LILT-id-warmupv0.2

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: 0.4560
• Precision: 0.9156
• Recall: 0.9022
• F1: 0.9089
• Accuracy: 0.9499

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: 2
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	2.4390	200	0.3327	0.8649	0.8142	0.8388	0.9063
No log	4.8780	400	0.2834	0.8866	0.8606	0.8734	0.9305
0.5668	7.3171	600	0.3369	0.8825	0.8631	0.8727	0.9273
0.5668	9.7561	800	0.3568	0.8905	0.8753	0.8829	0.9321
0.1227	12.1951	1000	0.4180	0.8881	0.8729	0.8804	0.9354
0.1227	14.6341	1200	0.4567	0.8845	0.8802	0.8824	0.9321
0.1227	17.0732	1400	0.4374	0.9037	0.8949	0.8993	0.9418
0.0219	19.5122	1600	0.4580	0.9042	0.8998	0.9020	0.9435
0.0219	21.9512	1800	0.4570	0.9037	0.8949	0.8993	0.9418
0.0016	24.3902	2000	0.4560	0.9156	0.9022	0.9089	0.9499
0.0016	26.8293	2200	0.4563	0.9059	0.8949	0.9004	0.9435
0.0016	29.2683	2400	0.4642	0.9057	0.8924	0.8990	0.9418

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1