15epoch_test_march19

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

• Loss: 0.1852
• Precision: 0.9341
• Recall: 0.9304
• F1: 0.9323
• Accuracy: 0.9727

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: 8
• eval_batch_size: 8
• 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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.7937	100	0.2436	0.8360	0.8644	0.8500	0.9391
No log	1.5873	200	0.1391	0.9038	0.9044	0.9041	0.9629
No log	2.3810	300	0.1335	0.9178	0.9386	0.9281	0.9706
No log	3.1746	400	0.1400	0.9369	0.9210	0.9289	0.9706
0.246	3.9683	500	0.1348	0.9377	0.9184	0.9279	0.9705
0.246	4.7619	600	0.1324	0.9338	0.9262	0.9300	0.9717
0.246	5.5556	700	0.1422	0.9257	0.9359	0.9308	0.9718
0.246	6.3492	800	0.1484	0.9334	0.9228	0.9281	0.9713
0.246	7.1429	900	0.1558	0.9346	0.9237	0.9291	0.9714
0.0481	7.9365	1000	0.1509	0.9296	0.9392	0.9343	0.9731
0.0481	8.7302	1100	0.1540	0.9366	0.9376	0.9371	0.9743
0.0481	9.5238	1200	0.1642	0.9347	0.9332	0.9340	0.9729
0.0481	10.3175	1300	0.1665	0.9369	0.9325	0.9347	0.9734
0.0481	11.1111	1400	0.1720	0.9341	0.9334	0.9338	0.9732
0.0218	11.9048	1500	0.1673	0.9366	0.9319	0.9342	0.9733
0.0218	12.6984	1600	0.1867	0.9355	0.9302	0.9329	0.9730
0.0218	13.4921	1700	0.1872	0.9358	0.9372	0.9365	0.9736
0.0218	14.2857	1800	0.1852	0.9341	0.9304	0.9323	0.9727

Framework versions

• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.1