	9.0	45	0.3942	0.8538	0.7246	0.5814	0.9615	0.7850
0.3575	10.0	50	0.3077	0.9231	0.8780	0.8372	0.9231	0.9014
0.2808	11.0	55	0.2337	0.9308	0.8861	0.8140	0.9722	0.9012
0.2272	12.0	60	0.2053	0.9308	0.8889	0.8372	0.9474	0.9071
0.2462	13.0	65	0.2418	0.9	0.8539	0.8837	0.8261	0.8959
0.1188	14.0	70	0.2207	0.9231	0.8810	0.8605	0.9024	0.9072

Framework versions

• Transformers 4.56.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.4