bert_david

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1726
• Accuracy: 0.9801
• F1: 0.9801
• Precision: 0.9808
• Recall: 0.9801

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.1124	1.0	38	1.0985	0.3709	0.2675	0.2579	0.3709
1.0094	2.0	76	0.9154	0.8742	0.8726	0.8883	0.8742
0.4876	3.0	114	0.2264	0.9603	0.9603	0.9604	0.9603
0.169	4.0	152	0.1434	0.9669	0.9670	0.9700	0.9669
0.0587	5.0	190	0.1836	0.9603	0.9602	0.9611	0.9603
0.1567	6.0	228	0.5538	0.9073	0.9063	0.9212	0.9073
0.0874	7.0	266	0.0742	0.9868	0.9868	0.9869	0.9868
0.0276	8.0	304	0.2171	0.9735	0.9736	0.9755	0.9735
0.0958	9.0	342	0.0921	0.9801	0.9800	0.9807	0.9801
0.0356	10.0	380	0.2225	0.9669	0.9672	0.9691	0.9669
0.0008	11.0	418	0.0458	0.9868	0.9867	0.9872	0.9868
0.0281	12.0	456	0.2725	0.9603	0.9601	0.9635	0.9603
0.001	13.0	494	0.1389	0.9735	0.9734	0.9748	0.9735
0.0003	14.0	532	0.0618	0.9934	0.9934	0.9935	0.9934
0.0007	15.0	570	0.3717	0.9470	0.9468	0.9531	0.9470
0.0132	16.0	608	0.2814	0.9669	0.9670	0.9700	0.9669
0.0318	17.0	646	0.2333	0.9735	0.9734	0.9748	0.9735
0.0405	18.0	684	0.2297	0.9669	0.9668	0.9671	0.9669
0.0002	19.0	722	0.1733	0.9801	0.9801	0.9808	0.9801
0.0002	20.0	760	0.1726	0.9801	0.9801	0.9808	0.9801

Framework versions

• Transformers 4.50.0.dev0
• Pytorch 2.5.1+cu121
• Datasets 3.4.1
• Tokenizers 0.21.0